Saturday, May 28, 2016
Joined Oct 26, 2011
136 Blog Posts

Intermediate Term Consolidation

vicegrip

The short term consolidation score looks at:
Daily Change
Daily Change Relative To ATR
Daily Change Relative To Weekly.

The short term represents any movement less than 1 week, but really only involves the daily moves because of lack of data between 1-6 days. It also scales the measurements so that as a stock moves less, it gets more points added to the score.

The intermediate term consolidation gives us more data.
Here are a few possible considerations to score:
1)Weekly change vs monthly performance
2)Weekly volatility vs monthly volatility
3)ATR (14 average daily move) ratios…*
4)Weekly volatility vs beta
5)Weekly volatility overall
6)Weekly performance vs beta
7)Absolute weekly performance not too extreme in either direction.
8)Distance from 20 day moving average?
9)Distance from 20 day moving average relative to some volatility metrics**

We can tinker a bit more with how one thing relates to another in different layers to provide a lot more clarity on whether or not the stock is undergoing volatility compression or volatility expansion in the last 7-20 days.
———————-
*ATR divided by price gives us a percentage daily move as an average over the last 14 days. While this only refers to the daily movement, it’s a function of 14 day volatility on daily basis. If the weekly volatility is smaller relative to the ATR (14 day volatility), it tends to represent a stock that is contracting in volatility. If the ATR is smaller than the monthly, it also tends to represent a stock that has contracted in volatility over the last 14 days more so than the last 30. You can also apply a beta adjusted bonus so high beta stocks (stocks with typically more volatility over a longer time frame) that also have a low ATR (less volatility over last 14 days), score well.

**A stock that is closer to it’s 20 day moving average will tend to either have moved less over the last 20 days OR be near it’s “balance” in it’s range suggesting it’s close to the Apex if the chart is fully consolidated. However,a stock with a higher beta, or ATR, or monthly or weekly volatility may be able to tolerate a little bit more movement from the 20 day volatility and still have no confirmed breakout or breakdown. Also, a stock that has moved less in the last 14 days or week than it has over it’s 20 day average may be consolidating. The closer a stock is to the 20 day moving average, typically the less it’s moved over those 20 days, or at least the more likely that it has stayed in an equally proportionally rangebound area or regressed to the mean recently.

Also:A stock in an uptrend with a 20 day moving average under the 50 day may represnet short term weakness and consolidation within an uptrend. A stock in a downtrend with the 20 day above the 50 day may represent a stock that may be consolidating and possibly forming a bottm, particularly if the stock is also above the 20 day moving average. This sort of deduction from data probably won’t be used, but may be a decent idea to pair with consolidation data to find chart patterns. It would best be used in a binary (1 for yes, 0 for no) and as an additional filter (E.G. you can set up a table to show you stocks that score over 80 that also pass this filter of “patterning” from uptrend.) Defining the stock as being in a longer term uptrend can be based upon a stock being above a 200 day moving average or 50 day being above the 200, distance from 52 week low, distance from 52 week high or other data. A little off topic here so I will scratch it.

A volatility measurement tends to measure a stock’s standard deviation of movement over a particular time frame. A stock will move within 1 std deviation ~68% of the time if it is normally distributed. 95% will be contained within 2 std deviations and 99.7% of the data set will be contained within 3 standard deviations. Whether or not stocks movements are actually normally distributed has famously come into question by Nassib Taleb in his books “Black Swan” and “fooled by randomness”, but for the purpose of measuring volatility, this doesn’t matter unless we are going to sell option premiums expecting moves within a particular range.

This is a bit different than looking at a stock’s average movement, but still relate to movement over a time period.

So as long as we evaluate each stock with the same measurement, comparing ATR average daily movement over 14 days) to weekly or monthly volatility (standard deviation of movement over the time period) should still get us an idea of how volatility has changed over time.

