About The OABOT

180 views

I haven’t posted all that much on the OABOT. Regrettably I haven’t put the kind of time into the development of it in quite some time. Fortunately I still have a lot of prepared material on it. First off you have to consider “what makes a stock worth buying?” Such a question is what got me started on the OABOT.

Here is a link to the Spreadsheet mapping out my early concept for OABOT. Reading it you may have a better understanding at how I was able to construct the OABOT and what my thoughts and planning was going into it.

Past posts on OABOT:

OABOT demonstration

A vision for the future of OABOT

I also constructed this OABOT document to explain what it is and how it works.

Lately the way I like to use it is grab 80 names from each “risk category” then put it into finviz and scan 400 stocks and narrow the list. There are two ways to rank stocks either taking into account “what’s working” to boost stocks that are in the right group, and just by ranking by overall setup score. Usually I like maybe 10% of the setups when doing it this way which gives me a pretty good list. If I use the summary tab to find the best themes, and then categorize the exact industry in that theme and determine what phase of the risk cycle is working in that idea or the next one, I have a very concentrated list that in a couple examples I liked about 30-40% of the names I picked. This really confirmed for me that finding a group that sets up together and finding the right classification of stock within that group will really boost the accuracy of what I’m doing and definitely will be a major part of improving the tool in the future.  Unfortunately adding a multiplier combining setup score AND which groups are working ran into problems since it over rated a lot of very small industry groups with less then half a dozen stocks in them.

What Makes A Stock Worth Buying?

18 views

It is a simple question but the answer is very difficult to answer in a way that a computer can understand. Attempting to do so allowed me to learn a lot. When I started the task of trying to put Option Addict’s teaching into code almost a year ago, I wanted to explain it in a way that a computer could understand and assist me in speeding up the process. In doing so I had to put the process under a microscope and learn to think about things in a different way. The only thing all stocks should have in common is the upside should significantly outweigh the downside. However, telling a computer how to determine that isn’t likely. One commonality that I like is contracting volatility. Unfortunately the dataset I am using only has performance and volatility on set intervals such as weekly or monthly so just because price as moved up or sideways from point A to point B as volatility contracted from monthly basis to weekly basis doesn’t mean the setup is good right now. Additionally, what makes a stock worth buying near the highs is totally different than what makes a stock worth buying near the lows.

 

buy

Ultimately a good stock to buy is simply one with asymmetric risk. (a risk/reward ratio that works in your favor). We typically look for a spot where the volatility is contracted severely in a stock and a break one way or another is likely to occur soon. The resolution of that break tends to result in explosive price swings in on direction or another often enough for us to capture big winners. If it goes against us we can salvage premium or sell stock minimizing the loss while letting winners run. There of course is more taught by option addict on how to know what type of stocks to focus on but subscribers of after hours already know that. I chose these 6 stocks among others on 11/4 (see comment in OA’s post 60% in 24 hours) with a lot of help from the “OABOT” which attempts to put much of Option Addict’s teachings into code. I wanted to show these 6 because it is enough to illustrate the drastic difference in a stock’s characteristics near the highs, near the lows, and everywhere between. Each of these stocks were at least in the top 80 of their respective “categories” and were selected out of nearly 7000 different stocks total. Not every stock can be given a rating and not every stock ends up in the right classification and not every stock with the right classification and high rating turns out like you hope. However, by characterizing a few things and breaking the stocks up into groups you can at least treat stocks with certain characteristics differently, and have EACH classification scored individually. Although it is no certainty that a stock with no dramatic moves over past month or week, with contracting volatility and daily move less than 2.5 times the ATR (you want to buy something currently in a tight range relative to the last several days as well as contracting in volatility over the entire week.) That tends to be a very good starting point. Rather than filter OUT all stocks that don’t meet these characteristics points can be awarded IF a stock meets criteria A OR B and you can program the excel spreadsheet using IF (Criteria A) AND (B),OR (C) AND (D) THEN (add X points) type language. But stocks near the low need to see a sign of bottoming and be such that it is starting to curl up and then consolidate where as stocks with strong trends you wait for recent weakness and for it to consolidate without taking out prior lows. In terms of what you tell the program this is drastically different so you must code it such that IF criteria such as percentage off the highs or lows is met THEN classify the stock differently.

For example, a “trash” stock that has been chronically underperforming should ideally see some recent short term strength and be turning the corner on the short term and consolidating upwards off the lows and short term be showing signs of a new uptrend such as a stock not being far below, and ideally being above the 20 day moving average. A laggard stock’s who’s just recently been dumped on the other hand will probably be below the 20 day moving average so that criteria might not even be used. It should either still be in a strong long term uptrend and/or be seeing some sign of selling exhaustion, oversold condition and perhaps some short term consolidation along with it still being up from it’s 52 week low and possibly above the 50 day low so that it is likely to be making prior lows. The laggard was the most difficult stock to classify and rank as it represented almost all of the “leftovers” that were not close enough to highs or institutionally owned enough to be considered “quality or “momentum” but were not so illiquid and chronically underperforming enough to be labeled “trash stocks”. Ultimately I had to break it up into 3 separate categories to be able to apply different scoring metrics while still lumping all 3 of them in the laggard category.

I knew that every stock should be consolidating in some way, however in some cases consolidation could be more of a continuation pattern to the downside where as others it could be reversal pattern from the upside back down. The fact that it is consolidating on it’s own might not be useful. So each metric of consolidation must be first evaluated and scored individually and manually looked at within the context of other evaluation.

I decided to integrate fundamentals at first but in hindsight I wish I would have kept that separate and have separate classifications for fundamental scores as well so that it would be easier to filter out at will. At some point I will probably end up undoing the fundamentals. For example, for “momentum stocks” I had rewarded accelerating earnings growth substantially and as a result it is a lot more difficult to use the ranking to find good technical setups in “momentum stocks” unless they also are showing earnings growth. For “quality stocks” I decided to look at stocks that had plenty of liquidity, and insider and institutional ownership along with positive earnings growth.  The problem with that of course is there can be biotech and speculative companies with high quality charts which are still leading their respective industries without positive earnings. There are many challenges faced with classifying stocks. Do you neglect some stocks and have some good stocks that you miss or get miscategorized? Or do you risk grabbing too many stocks including those you don’t really have any interest in. Of course with additional complexity it would be possible to only set up a score relative to the sector or relative to the industry or both. I didn’t involve fundamentals for any other classification as I realized at some point I may want a separate ranking. Plus I didn’t want to have a ton of uncategorized stocks that I couldn’t rank.

 

 

Feel The Weight of a Thousand Tonnes of Gold on Your Chest!

89 views

goldsuit

gold
Gold under $1200 is at a tipping point. The weight of all those who own gold at a higher price and want out will begin to weigh on those who want in and the stock price will likely behave like gold in water… sinking. Volume profiles provide context for the collective psychology of any market. People tend to fear loss more than they appreciate or are anxious for gain. When they are under water they are looking to sell and break even or reduce the loss and they are not thinking about gains. For this reason you can anticipate speed and direction with volume profiles.
Once everyone gets in a market in a mania and there is no longer any bid to support higher prices, prices begin to decline. As they decline eventually buyer after buyer ends up under water and soon it is only a matter of time before it is a race for the exits. This by no means is a certainty, just an edge that you can gain. However, allow me to show why the odds are heavily in the gold seller’s favor and why the man in the gold suit may be like someone in a goldsuit literally underwater, unable to shed gold soon enough to reduce the weight and swim to the surface.

goldpsychology

You can see why gold under 1600 led to a sharp decline as there were fewer people likely to step in and buy and a lot of bagholders. Some of those sold to those who bought between 1200-1400 and new players entered the game. Some of those who bought above 1600 are still in the game. But now those who bought between 1200-1400 are now feeling the pain as well and those who bought into the mania top are in deep trouble. It’s likely only a matter of time before panic sets in. Failing to panic will only prolong the malaise in this market that lasts years, as after enough time, those in gold will be sick of its underperformance, but it could very well trap new players in the meantime and grind sideways for a very long time. The best thing the gold longs will have going for them is the possibility of a panic to flush out as many gold bugs as possible where new money can enter and the psychology can invert and flip in the bulls favor.

One interesting thing to note is gold is an international asset and the dollar is rising. The other thing the bulls may have going for them is that the dollar is strong. That seems to run contrary to what most gold bugs have been “pitched” but if gold can panic on a strong dollar and form a bottom on a strong dollar, it will have the majority of other currency behind it followed by the dollar. When the dollar is strong other currencies are weak and other countries may seek the dollar AND GOLD as a hedge to their declining currencies. When you price the gold in yen or euro for example, gold is not looking as bearish as the yen has also declined sharply. If gold can flush and panic can take over, volume can spike as the headline prints “gold under $1000!” and every gold bug capitulates you will have a short term constructive volume pocket above at that point and depending on the volume when gold hits around $1000, you may just begin to see the scales begin to tip in the favor of the gold buyer. However, right now it would appear the odds are in favor of the gold bears by around maybe 8 to 1 or more. And if $900 gives way, the weight will be CRUSHING to the gold bugs. Personally I think gold under $1000 is the low because that will attract the attention of a ton of new buyers and cause panic among soccer mom’s and dad’s. However, if there aren’t enough new buyers to SUBSTANTIALLY tip the scales back the other way, you could see a lot of sideways action again and an eventual decline again that is only made WORSE by all the new buyers who eventually find themselves underwater and become sellers.

Of course, buyers could still come in but if they enter they would have to come from somewhere else. The people that are supposed to stay short or stay away could cover and come back in and the buyers that are supposed to panic could double down and buy more. There’s tons of money in other markets relative to gold so liquidation of bonds or stocks to buy gold, or another market would have to grow or wealth in India would have to skyrocket as buying gold is part of their culture could save it. But it would need to happen quick and gold would need to quickly reject new lows and retake 1300 before it could start to have the odds in the bulls favor. But anything is possible.

However, being long gold is playing some theory without respect to the odds and payout. HOPE is not an investment strategy, unless you want your strong dollar and crashing gold leaving you with very little remaining CHANGE.

 

Primer on Breadth

9 views

As some of you know, I created a little research tool that with a press of a button updates the latest finviz data and runs some calculations. I can use it to do a lot of different things but it is really an unfinished project to narrow the list of stock picks but I added other features. One of them is that it updates market breadth data.
breadth tracking

Market breadth is some form of measurement of total advancing stocks vs declining stocks. There are many ways to look at breadth but in it’s simplest form it basically is a way of offsetting the market cap weighted average by telling you if participation is broad among all stocks on more of an equal weighted basis and a way to tell if market momentum favors the bulls or bears. I have set up these different measurements of breadth to focus on some of the more extreme movers in the market only. By focusing on the extreme moves I get a better litmus of what has moved substantially enough for one to conclusively say that the movement is more than just “noise” or temporary emotion. I am focusing on the conviction moves where people chased the stock one direction or another. Although the logic is simple behind breadth, there are a variety of ways to use it. There are longer term signals and shorter term signals, and there are breadth as a contrarian signal, or breadth to confirm a shift in sentiment.

% movers indicate momentum and sentiment shifts

We will start with the % movers.This focuses on stocks that have moved beyond a minimum threshold in either direction. To be fair, I created an adjustment so that if a stock has moved up 150%, the down equivalent was the amount needed to bring the stock back to it’s starting point after a 150% move upwards, or a 60% decline. Initial thrusts of breadth upwards after bearish conditions are the proverbial “green shoots” that may begin to trigger/signal a shift in sentiment. One or two of them might happen in a bear market too as the rips higher are fast and violent. So depending on how aggressive you are with trying to time every move and whether you try to anticipate the next move by buying the oversold conditions, or waiting for the shift in sentiment to take place, you may or may not want to wait for more substantial confirmation. Tracking these results on a daily basis and creating a 5 and/or 10 day moving average is one way to go about monitoring the movements for fast rips and sustainable shifts in sentiment. Typically the more bearish and greater declines that precede such a shift sets up more bullish conditions once sentiment flips to bullish and all the cash on the side and value created will trigger value buying plus growth and momentum buyers and retail trader chasing higher following conditions where everyone that was ever going to sell already had done so.

new highs/lows as contrarian signal

Tracking new 52week highs and lows (or in this case, within 1% of those marks, is a way to look at longer term accumulation vs declines and can be useful as a contrarian indicator or early-middle bull market/ middle-late bear market indicator. When there are little to no stocks at or near their lows, you may want to consider raising a bit of cash, position sizing a bit smaller, and being a bit more cautious and/or hedging. Tops are gradual, but short-intermeidate declines can be sharp and painful if they are correlated and violent. New 50 day high/low is the same principal, but can be used to confirm the longer term signal or on a shorter term basis for a more active signal. If stocks are 90% near highs vs lows but those within 1% of their 50 day high/low does not confirm, there are still perhaps some stocks near intermediate term lows offering buy the dip opportunities as opposed to a euphoric mania. If stocks are over 90% near 50 day high/low but perhaps on a 52week high/low they aren’t giving a signal, you could be near a temporary swing high and perhaps some minor caution in preparation to buy at a better price might be warranted.

Moving average breadth as trending indicator

So another form of breadth is looking at moving averages. You can use moving averages to indicate either a recent reaction and mean/reversion or as trending indicators depending on how you set them up. You can use these at whatever duration of moving averages that you want. I have just set up the standard 20 day 50 day and 200 day moving averages. I want to look at the % of stocks above each moving average (uptrend) vs the % below each moving average (downtrend) If instead you only look at those significantly above each moving average, you can look at it as the % of stocks at overbought or oversold extremes as well for mean reversion and a more contrarian oriented signal.

Then I looked at stocks with accelerating trends or an indication of a more convincing trend as it lines up on multiple time frames to be trending. To do this I looked at stocks with their 20 day moving average above their 50 day moving averages AND the price above their 20 day moving averages… vs stocks with 20 day average below their 50 day moving average and price below the 20 day moving averages. Most likely this would often produce very similar results as saying the 5 day moving average must be above the 20 day and 20 above the 50, or 5 day below the 20 day and 20 below the 50. I repeated the process with 50 day and 200 in place for the 20 day and 50 day for longer term trends.

Breadth Divergences of leadership:

One of the reasons I track TWO moves of each timeframe on the % movers is to look for leadership. One of the movers is a significant, but lesser extreme than the other. When I look at 4% movers I like to see these MORE bullish than the 1% movers and if there is to be a shift off lows, I like to see it on increased leadership/aggressive chasing that is more indicative of a paradigm shift than just your increase in buying equally due to temporary emotion on news without any clear leaders. If the 1+% movers are for example at 30% and the 4% start to creep up towards and above 50% first, this to me tells me there is a shift of sentiment and people are willing to chase a select few stocks higher, which may become the leaders of the future. A healthy market will have leadership emerge first, and that will give you an opportunity to get in before the leaders lift the rest of the stocks.

Breadth Divergences of TIME:The other kind of divergence is one of time. This is when the breadth signal on monthly data is bearish and the shorter term signal on the week and/or day shift bullish. The problem here will be how to interpret the data… Either it is a rip to sell into, or the start of a shift which will turn the weekly and monthly data positive and lead to a greater, longer term sustainable move. This signal in and of itself is not useful unless you can put it into proper context.

Breadth Intraday Shift:Tracking these results a few times a day can illustrate how things change over the course of the day, particularly when held in context. The best signal I have got since tracking this was when the Russian invasion of Ukraine took place. First everything was down as one might suspect, but the 4% movers were less bearish. This started to drag down even the 4% movers to become slightly more bearish at first as panic selling spilled over to whatever whas up and those down moderately spilled to more substantial losses. Interpreting this was initially difficult. Was it that the smart money that had a lot of capital to move accumulating on the fear and wasn’t believing in the decline but the broad sentiment and panic caused them to sell off, or did the smart money begin to shift as well, recognizing that things could get worse. But then towards the end of the day the 4% movers started ripping first, and turned very bullish, followed by the 1% movers turning moderately bullish. That signaled that all the fear based selling was over and the market began to recognize it for what it was… a buying opportunity on fears of a world war that may not materialize. This continued into the next day along with the news that came that Russia was withdrawing their invasion and for the time being the signal got you near a nice swing market low and if you followed along with OA’s risk cycle, it was a fortunate situation as you knew what to buy and you had a signal to get in the market ahead of the big part of the move.

 

Adding Volume Filters: One thing you may want to do is only look for stocks that are up or down on the day on volume that is significantly greater than usual 1.5x, 2x,3x. This signals more active participation than usual and will better measure of participation as opposed to thin volume movements that may not be as telling of an aggressive change in sentiment as when all the stocks advancing are doing so on increased volume. The problem with this is you can only really use it for the daily moves as far as I am aware of as it is more difficult to track volume on a weekly, monthly and quarterly basis.

 

Mark This Day On Your Calander! Overall Breadth Oversold

47 views

Since the “breadth” indicators looks at the percentage bullish vs % bearish on multiple time frames and uses multiple ways to look at breadth it is VERY rare that you see the overall average ratings outside of the normal range of say 40% to 60%.
breadth oversold
The highest I’ve seen it to compare the opposite was 74.6% on July 3rd. That created a meaningful upper range of the S&P and practically top picked the Russel just in time one last day before the major selloff, from which we have sold off over 10% as of today. The opposite reading would effectively be 25.4% on the bearish side. Instead we currently have a rating under 20% for the first time since I’ve been watching this in April and (semi) actively tracking since late June.
Many students of breadth will tell you to wait for a “breadth thrust” or dramatic and significant flip FROM oversold or overbought levels as you then have the change in sentiment which triggers shorts to chase from oversold and sellers to pile on from overbought. You also can potentially look for leadership to emerge which can be evident from the larger of the two numbers on any time frame acting more bullish than the lesser of the two numbers. However, you can also look at the opposite process of trying to dollar cost average or scale in/out as well. You might use it as a signal to transfer money to more aggressively buy the rare historic event. Of course, it is worth mentioning that bull markets tend to remain overbought for quite some time and bear markets can remain oversold for some time. Nevertheless, this type of substantial selling could represent ultimate discouragement lows…. Ultimately the trick is putting the breadth into context with sentiment and relevant context. That is difficult to do from a stale bull which has yet to receive public participation, but I am going to bet that this is a significant shakeout that has far reaching global implications like 1998 but one that is still in the context of credit expansion and a bullish business cycle with credit still remaining very loose. You could see huge ramifications from sanctions on russia, ebola, ISIS, global tensions and increased fear but you cannot reverse and manipulate the primary trend which I believe is still higher as you do not have confirmation in the S&P, Nasdaq and dow. The russel is concerning but markets don’t act in isolation forever. So while some say the russel is a (complex) head and shoulders breakdown (or double top), I say it is a head and shoulders FAKEOUT until proven otherwise.
Having the discipline to rotate capital into risk here is certainly not easy however… particularly leveraged option buying strategies which tend to capitalize off of low volatility when the volatility is high. That makes this a bit more challenging, but there is still a good roadmap of which stocks to focus on in After Hours with Option Addict and I believe the opportunity is also good for buying TNA and XIV

 

The moves are adjusted for the amount that would erase a move. For example 100% movers corresponds to 100% up movers vs 50% down movers since a 50% down move following a 100% up move would bring you back to even. A 150% move up corresponds to 60% decline. 10% up = 9.0909% down. etc

Scanning For Themes

2 views

I decided to set the OAbot up so I could get a quick glance and look for themes. It was initially with the intention that I would completely automate the process, but in it’s current format the OAbot works better as a research tool than an automatic setup generator. One of the rankings looks at the average “setup score” of each stock within an industry and comes up with an average.The setup score typically (depending on the classification of each stock) will look for strong uptrends, recent consolidation and volatility compression using monthly volatility, weekly volatility, and daily move as a % of ATR).