By looking for contracting volatility over time and comparing volatility over various time frames as well as the magnitude of the change, we are able to look on a 1 day, 1 week, 2 week, and 1 month period of time and compare volatility relative to a stock’s long term comparative movements relative to the S&P (beta), and not only get a good idea of whether or not a stock is located within normal bounds of a range, but whether or not those ranges are contracting, to what extent they are contracting, and how the recent “quietness” compares to the historical movement of a stock.

volatility time volatilty compression2

The pictures above don’t fairly represent how our scores also consider relative movement, relative volatility, and average daily moves and standard deviations over time periods. Overall, using all the data presented will provide a much greater chance of having consolidation patterns rise to the top than may be represented by the images. I covered this in a different post.

Again– once we gather this data, we can have the spreadsheet sort out the data by certain categories– We then can make adjustments to the final individual score based upon the average of the group. A stock showing signs of consolidation in a group with an amazing consolidation score may be worth more than an “ideal” setup in an average group or worse. When you see multiple stocks in an industry setting up, it’s less likely to just be randomness or a fake. If every single stock in a group is setting up this represents a large amount of capital preparing to rotate into a theme. One or two signals can be wrong, but say 20/25 names in the group setting up? The probability of you catching an idea before it makes a big move is greater.

I feel similarly about seasonal data. Seasonal data on an individual stock may be due to a few quarters of earnings at a certain time that happened a few years and skewed the data. But seasonal data that suggests both the individual stock outperforming the industry and an entire industry will outperform the sector, and the sector will outperform at a particular time of year is more likely due to a causal relationship such as capital moving in reaction to or in anticipation of holiday shopping leading to an increase in earnings and a fundamental reason as to why a particular stock has been more effective at capitalizing on this seasonality than it’s peers in the industry. This is a little off topic here as well.

I don’t necessarily need to understand the cause as long as there is evidence the move will continue to correlate with the timeframe as opposed to it just happened to correlate without cause that will be operationally random moving forward.

If you don’t understand a cycle and are just selecting a point your results will be normally distributed as if it were random. But if instead it were a cycle with waves that expand and contract in duration, and you were able to identify this cycle and buy closer to the low and sell closer to the high, then you would be able to show superior results. Your results would come at the expense of someone on the other side, so again, analyzing the results would show a normal distribution. If you are an outlier to the upside, someone or a group of people will be an outlier to the downside, and overall the data set would be within a range with little evidence that you were actually able to exploit a tendency. That is the nature of a interconnected system where the wins of one person correlate with the loss of another around a collective average. A little off topic again.

So we basically have a good outline for the different things we can consider when making our intermediate term volatility rank which I have begun working on. I have finished the short term consolidation rank unless I decide that something in the intermediate term rank belongs in the short term or I come up with a different idea of what I can add.

Comments »

Short Term Consolidation Score – Daily

volatilty compression2

Consolidation Rating The Heart Of The OABOT2.0 –

If the heart of the OABOT is the consolidation score, than the short term consolidation score are the aorta and other major arteries, and the intermediate term is the blood, with the long term being the veins.

Although there’s not too many possible data points that go into a daily score, there’s a surprising number of pieces of the formula that can be applied by combining these variables. The short term consolidation rank helps punish stocks that are breaking down or breaking out, while rewarding stocks that are moving less today than recent average movement over various periods of time. When combined with high intermediate scores, this generally allows you to identify stocks that are closer to the apex. The stocks that show an “inside day” tend to rise to the top a little more often with all other things being equal. It also avoids looking at stocks that may be consolidating, but either are breaking out or aren’t “quiet” today, potentially suggesting they may need more sideways work, or that it may be too late to chase.

Before I get into the nuts and bolts of how I am preparing the daily score, I want to show you how this spreadsheet will be “cleaner” than the last one, and thus, much easier to tweak.

I have a “data” tab (where the data can be updated and replaced with new data completely), an “all tab” where all the data is repeated so that formulas can be applied, scores can be added (and re-added if/when I put the summary tab back on), and the “formula tab” where specific numbers and details within the formula can be tweaked.