Although there may be other valuable metrics such as relative volume of the industry and breadth, ideally the only necessary component to identifying anticipatory entries is all stocks within a group saying the same thing. I can repeat this scan across sectors or classification types, but there is an inherent bias towards stocks that are near the highs as any laggard that just exploded to new highs will then be looked at as a momentum or quality stock. As such aside from trying to subtract the recent day move from the high to see what the stock was classified as before the move, and/or having some sort of metric to track over time at what % of each group has broken out and is no longer classified as a laggard or trash near lows, there is not much that can be done… I don’t have the time to put a lot of work into the OABOT right now as I once did.

So for now we can scan for themes quickly. The following is all industries with the # of stocks in the industry over 20, an individual setup score average over 90, and sorted by avg setup score in the industry.

Major Integrated Oil & Gas 106.474303
Oil & Gas Pipelines 105.618511
Trucking 101.401134
Residential Construction 99.1633571
Rental & Leasing Services 98.1455953
Oil & Gas Refining & Marketing 97.1936567
Semiconductor – Broad Line 96.6590651
Gas Utilities 96.1367812
Textile – Apparel Clothing 96.0117017
REIT – Residential 94.6209923
Oil & Gas Equipment & Services 93.8425753
Specialty Chemicals 93.7406842
Auto Parts 93.4825295
Property & Casualty Insurance 93.4359768
Credit Services 93.4128977
Gold 93.3070126
Chemicals – Major Diversified 93.1052147
REIT – Diversified 92.6859943
Telecom Services – Domestic 91.2492766
Drug Manufacturers – Other 91.1494354
REIT – Retail 90.8556666
Electric Utilities 90.6886449
Oil & Gas Drilling & Exploration 90.1819049
Independent Oil & Gas 90.1435737

 

The individual setup score breaks down the stock differently depending upon the classification. Stocks near their lows are evaluated differently than those near their highs. Stocks flagged as “short squeeze candidates” are evaluated partially by their float and % of float short in addition to weightings from each category of classifications. Stocks that are liquid with good fundamentals and growth prospects are looked at differently. Stocks with accelerating momentum and growth are looked at differently. There are 3 different types of “laggards” and each has a different way to evaluate the score. The score is very dynamic in that if certain things are true it is evaluated a certain way. If either of a number of things are true, it may be given a bonus to the score. If a combination of things exceed a certain value it may contribute to the score. Should I find the time, I will put a lot more thought into the exact metrics, weightings, and components that go into the score by tracking which setups look better after making tweaks over a longer period of time until I have more ideal rankings. Once I am able to further fine tune everything, and possibly even track price across time and automate the tracking, THEN I feel I may be able to construct a tool that is more automated, particularly if I implement many of the things that are computed, but not factored into the end ranking just yet. Rather than use the OAbot individual stock scores at this time, I think it is quicker just to go to finviz and scan through each of these industries until you have enough setups or identify a theme or two that you are satisfied with.

I’ve already manually from top down analysis concluded energy was a good setup a couple weeks ago… so all the various oil&gas plays coming up is additional confirmation.

I like how trucking stocks while a very diverse group (some stocks near highs, others near lows), still shows a lot of consolidation and bullish looking setups. Look at HTLD and UACL as an example.  Very different stocks right now, but both look like they are working sideways to set up a bullish move. Even among the worst stocks of the group as determined by their % off of 50 day highs are names like SWFT that are at least consolidating above a recent balance area after the sharp drop and rejecting new lows below $20 so far. Not at all a bullish chart after making the equal low longer term, but you could easily see a move to 22 before declines and recent action is at least decent considering the technical damage done to it. There are still a few stocks of the group that look like while contracting in volatility, they still have more sideways work to do for a couple weeks. I think the industry may need a bit more time for some of the leading stocks to consolidate and some of the others to work sideways or breakout and retest as the others set up, but overall it looks like an industry where you could identify a select few setups, then move onto the next, and possibly at some point you might see more of them beginning to correlate… Either way it may be a good theme to watch.

I am surprised to see residential construction score so well, I never would have even looked right now. It has been a dog of an industry but a handful of stocks are making bull flags and consolidating after a bounce from lower prices. I don’t love the industry, but every now and then you’ll see an industry flagged you wouldn’t have thought to even look at and get some ideas. It’s nice to have a preset sort of program that doesn’t have bias aside from the one you programmed in that can then be checked critically with a human eye as you have the machine that doesn’t get tired or miss anything, and then the human eye to critically assess and the intuition developed from experience (that cannot easily be programmed) to filter down the process. Since currently I am using more of an intuitive feel based identification of theme and selection of stocks, it is not a bad idea to combine objective filters such as selecting from those in sectors with seasonality that suggests we are near a low or have plenty of upside ahead.

This process takes much longer when I am analyzing it and then converting analysis to words and typing them out and it may not be exciting to read, so I won’t go through all of them. I may in the future just look to create only a volatility compression score only as being able to limit the search to groups with plenty of volatility contraction may be a little easier, particularly if I combine with a decent overall setup score in the way that it will filter down the groups to those over 85 or 90 rating first and then will sort by volatility compression scores.

 

the end for now.

 

Breadth Marches On

1 views

One of the interesting I observed about breadth recently is that off of oversold levels the bounce was so strong that it was difficult to distinguish leaders from anything else. “buy everything” was the motto just as on a longer fractal it was off the 2009 lows. Breadth still flipped aggressively and there were still a large percentage of stocks up, as well as a large percentage of big movers up, but it was not a market that you could say was pulled up by leaders, but instead one in which everything rallied, and a small percentage of stocks rallied a lot. That could be interpreted as shorts merely getting squeezed out and an over enthusiastic buy after a sharp sell off before another leg down, or as a sell off that had gotten so far oversold that the opportunity was so good that the buyers didn’t have time to be picky and do thorough research, they wanted in on anything and everything they could. The key to evaluating the breadth sometimes is patience when you are unable to have an edge in putting the breadth into proper context. Coming off oversold is tough.
A breadth flip from bearish breadth to bullish breadth is generally really good off of oversold levels, but the difficult part is assessing if it will be a market of stocks or a stock market and if the move will represent a V reversal or if those lows will be retested? Is it going to run away from here or is it going to retest resistance and remain in a range? Sometimes clarity is not provided until later on.

Fortunately, the last few sessions have seen leadership as markets have gotten more stretched from the lows. We do see more modest breadth overall, but this is clearly not “just noise” as some of the 1% movers may be and not a reactionary rally, but one driven by big movers clearly outnumbering the small moves.

For example today just looking at one day moves, 60% of all stocks moving were up today. Of those that moved 1% or more, 61.2% were up. Of all 4% or more moves, 64% were up. While the extreme moves are more rare, the more extreme the filter got, the more bullish the breadth is as a % of movers of a certain threshold. In a strongly bullish environment, the breadth being even more bullish as the moves become more extreme is typically a good sign. That may signal a number of things such as conviction to the upside, positive earnings and revisions and fundamental factors that drive stocks up, generally positive news as news driven moves may be of greater magnitude, and an increased amount of capital in the system allowing them to chase. All of these are typically bullish except when at overbought extremes in bubble territory and indicative of sentiment being too extreme when 4% movers make up a larger and larger percentage of up moves.

One sign for extreme sentiment might be breadth on a longer term time frame showing extreme up moves composing a higher percentage than strong up moves, and a large overall number. We have not seen that lately, as sentiment is not extreme at all on the longer term. We also have not seen much leadership on the extreme moves at all on longer term basis, which means this is still a swing trading environment, as opposed to the market entering run away trend mode where buy and hold for a run away move is king (I imagine the 80s and 90s showed plenty of leading growth stocks running gap and go or breakout and chase). This action in my view can be confirmed by the Russel still remaining in a consolidation range.

Monday August 11th (possibly the prior Friday) and for sure Wednesday, August 13th saw huge reversals from bearish breadth that continued to stack up over the next couple weeks. Although breadth on the daily move is not above the 80% range as it was a week ago, the fact that there still appears to be greater convictions on the big movers to the upside shows that leaders can continue to pull the market higher. The % of 1% movers moving at least 4% is around 15% so letting your winners run still appears to be a good strategy as well which supports the swing trader’s and stock picker’s cause.

Today here are some signs the bulls are winning aside from up moves outnumbering down moves and the market being up:
% of up stocks moving 4%+
5.80%

% of down stocks moving >-4%
4.92%
% of 1%+ stocks moving 4%+
14.86%

% of -1% stocks moving >-4%
13.23%
% of up stocks moving 1%+
39.01%
% of down stocks moving >-1%
37.18%

 

breadth tracking

Trading Systems Part 2: Accounting for Multiple Outcomes

36 views

In a basic “pot odds” calculation as described in the last post trading systems: pot odds and trading, if you can let your winners run so that you sell your option for 4 times the cost, netting 3 times your risk or 300%, you are getting 3:1 and thus only have to be right more than 1/4. (lose 3, win on 3 the 4th time and break even). However, options are more complicated than that. While you may be right or wrong on whether the stock goes to a price such that you earn 300%, the stock could reach the target earlier, in which case you may make more than 300%. Also, the timing could be late but direction still right, or the stock could gap up or run past the target and exceed your return. As a result, you might have a number of outcomes ranging between the extreme 1000% or so outlier trades to 0% depending on your ability to let the winner run and your tolerance for volatility. A simple pot odds calculation becomes tricky for determining whether or not the system is profitable, and how much. For this you can use “expected value”, but that really doesn’t grasp the expectancy of the system per trade over time. Nevertheless, since 1% risk represents a small amount of capital lost through volatility, it can still be looked at from the concept of how many 1% position trades are taken per year, and you can approximate that the growthrate that occurs in your portfolio is close to 1% of the expected value of the system as a decent starting point. With such a position size, you can take more trades on at once and as a result the gains can compound faster. There is still some correlation (a LOT of correlation during a significant correlated selloff), but by diversifying trades over multiple time frames and spreading out your purcahses and using the occasional hedge, you can mitigate the cost of volatility at very little cost to your overall results.

So let’s get into an example. Say you average out all your trades over 100%, You average all those over 30%, all those between -30 and 30% and all those from -30 to -90% and then all trades that lose more than 90%. That gives you 5 averages that can define your system in the total average of “large wins” “moderate wins” “break even” “salvaged premium” and “full loss”. The frequency of each event determines a number of trades out of all trades. Simply divide the number of trades by all trades for each of the 5 and you have an estimated probability of the return. Average the return of all trades in each category to get the average return.

From here I actually like to model my results as opposed to using expected value, but let’s just give an example:

Last year’s 2013 results before I booked a couple of large wins in TWTR and TSLA were

17% of trades averaged 294%, 20% of trades averaged 53%, 15.8% of trades averaged -1% 23.2% of trades averaged -63% and 24% of trades I rounded down to the full 100% loss.

To do an “expected value” calculation, you simply multiply the probability of each event by the return and then add them together. So (.17*2.94)+(.2*.53)+(.158*-.01)+(.232*-.632)+(.24*-1)=.217596=~21.76% gain per option trade. By letting the winners run in 2014 from Jan until Jun before a few large wins in AMZN and PCLN and others, this yield was just above 40% but it was also substantially helped by a few large outliers well over 500% (GMCR,AAPL,etc) . So how does that translate to portfolio growth? There is an easy calculation that can be done if the position size is small.

Let’s put it at 1% risk or 1% of .2176 expectation=.002176% per trade. In other words, we multiply at a factor of 1+.002176 per trade. (1.002176^X)-1=annualized rate of return where X is the number of 1% trades you can place with this system in a year.  Make 300 trades and you are looking at about a 1.002176^300=~1.919-1=~.92=92% annualized rate of return.

The smaller the position size, the more consistent the system, the lower the correlation between trades, and the greater the edge, the more accurate the estimation is and typically the more consistent the results are with the expectation. However, keep in mind the relationship of return and risk as you look at how it becomes less accurate with increased position size. The assumption that increasing the risk will proportionally increase the volatility and the reward assumes a linear relationship which is not true. Instead the relationship looks as follows:

Kelly-Criterion

Nevertheless, we can use a monte carlo simulation to pull a random number between 0 and 1. Since there is a 17% chance of the upper result if it pulls a 0.17 or less, this corresponds to the 294% return result occuring, where as a 20% chance of a 53% return corresponds to a .17 to .37 being pulled and so on. That can allow us to randomly pull the results of a thousand trades, and then through a monte carlo simulation tool we can simulate thousands of trials of 1000 trades and model any particular result or formula as a consequence of each result to get the intended information we need and reflect the results we want. Just by using the average return and ignoring the substantial probability of effective ruin

At 1% with ZERO fees, we get an average return of 91.35% compared to our expected 91.9%

With 2% with zero fees we get an average return of 265.7% compared to an expected ((2*0.002176)+1)^300=268%. Being able to place the same volume of trades as the risk gets too much larger becomes increasingly difficult and increases the correlation risk that is not modeled here but will negatively effect returns. As we increase the position size the difference between the average simulated return and the calculated return grow as the simulation is less and less close to the calculated returns due to losses from volatility. The other thing is, as bet size increased it becomes increasingly difficult to place the same number of trades per year.

What this growing difference does not show is the increasing skew of results. In other words, if 1000 traders were to trade this system over 300 trades at a fixed position size, a smaller and smaller percentage of traders would have a greater and greater result as position size increased. The magnitude of the outliers would increase which would skew the AVERAGE expected results. The opposite side of it would be an increasing probability of poor results as well. Effectively, the more you increase the position size the more you turn it into a lotto ticket, until you eventually take your skill edge out of it and your results actually begin to decline and eventually even become negative if you get too carried away. Even if you are adjusting the percentage risked, you should never go over the full kelly, and since you don’t have an unlimited amount of time to recover the drawdown, and you can only place a few hundred trades per year with overlapping correlation, and there is greater uncertainty… I would absolutely drastically reduce the bet from the kelly. Calculating the “kelly” is a process that has many applications and is another story for another time.

Trading Systems Part 1: Pot Odds and Trading

56 views

I wanted to go over some basics of building a profitable trading system by looking at “pot odds” a term applied in the game of poker and “expected value” and how it is interchangeable to trading. I am usually not a fan of looking at expected value because it doesn’t properly frame in the potential for massive account volatility, but it is still useful in understanding the basic requirements for a profitable system and can assist people new to trading or who have not yet learned to be profitable why they might have some “leaks in their game”. “pot odds” looks at the amount you can win, vs the amount you have at risk to determine the necessary probability of your decision in order to be profitable. From it you can also determine “expected value (EV) per hand (or per trade) ASSUMING the same amount risked every time.

In other words, say your opponent moved all in for the size of the pot on the flop. You get paid 2:1 to call since you must match your opponents bet, but get twice that (the pot AND the opponent’s stack) if you are right. You can then lose twice, win on the third and win everything you lost back. Therefore, you need only just better than a 1 out of 3 chance of winning for the call to be “profitable” in the long run.

If you have your entire bankroll at risk rather than a small percentage you are overextended and it is not a long term profitable decision because it will inhibit your ability to earn in the future as you will not be able to recover from a loss. “bankroll management” as it is called in poker and “position sizing” as it is called in stock trading is an important seperate topic that combined with “pot odds” greatly influences your long term profitability.

What novice poker players in cash games and traders often ignore is the fact that they will lose due to volatility. Lose 10% of capital, you need 11% just to get back to even, so 1% is lost due to volatility (it is actually less since multiple gains will also compound at a less substantial rate). Lose 50% of your capital and you need 100% return to get back to even.

kellycriterionbetting more than 2 times the “full kelly” actually turns a profitable strategy unprofitable over ANY time frame, and erases any skill edge you have. While your winners may also compound, your losses create disproportionately large drawdowns that require a greater skill or edge to overcome to get back to even, and due to simple chance those drawdowns are certain to occur over a long enough time horizon with a correlated, unhedged system. Your edge is not as profitable as the expected value calculation in reality, at least not without introducing a “risk of ruin”. But I digress, let’s keep things simple… Pot odds. To keep things simple and ignore “uncertainty”, I want to use a situation in trading which the numbers tend to be a bit more concrete as opposed to uncertainty, so I want to give the example of “risk arbitrage” through buying into mergers and acquisitions.

According to this blog entry, 97 deals were completed, only 2 deals failed. However, many others remain unresolved and sometimes it may take longer than anticipated to close which doesn’t eat into your total profits for the trade, but does eat into your annualized profits since it takes longer for you to be able to reuse that capital. Since we are concerned in this article about profitability and pot odds, time will not be factored in. Even though almost 98% of deals closed, we will leave a bit more margin of error since some deals can linger on for years and then not close which could skew the numbers a bit. To be safe, we will say over 90% of deals go through and we can say that with pretty substantial confidence. At 90% chance of being right to win X and 10% chance to be wrong and lose 100% we must solve for X such that -.9x=-1*.10 or.9x=-.1 or -.1/-.9=1/9=.111111=11.1111%.

If you would lose 100% and win 11.111111% when you are right, the strategy of buying into these deals post announcement would be break even. So if you put your entire capital at risk, you would have the “pot odds” to buy any time there is a 11.11% payout or better. But when deals fail, they don’t go to zero, they drop down to around their pre-buyout price on average. This can change from deal to deal and some deals may just be in a constant downtrend.

If you had a downside of only 10% and an upside of only 5%, how would it compare with a trade with a downside of 100% and upside of 20%? You can confirm the expected value by downside of a loss (expressed as negative percentage) times probability of a loss plus upside of gain plus probability of a gain.

System 1 10% downside, 5% upside:

(-.1*.1)+(.05*.9)=.035=3.5% per completed deal.

(-.1*.1)+(.2*.9)=.08 or 8% per completed deal.

This shows why I don’t like expected value in comparing two systems, you risk insolvency by putting 100% of your capital into the trade with the “higher expected value” where as you only put 10% of your account at risk with the “lower expected value trade” If you were to risk 10% of capital on the 2nd strategy, you would only grow your portfolio by 0.80% per deal. As such the deal providing the better risk reward assuming equal probabilities of success/failure is almost always going to provide the better return on risk. You must evaluate system on an equal portfolio risk basis to truely determine which is better. The kelly criterion is a good metric for comparing one system to another on an equal risk basis. However, the kelly criterion should not be used for position sizing as it is probably 5 times more aggressive than it should be (or more) due to the false assumptions the formula makes about having an “infinite time frame”, a “certain, predefined known edge” and “complete emotional tolerance for all volatility that does not effect your edge” and that multiple bets at a lower percentage with a low correlation provides a better return on risk overall.

Additionally, if you could complete 12 trades a year in the 3.5% expected value system and the  8% per deal system took over a year, it would be much more profitable, so a raw calculation of “expected value per trade” must first be “normalized” (in this case normalized means something very different than “normalizing” volatility” as instead it means it should be set up to reflect the “normalized rate of return at a given level of risk” to reflect overall growth on portfolio given equal risk via position sizing). Then it must be also “annualized” so that they are equalized on a particular time frame after fees… However, due to uncertainty, you also must consider that the large edge that compounds fewer times is less vulnerable to small changes that may negatively impact the ROI more than you thought.

I don’t want to get any more complicated than I already have in this article, but for now I will tell you to practice by looking at your “risk” and your “reward” and measuring out your probabilities required to break even. For example if you have a target price that nets 3 times your risk, you need to hit your target 1/4 times in order to break even.  With options if you try to hit around “break even” on hitting your targets or even lower, the trades that expire in the money and have to be sold for a slight gain, break even and only a slight loss as opposed to the entire risk of the option PLUS the occasional trade that runs beyond the target will create profitability in the system.  I may get into more specific and more realistic trading systems in a future next post.