Rather than looking at some nonsense like:

=IF(OR((all!BO2*BN2)>(formula!C43*all!AX2),(all!BO2*BN2)<(-formula!C43*all!AX2)),formula!C44,formula!C45)+IF(OR((all!BO2*BN2)>(formula!C46*all!AX2),(all!BO2*BN2)<(-formula!C46*all!AX2)),formula!C47,formula!C48)+IF(OR((all!BO2*BN2)>(formula!C49*all!AX2),(all!BO2*BN2)<(-formula!C49*all!AX2)),formula!C50,formula!C51)

and then having to interpret the nonsense and substitute certain variables and change the formula itself, I have instead set it up so the “all” tab pulls from both the data and the formula tab where necessary. Even though the formula still looks like crazy nonsense, I don’t have to look at it and deconstruct it.

If I want to make tweaks, I look at this clean version where I can read what’s happening, and I can make adjustments without adjusting the formula entirely. If I want to make structural changes to how the rule is applied or apply a different rule, I can more easily read what’s actually taking place and what cell on the all tab contains this formula.

formula

Now we can resume talking about the short term consolidation rank tab….

For the time being, I broke the short term consolidation range into 3 subsections. Ultimately I am combining them all, but if later I decide I want a weighted average so that one of them count more towards the final score I can do that.

The categories are:

1)Changes as a multiple of the ATR (A stock that moves less than 0.5 times the ATR will score more than a stock that moves more than 2x the ATR)

2)Absolute Changes (A stock that moves somewhere between -1% and 1% will score more than a stock that moves between -4.5% and 4.5%)

3)Changes relative to the weekly performance.

The 3rd variable is the trickiest, but important. To avoid confusion, I chose to classify the weekly performance as “narrow weekly range” for a smaller weekly move in either direction or “wide weekly range” for a larger weekly move in either direction. I also classify it as “positive” for a stock that is up on the week, and “negative” so I don’t get in trouble with the less than and greater than sign when negatives get involved.

This helps me avoid confusion in making the formula since you may have a weekly positive but a daily negative or vise versa and writing the formula without separating these gets tricky.

Let’s say you want the daily range to be within the weekly move. Let’s say a stock moved a positive 1% on the week. If a stock moves less than 1% but more than -1% it represents consolidation. However if instead the number is negative the same formula would say less than -1% but more than 1% and represent the opposite of what you are trying to express. So by setting up separate formulas for positive and negative numbers it makes things a bit easier.

If a stock is not moving strongly on the week, and is showing signs of the stock moving less today than that weekly action, it’s probably looking like a better target. This is more true if the daily action is less than half of the weekly move, and more true if the daily action is less than 0.25 of the daily move.

While it’s possible a stock started the week moving up 10% and then went down 8% to finish only up 1.2% suggesting less volatility than there actually is, if the stock on the day is only moving 0.30%, you are still looking at a stock that is less volatile on the day than over the course of that week.  Not all stocks that show less movement on the day than the week will be consolidating, but most stocks consolidating will see less movement on the day than the week.

When backed by weekly (last 7 days) being less than a multiple of the ATR, it begins to paint a picture of stocks that are in narrowing ranges, even if that contraction may not be orderly.

volatilty compression2

The above graphic shows that inverse head and shoulder patterns as well as the really steep moves with significant declines on the week in falling wedges may not always pass with a high consolidation score. Also there may be some stocks that pass within the thresholds set but don’t actually show a pattern.

We can try to get the falling wedges after significant increases in price to have a boost by relative comparisons of monthly volatility relative to weekly volatility and movement relative to others.