About The OABOT

180 views

I haven’t posted all that much on the OABOT. Regrettably I haven’t put the kind of time into the development of it in quite some time. Fortunately I still have a lot of prepared material on it. First off you have to consider “what makes a stock worth buying?” Such a question is what got me started on the OABOT.

Here is a link to the Spreadsheet mapping out my early concept for OABOT. Reading it you may have a better understanding at how I was able to construct the OABOT and what my thoughts and planning was going into it.

Past posts on OABOT:

OABOT demonstration

A vision for the future of OABOT

I also constructed this OABOT document to explain what it is and how it works.

Lately the way I like to use it is grab 80 names from each “risk category” then put it into finviz and scan 400 stocks and narrow the list. There are two ways to rank stocks either taking into account “what’s working” to boost stocks that are in the right group, and just by ranking by overall setup score. Usually I like maybe 10% of the setups when doing it this way which gives me a pretty good list. If I use the summary tab to find the best themes, and then categorize the exact industry in that theme and determine what phase of the risk cycle is working in that idea or the next one, I have a very concentrated list that in a couple examples I liked about 30-40% of the names I picked. This really confirmed for me that finding a group that sets up together and finding the right classification of stock within that group will really boost the accuracy of what I’m doing and definitely will be a major part of improving the tool in the future.  Unfortunately adding a multiplier combining setup score AND which groups are working ran into problems since it over rated a lot of very small industry groups with less then half a dozen stocks in them.

What Makes A Stock Worth Buying?

18 views

It is a simple question but the answer is very difficult to answer in a way that a computer can understand. Attempting to do so allowed me to learn a lot. When I started the task of trying to put Option Addict’s teaching into code almost a year ago, I wanted to explain it in a way that a computer could understand and assist me in speeding up the process. In doing so I had to put the process under a microscope and learn to think about things in a different way. The only thing all stocks should have in common is the upside should significantly outweigh the downside. However, telling a computer how to determine that isn’t likely. One commonality that I like is contracting volatility. Unfortunately the dataset I am using only has performance and volatility on set intervals such as weekly or monthly so just because price as moved up or sideways from point A to point B as volatility contracted from monthly basis to weekly basis doesn’t mean the setup is good right now. Additionally, what makes a stock worth buying near the highs is totally different than what makes a stock worth buying near the lows.

 

buy

Ultimately a good stock to buy is simply one with asymmetric risk. (a risk/reward ratio that works in your favor). We typically look for a spot where the volatility is contracted severely in a stock and a break one way or another is likely to occur soon. The resolution of that break tends to result in explosive price swings in on direction or another often enough for us to capture big winners. If it goes against us we can salvage premium or sell stock minimizing the loss while letting winners run. There of course is more taught by option addict on how to know what type of stocks to focus on but subscribers of after hours already know that. I chose these 6 stocks among others on 11/4 (see comment in OA’s post 60% in 24 hours) with a lot of help from the “OABOT” which attempts to put much of Option Addict’s teachings into code. I wanted to show these 6 because it is enough to illustrate the drastic difference in a stock’s characteristics near the highs, near the lows, and everywhere between. Each of these stocks were at least in the top 80 of their respective “categories” and were selected out of nearly 7000 different stocks total. Not every stock can be given a rating and not every stock ends up in the right classification and not every stock with the right classification and high rating turns out like you hope. However, by characterizing a few things and breaking the stocks up into groups you can at least treat stocks with certain characteristics differently, and have EACH classification scored individually. Although it is no certainty that a stock with no dramatic moves over past month or week, with contracting volatility and daily move less than 2.5 times the ATR (you want to buy something currently in a tight range relative to the last several days as well as contracting in volatility over the entire week.) That tends to be a very good starting point. Rather than filter OUT all stocks that don’t meet these characteristics points can be awarded IF a stock meets criteria A OR B and you can program the excel spreadsheet using IF (Criteria A) AND (B),OR (C) AND (D) THEN (add X points) type language. But stocks near the low need to see a sign of bottoming and be such that it is starting to curl up and then consolidate where as stocks with strong trends you wait for recent weakness and for it to consolidate without taking out prior lows. In terms of what you tell the program this is drastically different so you must code it such that IF criteria such as percentage off the highs or lows is met THEN classify the stock differently.

For example, a “trash” stock that has been chronically underperforming should ideally see some recent short term strength and be turning the corner on the short term and consolidating upwards off the lows and short term be showing signs of a new uptrend such as a stock not being far below, and ideally being above the 20 day moving average. A laggard stock’s who’s just recently been dumped on the other hand will probably be below the 20 day moving average so that criteria might not even be used. It should either still be in a strong long term uptrend and/or be seeing some sign of selling exhaustion, oversold condition and perhaps some short term consolidation along with it still being up from it’s 52 week low and possibly above the 50 day low so that it is likely to be making prior lows. The laggard was the most difficult stock to classify and rank as it represented almost all of the “leftovers” that were not close enough to highs or institutionally owned enough to be considered “quality or “momentum” but were not so illiquid and chronically underperforming enough to be labeled “trash stocks”. Ultimately I had to break it up into 3 separate categories to be able to apply different scoring metrics while still lumping all 3 of them in the laggard category.

I knew that every stock should be consolidating in some way, however in some cases consolidation could be more of a continuation pattern to the downside where as others it could be reversal pattern from the upside back down. The fact that it is consolidating on it’s own might not be useful. So each metric of consolidation must be first evaluated and scored individually and manually looked at within the context of other evaluation.

I decided to integrate fundamentals at first but in hindsight I wish I would have kept that separate and have separate classifications for fundamental scores as well so that it would be easier to filter out at will. At some point I will probably end up undoing the fundamentals. For example, for “momentum stocks” I had rewarded accelerating earnings growth substantially and as a result it is a lot more difficult to use the ranking to find good technical setups in “momentum stocks” unless they also are showing earnings growth. For “quality stocks” I decided to look at stocks that had plenty of liquidity, and insider and institutional ownership along with positive earnings growth.  The problem with that of course is there can be biotech and speculative companies with high quality charts which are still leading their respective industries without positive earnings. There are many challenges faced with classifying stocks. Do you neglect some stocks and have some good stocks that you miss or get miscategorized? Or do you risk grabbing too many stocks including those you don’t really have any interest in. Of course with additional complexity it would be possible to only set up a score relative to the sector or relative to the industry or both. I didn’t involve fundamentals for any other classification as I realized at some point I may want a separate ranking. Plus I didn’t want to have a ton of uncategorized stocks that I couldn’t rank.

 

 

Feel The Weight of a Thousand Tonnes of Gold on Your Chest!

89 views

goldsuit

gold
Gold under $1200 is at a tipping point. The weight of all those who own gold at a higher price and want out will begin to weigh on those who want in and the stock price will likely behave like gold in water… sinking. Volume profiles provide context for the collective psychology of any market. People tend to fear loss more than they appreciate or are anxious for gain. When they are under water they are looking to sell and break even or reduce the loss and they are not thinking about gains. For this reason you can anticipate speed and direction with volume profiles.
Once everyone gets in a market in a mania and there is no longer any bid to support higher prices, prices begin to decline. As they decline eventually buyer after buyer ends up under water and soon it is only a matter of time before it is a race for the exits. This by no means is a certainty, just an edge that you can gain. However, allow me to show why the odds are heavily in the gold seller’s favor and why the man in the gold suit may be like someone in a goldsuit literally underwater, unable to shed gold soon enough to reduce the weight and swim to the surface.

goldpsychology

You can see why gold under 1600 led to a sharp decline as there were fewer people likely to step in and buy and a lot of bagholders. Some of those sold to those who bought between 1200-1400 and new players entered the game. Some of those who bought above 1600 are still in the game. But now those who bought between 1200-1400 are now feeling the pain as well and those who bought into the mania top are in deep trouble. It’s likely only a matter of time before panic sets in. Failing to panic will only prolong the malaise in this market that lasts years, as after enough time, those in gold will be sick of its underperformance, but it could very well trap new players in the meantime and grind sideways for a very long time. The best thing the gold longs will have going for them is the possibility of a panic to flush out as many gold bugs as possible where new money can enter and the psychology can invert and flip in the bulls favor.

One interesting thing to note is gold is an international asset and the dollar is rising. The other thing the bulls may have going for them is that the dollar is strong. That seems to run contrary to what most gold bugs have been “pitched” but if gold can panic on a strong dollar and form a bottom on a strong dollar, it will have the majority of other currency behind it followed by the dollar. When the dollar is strong other currencies are weak and other countries may seek the dollar AND GOLD as a hedge to their declining currencies. When you price the gold in yen or euro for example, gold is not looking as bearish as the yen has also declined sharply. If gold can flush and panic can take over, volume can spike as the headline prints “gold under $1000!” and every gold bug capitulates you will have a short term constructive volume pocket above at that point and depending on the volume when gold hits around $1000, you may just begin to see the scales begin to tip in the favor of the gold buyer. However, right now it would appear the odds are in favor of the gold bears by around maybe 8 to 1 or more. And if $900 gives way, the weight will be CRUSHING to the gold bugs. Personally I think gold under $1000 is the low because that will attract the attention of a ton of new buyers and cause panic among soccer mom’s and dad’s. However, if there aren’t enough new buyers to SUBSTANTIALLY tip the scales back the other way, you could see a lot of sideways action again and an eventual decline again that is only made WORSE by all the new buyers who eventually find themselves underwater and become sellers.

Of course, buyers could still come in but if they enter they would have to come from somewhere else. The people that are supposed to stay short or stay away could cover and come back in and the buyers that are supposed to panic could double down and buy more. There’s tons of money in other markets relative to gold so liquidation of bonds or stocks to buy gold, or another market would have to grow or wealth in India would have to skyrocket as buying gold is part of their culture could save it. But it would need to happen quick and gold would need to quickly reject new lows and retake 1300 before it could start to have the odds in the bulls favor. But anything is possible.

However, being long gold is playing some theory without respect to the odds and payout. HOPE is not an investment strategy, unless you want your strong dollar and crashing gold leaving you with very little remaining CHANGE.

 

Primer on Breadth

9 views

As some of you know, I created a little research tool that with a press of a button updates the latest finviz data and runs some calculations. I can use it to do a lot of different things but it is really an unfinished project to narrow the list of stock picks but I added other features. One of them is that it updates market breadth data.
breadth tracking

Market breadth is some form of measurement of total advancing stocks vs declining stocks. There are many ways to look at breadth but in it’s simplest form it basically is a way of offsetting the market cap weighted average by telling you if participation is broad among all stocks on more of an equal weighted basis and a way to tell if market momentum favors the bulls or bears. I have set up these different measurements of breadth to focus on some of the more extreme movers in the market only. By focusing on the extreme moves I get a better litmus of what has moved substantially enough for one to conclusively say that the movement is more than just “noise” or temporary emotion. I am focusing on the conviction moves where people chased the stock one direction or another. Although the logic is simple behind breadth, there are a variety of ways to use it. There are longer term signals and shorter term signals, and there are breadth as a contrarian signal, or breadth to confirm a shift in sentiment.

% movers indicate momentum and sentiment shifts

We will start with the % movers.This focuses on stocks that have moved beyond a minimum threshold in either direction. To be fair, I created an adjustment so that if a stock has moved up 150%, the down equivalent was the amount needed to bring the stock back to it’s starting point after a 150% move upwards, or a 60% decline. Initial thrusts of breadth upwards after bearish conditions are the proverbial “green shoots” that may begin to trigger/signal a shift in sentiment. One or two of them might happen in a bear market too as the rips higher are fast and violent. So depending on how aggressive you are with trying to time every move and whether you try to anticipate the next move by buying the oversold conditions, or waiting for the shift in sentiment to take place, you may or may not want to wait for more substantial confirmation. Tracking these results on a daily basis and creating a 5 and/or 10 day moving average is one way to go about monitoring the movements for fast rips and sustainable shifts in sentiment. Typically the more bearish and greater declines that precede such a shift sets up more bullish conditions once sentiment flips to bullish and all the cash on the side and value created will trigger value buying plus growth and momentum buyers and retail trader chasing higher following conditions where everyone that was ever going to sell already had done so.

new highs/lows as contrarian signal

Tracking new 52week highs and lows (or in this case, within 1% of those marks, is a way to look at longer term accumulation vs declines and can be useful as a contrarian indicator or early-middle bull market/ middle-late bear market indicator. When there are little to no stocks at or near their lows, you may want to consider raising a bit of cash, position sizing a bit smaller, and being a bit more cautious and/or hedging. Tops are gradual, but short-intermeidate declines can be sharp and painful if they are correlated and violent. New 50 day high/low is the same principal, but can be used to confirm the longer term signal or on a shorter term basis for a more active signal. If stocks are 90% near highs vs lows but those within 1% of their 50 day high/low does not confirm, there are still perhaps some stocks near intermediate term lows offering buy the dip opportunities as opposed to a euphoric mania. If stocks are over 90% near 50 day high/low but perhaps on a 52week high/low they aren’t giving a signal, you could be near a temporary swing high and perhaps some minor caution in preparation to buy at a better price might be warranted.

Moving average breadth as trending indicator

So another form of breadth is looking at moving averages. You can use moving averages to indicate either a recent reaction and mean/reversion or as trending indicators depending on how you set them up. You can use these at whatever duration of moving averages that you want. I have just set up the standard 20 day 50 day and 200 day moving averages. I want to look at the % of stocks above each moving average (uptrend) vs the % below each moving average (downtrend) If instead you only look at those significantly above each moving average, you can look at it as the % of stocks at overbought or oversold extremes as well for mean reversion and a more contrarian oriented signal.

Then I looked at stocks with accelerating trends or an indication of a more convincing trend as it lines up on multiple time frames to be trending. To do this I looked at stocks with their 20 day moving average above their 50 day moving averages AND the price above their 20 day moving averages… vs stocks with 20 day average below their 50 day moving average and price below the 20 day moving averages. Most likely this would often produce very similar results as saying the 5 day moving average must be above the 20 day and 20 above the 50, or 5 day below the 20 day and 20 below the 50. I repeated the process with 50 day and 200 in place for the 20 day and 50 day for longer term trends.

Breadth Divergences of leadership:

One of the reasons I track TWO moves of each timeframe on the % movers is to look for leadership. One of the movers is a significant, but lesser extreme than the other. When I look at 4% movers I like to see these MORE bullish than the 1% movers and if there is to be a shift off lows, I like to see it on increased leadership/aggressive chasing that is more indicative of a paradigm shift than just your increase in buying equally due to temporary emotion on news without any clear leaders. If the 1+% movers are for example at 30% and the 4% start to creep up towards and above 50% first, this to me tells me there is a shift of sentiment and people are willing to chase a select few stocks higher, which may become the leaders of the future. A healthy market will have leadership emerge first, and that will give you an opportunity to get in before the leaders lift the rest of the stocks.

Breadth Divergences of TIME:The other kind of divergence is one of time. This is when the breadth signal on monthly data is bearish and the shorter term signal on the week and/or day shift bullish. The problem here will be how to interpret the data… Either it is a rip to sell into, or the start of a shift which will turn the weekly and monthly data positive and lead to a greater, longer term sustainable move. This signal in and of itself is not useful unless you can put it into proper context.

Breadth Intraday Shift:Tracking these results a few times a day can illustrate how things change over the course of the day, particularly when held in context. The best signal I have got since tracking this was when the Russian invasion of Ukraine took place. First everything was down as one might suspect, but the 4% movers were less bearish. This started to drag down even the 4% movers to become slightly more bearish at first as panic selling spilled over to whatever whas up and those down moderately spilled to more substantial losses. Interpreting this was initially difficult. Was it that the smart money that had a lot of capital to move accumulating on the fear and wasn’t believing in the decline but the broad sentiment and panic caused them to sell off, or did the smart money begin to shift as well, recognizing that things could get worse. But then towards the end of the day the 4% movers started ripping first, and turned very bullish, followed by the 1% movers turning moderately bullish. That signaled that all the fear based selling was over and the market began to recognize it for what it was… a buying opportunity on fears of a world war that may not materialize. This continued into the next day along with the news that came that Russia was withdrawing their invasion and for the time being the signal got you near a nice swing market low and if you followed along with OA’s risk cycle, it was a fortunate situation as you knew what to buy and you had a signal to get in the market ahead of the big part of the move.

 

Adding Volume Filters: One thing you may want to do is only look for stocks that are up or down on the day on volume that is significantly greater than usual 1.5x, 2x,3x. This signals more active participation than usual and will better measure of participation as opposed to thin volume movements that may not be as telling of an aggressive change in sentiment as when all the stocks advancing are doing so on increased volume. The problem with this is you can only really use it for the daily moves as far as I am aware of as it is more difficult to track volume on a weekly, monthly and quarterly basis.

 

Mark This Day On Your Calander! Overall Breadth Oversold

47 views

Since the “breadth” indicators looks at the percentage bullish vs % bearish on multiple time frames and uses multiple ways to look at breadth it is VERY rare that you see the overall average ratings outside of the normal range of say 40% to 60%.
breadth oversold
The highest I’ve seen it to compare the opposite was 74.6% on July 3rd. That created a meaningful upper range of the S&P and practically top picked the Russel just in time one last day before the major selloff, from which we have sold off over 10% as of today. The opposite reading would effectively be 25.4% on the bearish side. Instead we currently have a rating under 20% for the first time since I’ve been watching this in April and (semi) actively tracking since late June.
Many students of breadth will tell you to wait for a “breadth thrust” or dramatic and significant flip FROM oversold or overbought levels as you then have the change in sentiment which triggers shorts to chase from oversold and sellers to pile on from overbought. You also can potentially look for leadership to emerge which can be evident from the larger of the two numbers on any time frame acting more bullish than the lesser of the two numbers. However, you can also look at the opposite process of trying to dollar cost average or scale in/out as well. You might use it as a signal to transfer money to more aggressively buy the rare historic event. Of course, it is worth mentioning that bull markets tend to remain overbought for quite some time and bear markets can remain oversold for some time. Nevertheless, this type of substantial selling could represent ultimate discouragement lows…. Ultimately the trick is putting the breadth into context with sentiment and relevant context. That is difficult to do from a stale bull which has yet to receive public participation, but I am going to bet that this is a significant shakeout that has far reaching global implications like 1998 but one that is still in the context of credit expansion and a bullish business cycle with credit still remaining very loose. You could see huge ramifications from sanctions on russia, ebola, ISIS, global tensions and increased fear but you cannot reverse and manipulate the primary trend which I believe is still higher as you do not have confirmation in the S&P, Nasdaq and dow. The russel is concerning but markets don’t act in isolation forever. So while some say the russel is a (complex) head and shoulders breakdown (or double top), I say it is a head and shoulders FAKEOUT until proven otherwise.
Having the discipline to rotate capital into risk here is certainly not easy however… particularly leveraged option buying strategies which tend to capitalize off of low volatility when the volatility is high. That makes this a bit more challenging, but there is still a good roadmap of which stocks to focus on in After Hours with Option Addict and I believe the opportunity is also good for buying TNA and XIV

 

The moves are adjusted for the amount that would erase a move. For example 100% movers corresponds to 100% up movers vs 50% down movers since a 50% down move following a 100% up move would bring you back to even. A 150% move up corresponds to 60% decline. 10% up = 9.0909% down. etc

Scanning For Themes

2 views

I decided to set the OAbot up so I could get a quick glance and look for themes. It was initially with the intention that I would completely automate the process, but in it’s current format the OAbot works better as a research tool than an automatic setup generator. One of the rankings looks at the average “setup score” of each stock within an industry and comes up with an average.The setup score typically (depending on the classification of each stock) will look for strong uptrends, recent consolidation and volatility compression using monthly volatility, weekly volatility, and daily move as a % of ATR).