Capturing the inverse head and shoulders look from a low can actually be done by classifying stocks properly ahead of time so that it’s near the low but has started to see recent strength, but there will be a lot of false positives and a few false negatives. For example, if you have a stock that is close to the low, has seen some movement from the low, is above 20 day moving average with recent weekly move not surpassing a certain amount and hasn’t declined too much and a very mild boost for weekly volatility being less than monthly, and daily volatility data doesn’t matter, you may capture a few that have that look… However, I won’t be using OABOT to try to create a special formula to try to find these patterns at this time.

We are looking for stocks that have not passed certain thresholds on different timelines and getting a bit more strict about the movement as the time range decreases, and using different criteria and different measurements for volatility and movement.

We are looking for contracting volatility. That doesn’t necessarily imply there will always be a pattern of contracting volatility just because volatility on the month is less than the week and the day is less, and the ranges are contained, and the ATR is small, and the beta adjusted movement on all time frames is low. But it’s an excellent start.

This is also why we must look at the same question in different ways. It’s not enough that a stock passes a filter of not breaking out to the upside, but also that the daily movement relative to absolute terms, recent average volatility on the day, and relative to the week suggests consolidation. As the data suggests more consolidation we continue to improve the value of stocks with a high consolidation score.

A  stock may have started it’s consolidation awhile after the cutoff. For example a high tight flag will see the stock advance for 3 months straight before it consolidates so both the quarterly and monthly performances may not be suggesting consolidation. If the consolidation has only just started midway through the week, the weekly may not even capture it.

Nevertheless, in terms of finding stocks in consolidation patterns with good setups near the apex, I was able to get about a 40% hit rate on stocks with very high consolidation scores the first time around, and that’s when I wasn’t as nuanced with the short term, intermediate term and long term consolidation scores. However, the VIX at that time I believe was under 12 so setups were abundant.

That was when I also started to included separate criteria to evaluate stocks near the high differently than near the low, but I actually think I can eventually get it to where I have a higher hit rate than that.

This means in the right conditions can look through the top 100 stocks and find around 40 that are at least starting to set up or show some kind of consolidation. Not all of these 40 are going to be worth trading, and certainly many of them won’t trigger a trade, but they at least capture the “look” I was going for.

I don’t really need it to be that high, even in low volatility environments. However, starting out that high gives me the flexibility to be strict about other data that will represent other things I like about the stock such as being in an industry with a good score as well, or being in an industry that has performed well in the long run that is starting to show weakness AND a high setup score, or focusing on the higher beta stocks or better swings in the past, or looking at a separate filter to evaluate a stock’s “risk” and compare it to the phase of the risk cycle.

Reducing the workload from 7000 stocks to 400 or less nearly automatically is the intention, while also tracking a few things that may be useful in providing a “market overview”.

Comments »

Consolidation Rating The Heart Of The OABOT2.0 –

tinman

tinman

Perhaps the very most time effective things I’ve done with the OABOT is the Consolidation score.

The consolidation score looks to reward stocks that are consolidating with higher scores. It looks to avoid stocks that have already started the breakout, and identify stocks that have shown at least some evidence of consolidation. Not all of the stocks with a good consolidation score will have the look I want, but it certainly significantly helps narrow the field.

volatility compression

By applying a consolidation rank individually, it helps identify and rank the stocks in order of those who have the most coiled up setups.

By slicing a stock into numerous classification and then looking at each of those classifications, you can then apply an “average classification” for each category. You can also look at the total count of stocks in that classification and the number of stocks with a consolidation score in the top 20%, and come up with a percentage out of 100 that each “theme” has. You can then use excel to sort industries (or sectors, market cap size, risk classification, etc) to see which “themes” are setting up.

When I created OABOT the first time I complicated things a little bit by evaluating stocks differently based upon where they were in relationship to their 52week high, moving averages, from lows, and performance in the days, weeks, month, quarter, year, % of shares that float, short float %, and other factors to label each stock into categories. I still think that’s valuable, but in the process, I think I got a little bit away from the ability to quickly find “themes” in that I combined “what’s working now” with relative volume of themes to multiply scores and combined everything together.