Although there may be other valuable metrics such as relative volume of the industry and breadth, ideally the only necessary component to identifying anticipatory entries is all stocks within a group saying the same thing. I can repeat this scan across sectors or classification types, but there is an inherent bias towards stocks that are near the highs as any laggard that just exploded to new highs will then be looked at as a momentum or quality stock. As such aside from trying to subtract the recent day move from the high to see what the stock was classified as before the move, and/or having some sort of metric to track over time at what % of each group has broken out and is no longer classified as a laggard or trash near lows, there is not much that can be done… I don’t have the time to put a lot of work into the OABOT right now as I once did.

So for now we can scan for themes quickly. The following is all industries with the # of stocks in the industry over 20, an individual setup score average over 90, and sorted by avg setup score in the industry.

Major Integrated Oil & Gas 106.474303
Oil & Gas Pipelines 105.618511
Trucking 101.401134
Residential Construction 99.1633571
Rental & Leasing Services 98.1455953
Oil & Gas Refining & Marketing 97.1936567
Semiconductor – Broad Line 96.6590651
Gas Utilities 96.1367812
Textile – Apparel Clothing 96.0117017
REIT – Residential 94.6209923
Oil & Gas Equipment & Services 93.8425753
Specialty Chemicals 93.7406842
Auto Parts 93.4825295
Property & Casualty Insurance 93.4359768
Credit Services 93.4128977
Gold 93.3070126
Chemicals – Major Diversified 93.1052147
REIT – Diversified 92.6859943
Telecom Services – Domestic 91.2492766
Drug Manufacturers – Other 91.1494354
REIT – Retail 90.8556666
Electric Utilities 90.6886449
Oil & Gas Drilling & Exploration 90.1819049
Independent Oil & Gas 90.1435737

 

The individual setup score breaks down the stock differently depending upon the classification. Stocks near their lows are evaluated differently than those near their highs. Stocks flagged as “short squeeze candidates” are evaluated partially by their float and % of float short in addition to weightings from each category of classifications. Stocks that are liquid with good fundamentals and growth prospects are looked at differently. Stocks with accelerating momentum and growth are looked at differently. There are 3 different types of “laggards” and each has a different way to evaluate the score. The score is very dynamic in that if certain things are true it is evaluated a certain way. If either of a number of things are true, it may be given a bonus to the score. If a combination of things exceed a certain value it may contribute to the score. Should I find the time, I will put a lot more thought into the exact metrics, weightings, and components that go into the score by tracking which setups look better after making tweaks over a longer period of time until I have more ideal rankings. Once I am able to further fine tune everything, and possibly even track price across time and automate the tracking, THEN I feel I may be able to construct a tool that is more automated, particularly if I implement many of the things that are computed, but not factored into the end ranking just yet. Rather than use the OAbot individual stock scores at this time, I think it is quicker just to go to finviz and scan through each of these industries until you have enough setups or identify a theme or two that you are satisfied with.

I’ve already manually from top down analysis concluded energy was a good setup a couple weeks ago… so all the various oil&gas plays coming up is additional confirmation.

I like how trucking stocks while a very diverse group (some stocks near highs, others near lows), still shows a lot of consolidation and bullish looking setups. Look at HTLD and UACL as an example.  Very different stocks right now, but both look like they are working sideways to set up a bullish move. Even among the worst stocks of the group as determined by their % off of 50 day highs are names like SWFT that are at least consolidating above a recent balance area after the sharp drop and rejecting new lows below $20 so far. Not at all a bullish chart after making the equal low longer term, but you could easily see a move to 22 before declines and recent action is at least decent considering the technical damage done to it. There are still a few stocks of the group that look like while contracting in volatility, they still have more sideways work to do for a couple weeks. I think the industry may need a bit more time for some of the leading stocks to consolidate and some of the others to work sideways or breakout and retest as the others set up, but overall it looks like an industry where you could identify a select few setups, then move onto the next, and possibly at some point you might see more of them beginning to correlate… Either way it may be a good theme to watch.

I am surprised to see residential construction score so well, I never would have even looked right now. It has been a dog of an industry but a handful of stocks are making bull flags and consolidating after a bounce from lower prices. I don’t love the industry, but every now and then you’ll see an industry flagged you wouldn’t have thought to even look at and get some ideas. It’s nice to have a preset sort of program that doesn’t have bias aside from the one you programmed in that can then be checked critically with a human eye as you have the machine that doesn’t get tired or miss anything, and then the human eye to critically assess and the intuition developed from experience (that cannot easily be programmed) to filter down the process. Since currently I am using more of an intuitive feel based identification of theme and selection of stocks, it is not a bad idea to combine objective filters such as selecting from those in sectors with seasonality that suggests we are near a low or have plenty of upside ahead.

This process takes much longer when I am analyzing it and then converting analysis to words and typing them out and it may not be exciting to read, so I won’t go through all of them. I may in the future just look to create only a volatility compression score only as being able to limit the search to groups with plenty of volatility contraction may be a little easier, particularly if I combine with a decent overall setup score in the way that it will filter down the groups to those over 85 or 90 rating first and then will sort by volatility compression scores.

 

the end for now.

 

Breadth Marches On

1 views

One of the interesting I observed about breadth recently is that off of oversold levels the bounce was so strong that it was difficult to distinguish leaders from anything else. “buy everything” was the motto just as on a longer fractal it was off the 2009 lows. Breadth still flipped aggressively and there were still a large percentage of stocks up, as well as a large percentage of big movers up, but it was not a market that you could say was pulled up by leaders, but instead one in which everything rallied, and a small percentage of stocks rallied a lot. That could be interpreted as shorts merely getting squeezed out and an over enthusiastic buy after a sharp sell off before another leg down, or as a sell off that had gotten so far oversold that the opportunity was so good that the buyers didn’t have time to be picky and do thorough research, they wanted in on anything and everything they could. The key to evaluating the breadth sometimes is patience when you are unable to have an edge in putting the breadth into proper context. Coming off oversold is tough.
A breadth flip from bearish breadth to bullish breadth is generally really good off of oversold levels, but the difficult part is assessing if it will be a market of stocks or a stock market and if the move will represent a V reversal or if those lows will be retested? Is it going to run away from here or is it going to retest resistance and remain in a range? Sometimes clarity is not provided until later on.

Fortunately, the last few sessions have seen leadership as markets have gotten more stretched from the lows. We do see more modest breadth overall, but this is clearly not “just noise” as some of the 1% movers may be and not a reactionary rally, but one driven by big movers clearly outnumbering the small moves.

For example today just looking at one day moves, 60% of all stocks moving were up today. Of those that moved 1% or more, 61.2% were up. Of all 4% or more moves, 64% were up. While the extreme moves are more rare, the more extreme the filter got, the more bullish the breadth is as a % of movers of a certain threshold. In a strongly bullish environment, the breadth being even more bullish as the moves become more extreme is typically a good sign. That may signal a number of things such as conviction to the upside, positive earnings and revisions and fundamental factors that drive stocks up, generally positive news as news driven moves may be of greater magnitude, and an increased amount of capital in the system allowing them to chase. All of these are typically bullish except when at overbought extremes in bubble territory and indicative of sentiment being too extreme when 4% movers make up a larger and larger percentage of up moves.

One sign for extreme sentiment might be breadth on a longer term time frame showing extreme up moves composing a higher percentage than strong up moves, and a large overall number. We have not seen that lately, as sentiment is not extreme at all on the longer term. We also have not seen much leadership on the extreme moves at all on longer term basis, which means this is still a swing trading environment, as opposed to the market entering run away trend mode where buy and hold for a run away move is king (I imagine the 80s and 90s showed plenty of leading growth stocks running gap and go or breakout and chase). This action in my view can be confirmed by the Russel still remaining in a consolidation range.

Monday August 11th (possibly the prior Friday) and for sure Wednesday, August 13th saw huge reversals from bearish breadth that continued to stack up over the next couple weeks. Although breadth on the daily move is not above the 80% range as it was a week ago, the fact that there still appears to be greater convictions on the big movers to the upside shows that leaders can continue to pull the market higher. The % of 1% movers moving at least 4% is around 15% so letting your winners run still appears to be a good strategy as well which supports the swing trader’s and stock picker’s cause.

Today here are some signs the bulls are winning aside from up moves outnumbering down moves and the market being up:
% of up stocks moving 4%+
5.80%

% of down stocks moving >-4%
4.92%
% of 1%+ stocks moving 4%+
14.86%

% of -1% stocks moving >-4%
13.23%
% of up stocks moving 1%+
39.01%
% of down stocks moving >-1%
37.18%

 

breadth tracking

Trading Systems Part 2: Accounting for Multiple Outcomes

36 views

In a basic “pot odds” calculation as described in the last post trading systems: pot odds and trading, if you can let your winners run so that you sell your option for 4 times the cost, netting 3 times your risk or 300%, you are getting 3:1 and thus only have to be right more than 1/4. (lose 3, win on 3 the 4th time and break even). However, options are more complicated than that. While you may be right or wrong on whether the stock goes to a price such that you earn 300%, the stock could reach the target earlier, in which case you may make more than 300%. Also, the timing could be late but direction still right, or the stock could gap up or run past the target and exceed your return. As a result, you might have a number of outcomes ranging between the extreme 1000% or so outlier trades to 0% depending on your ability to let the winner run and your tolerance for volatility. A simple pot odds calculation becomes tricky for determining whether or not the system is profitable, and how much. For this you can use “expected value”, but that really doesn’t grasp the expectancy of the system per trade over time. Nevertheless, since 1% risk represents a small amount of capital lost through volatility, it can still be looked at from the concept of how many 1% position trades are taken per year, and you can approximate that the growthrate that occurs in your portfolio is close to 1% of the expected value of the system as a decent starting point. With such a position size, you can take more trades on at once and as a result the gains can compound faster. There is still some correlation (a LOT of correlation during a significant correlated selloff), but by diversifying trades over multiple time frames and spreading out your purcahses and using the occasional hedge, you can mitigate the cost of volatility at very little cost to your overall results.

So let’s get into an example. Say you average out all your trades over 100%, You average all those over 30%, all those between -30 and 30% and all those from -30 to -90% and then all trades that lose more than 90%. That gives you 5 averages that can define your system in the total average of “large wins” “moderate wins” “break even” “salvaged premium” and “full loss”. The frequency of each event determines a number of trades out of all trades. Simply divide the number of trades by all trades for each of the 5 and you have an estimated probability of the return. Average the return of all trades in each category to get the average return.

From here I actually like to model my results as opposed to using expected value, but let’s just give an example:

Last year’s 2013 results before I booked a couple of large wins in TWTR and TSLA were

17% of trades averaged 294%, 20% of trades averaged 53%, 15.8% of trades averaged -1% 23.2% of trades averaged -63% and 24% of trades I rounded down to the full 100% loss.

To do an “expected value” calculation, you simply multiply the probability of each event by the return and then add them together. So (.17*2.94)+(.2*.53)+(.158*-.01)+(.232*-.632)+(.24*-1)=.217596=~21.76% gain per option trade. By letting the winners run in 2014 from Jan until Jun before a few large wins in AMZN and PCLN and others, this yield was just above 40% but it was also substantially helped by a few large outliers well over 500% (GMCR,AAPL,etc) . So how does that translate to portfolio growth? There is an easy calculation that can be done if the position size is small.

Let’s put it at 1% risk or 1% of .2176 expectation=.002176% per trade. In other words, we multiply at a factor of 1+.002176 per trade. (1.002176^X)-1=annualized rate of return where X is the number of 1% trades you can place with this system in a year.  Make 300 trades and you are looking at about a 1.002176^300=~1.919-1=~.92=92% annualized rate of return.

The smaller the position size, the more consistent the system, the lower the correlation between trades, and the greater the edge, the more accurate the estimation is and typically the more consistent the results are with the expectation. However, keep in mind the relationship of return and risk as you look at how it becomes less accurate with increased position size. The assumption that increasing the risk will proportionally increase the volatility and the reward assumes a linear relationship which is not true. Instead the relationship looks as follows:

Kelly-Criterion

Nevertheless, we can use a monte carlo simulation to pull a random number between 0 and 1. Since there is a 17% chance of the upper result if it pulls a 0.17 or less, this corresponds to the 294% return result occuring, where as a 20% chance of a 53% return corresponds to a .17 to .37 being pulled and so on. That can allow us to randomly pull the results of a thousand trades, and then through a monte carlo simulation tool we can simulate thousands of trials of 1000 trades and model any particular result or formula as a consequence of each result to get the intended information we need and reflect the results we want. Just by using the average return and ignoring the substantial probability of effective ruin

At 1% with ZERO fees, we get an average return of 91.35% compared to our expected 91.9%

With 2% with zero fees we get an average return of 265.7% compared to an expected ((2*0.002176)+1)^300=268%. Being able to place the same volume of trades as the risk gets too much larger becomes increasingly difficult and increases the correlation risk that is not modeled here but will negatively effect returns. As we increase the position size the difference between the average simulated return and the calculated return grow as the simulation is less and less close to the calculated returns due to losses from volatility. The other thing is, as bet size increased it becomes increasingly difficult to place the same number of trades per year.

What this growing difference does not show is the increasing skew of results. In other words, if 1000 traders were to trade this system over 300 trades at a fixed position size, a smaller and smaller percentage of traders would have a greater and greater result as position size increased. The magnitude of the outliers would increase which would skew the AVERAGE expected results. The opposite side of it would be an increasing probability of poor results as well. Effectively, the more you increase the position size the more you turn it into a lotto ticket, until you eventually take your skill edge out of it and your results actually begin to decline and eventually even become negative if you get too carried away. Even if you are adjusting the percentage risked, you should never go over the full kelly, and since you don’t have an unlimited amount of time to recover the drawdown, and you can only place a few hundred trades per year with overlapping correlation, and there is greater uncertainty… I would absolutely drastically reduce the bet from the kelly. Calculating the “kelly” is a process that has many applications and is another story for another time.

Trading Systems Part 1: Pot Odds and Trading

56 views

I wanted to go over some basics of building a profitable trading system by looking at “pot odds” a term applied in the game of poker and “expected value” and how it is interchangeable to trading. I am usually not a fan of looking at expected value because it doesn’t properly frame in the potential for massive account volatility, but it is still useful in understanding the basic requirements for a profitable system and can assist people new to trading or who have not yet learned to be profitable why they might have some “leaks in their game”. “pot odds” looks at the amount you can win, vs the amount you have at risk to determine the necessary probability of your decision in order to be profitable. From it you can also determine “expected value (EV) per hand (or per trade) ASSUMING the same amount risked every time.

In other words, say your opponent moved all in for the size of the pot on the flop. You get paid 2:1 to call since you must match your opponents bet, but get twice that (the pot AND the opponent’s stack) if you are right. You can then lose twice, win on the third and win everything you lost back. Therefore, you need only just better than a 1 out of 3 chance of winning for the call to be “profitable” in the long run.

If you have your entire bankroll at risk rather than a small percentage you are overextended and it is not a long term profitable decision because it will inhibit your ability to earn in the future as you will not be able to recover from a loss. “bankroll management” as it is called in poker and “position sizing” as it is called in stock trading is an important seperate topic that combined with “pot odds” greatly influences your long term profitability.

What novice poker players in cash games and traders often ignore is the fact that they will lose due to volatility. Lose 10% of capital, you need 11% just to get back to even, so 1% is lost due to volatility (it is actually less since multiple gains will also compound at a less substantial rate). Lose 50% of your capital and you need 100% return to get back to even.

kellycriterionbetting more than 2 times the “full kelly” actually turns a profitable strategy unprofitable over ANY time frame, and erases any skill edge you have. While your winners may also compound, your losses create disproportionately large drawdowns that require a greater skill or edge to overcome to get back to even, and due to simple chance those drawdowns are certain to occur over a long enough time horizon with a correlated, unhedged system. Your edge is not as profitable as the expected value calculation in reality, at least not without introducing a “risk of ruin”. But I digress, let’s keep things simple… Pot odds. To keep things simple and ignore “uncertainty”, I want to use a situation in trading which the numbers tend to be a bit more concrete as opposed to uncertainty, so I want to give the example of “risk arbitrage” through buying into mergers and acquisitions.

According to this blog entry, 97 deals were completed, only 2 deals failed. However, many others remain unresolved and sometimes it may take longer than anticipated to close which doesn’t eat into your total profits for the trade, but does eat into your annualized profits since it takes longer for you to be able to reuse that capital. Since we are concerned in this article about profitability and pot odds, time will not be factored in. Even though almost 98% of deals closed, we will leave a bit more margin of error since some deals can linger on for years and then not close which could skew the numbers a bit. To be safe, we will say over 90% of deals go through and we can say that with pretty substantial confidence. At 90% chance of being right to win X and 10% chance to be wrong and lose 100% we must solve for X such that -.9x=-1*.10 or.9x=-.1 or -.1/-.9=1/9=.111111=11.1111%.

If you would lose 100% and win 11.111111% when you are right, the strategy of buying into these deals post announcement would be break even. So if you put your entire capital at risk, you would have the “pot odds” to buy any time there is a 11.11% payout or better. But when deals fail, they don’t go to zero, they drop down to around their pre-buyout price on average. This can change from deal to deal and some deals may just be in a constant downtrend.

If you had a downside of only 10% and an upside of only 5%, how would it compare with a trade with a downside of 100% and upside of 20%? You can confirm the expected value by downside of a loss (expressed as negative percentage) times probability of a loss plus upside of gain plus probability of a gain.

System 1 10% downside, 5% upside:

(-.1*.1)+(.05*.9)=.035=3.5% per completed deal.

(-.1*.1)+(.2*.9)=.08 or 8% per completed deal.

This shows why I don’t like expected value in comparing two systems, you risk insolvency by putting 100% of your capital into the trade with the “higher expected value” where as you only put 10% of your account at risk with the “lower expected value trade” If you were to risk 10% of capital on the 2nd strategy, you would only grow your portfolio by 0.80% per deal. As such the deal providing the better risk reward assuming equal probabilities of success/failure is almost always going to provide the better return on risk. You must evaluate system on an equal portfolio risk basis to truely determine which is better. The kelly criterion is a good metric for comparing one system to another on an equal risk basis. However, the kelly criterion should not be used for position sizing as it is probably 5 times more aggressive than it should be (or more) due to the false assumptions the formula makes about having an “infinite time frame”, a “certain, predefined known edge” and “complete emotional tolerance for all volatility that does not effect your edge” and that multiple bets at a lower percentage with a low correlation provides a better return on risk overall.

Additionally, if you could complete 12 trades a year in the 3.5% expected value system and the  8% per deal system took over a year, it would be much more profitable, so a raw calculation of “expected value per trade” must first be “normalized” (in this case normalized means something very different than “normalizing” volatility” as instead it means it should be set up to reflect the “normalized rate of return at a given level of risk” to reflect overall growth on portfolio given equal risk via position sizing). Then it must be also “annualized” so that they are equalized on a particular time frame after fees… However, due to uncertainty, you also must consider that the large edge that compounds fewer times is less vulnerable to small changes that may negatively impact the ROI more than you thought.

I don’t want to get any more complicated than I already have in this article, but for now I will tell you to practice by looking at your “risk” and your “reward” and measuring out your probabilities required to break even. For example if you have a target price that nets 3 times your risk, you need to hit your target 1/4 times in order to break even.  With options if you try to hit around “break even” on hitting your targets or even lower, the trades that expire in the money and have to be sold for a slight gain, break even and only a slight loss as opposed to the entire risk of the option PLUS the occasional trade that runs beyond the target will create profitability in the system.  I may get into more specific and more realistic trading systems in a future next post.