The reboot of OABOT will possibly look to simplify and just keep it contained to identifying

1)consolidation rank on short, medium and long term basis as well as overall consolidation score

2)Consolidation score by “groups” or “themes” (sectors, industry, market cap size, Location, etc)

3)The number of themes that have a certain NUMBER of scores over certain thresholds (For example, what percentage of stocks in each industry have a consolidation score over 80? Over 90?).

If every stock on average has a better score than the rest, OR if there are multiple stocks in the best 20% of setups or best 10% of setups, (even if the others are bad and skew the average to be low setup score for the theme) these metrics will help confirm that institutional capital is likely being prepared to enter a theme, even if they perhaps are being a bit more selective about how many within the theme they’re looking to enter.

This time around, I want to make sure I compartmentalize different aspects of what I’m looking at.

  • I want to quickly identify the next theme rotation based upon consolidation score only. At some point I might also be able to separately look at theme performance and what’s working well now so I can manually glance at them and see if they have anything in common “beneath the surface” (are they all showing a particular technical chart pattern or candlestick pattern? Is there a fundamental reason or news related reason why certain stocks are outperforming? Is there some idea that might be next based on this knowledge?)
  • Then I want to separately look at the top stocks in some of these top performing themes only; as well as separately looking at the top consolidation setups within themes that are working; as well as the top overall consolidation scores independent of industry.
  • It will eventually be valuable to also look at the risk rotation of stocks, and the “What risk cycle is working” vs “what risk cycle is next” in each theme in a more anticipatory fashion. Although this is far more time consuming as it requires a layer of classification, analysis of the theme, and then reclassification providing a second layer of classification.
  • And then look at the holistic score that factors in everything as an average.

The consolidation score has to look at a lot of different variables and apply them in different ways. Of the finviz stats could be used in a way that makes them relevant to consolidation:

1)Monthly Volatility

2)Weekly Volatility

3)ATR

4)Change (daily)

5)Beta

6)Weekly performance

7)Price

8)Monthly Performance

9)[Distance of price from moving averages]

10)Ratios and multiples of one score to the other.

I don’t believe I will be using the distance from moving averages, but It’s interesting to think about. A stock near the 20 day moving average relative to the ATR suggests movement in the last 20 days has been either quiet or nearly evenly distributed among the upper and lower range over the last 20 days. Or a stock above it’s 50 day but below it’s 20 day may suggest the trend has turned upwards, but the stock has recently begun moving downward slightly… If backed up by consolidation score, then it might be a wedge pattern or flag pattern… but it’s hard to tell just based upon this data if it’s a pullback or a orderly bullish consolidation or a sharp declined FOLLOWED by orderly consolidation (bear flag).

On it’s own it doesn’t necessarily suggest consolidation, but combined with consolidation it suggests recent order or that the stock may be near the apex of the consolidation. Perhaps instead sorting stocks over a certain score by distance away from 20 day moving average might be a better approach than factoring it into the score.

Additionally, it’s possible that a stock far above the 20 day moving average that is consolidating substantially in the last week might be bullish while the stock far below the 20 day may not. This is because the one far above that is consolidating may be a bull flag, while one far below is possibly a bear flag. So if you are going to use moving averages, you have to combine them with other data and possibly create separate classifications that combine certain elements.

I will probably do without the moving average data.

You want weekly volatility less than monthly volatility, price change less than 2x ATR but more than -2xATR, weekly change less than some multiple of ATR and more than the negative. When using ATR you either need to convert the change in % terms to $ terms or convert the ATR to a percentage terms. Daily and monthly volatility under a particular amount. Higher weekly and monthly volatility may be acceptable if the beta is higher, so you might adjust beta.

The ratio of weekly volatility to monthly volatility can influence how much you score the stock. The smaller the weekly and the larger the monthly, the more consolidation that’s taking place in the last week relative to the month.

The way you create a consolidation score is using IF formulas and possibly “AND”, as well as “OR” formulas in excel. For example One formula in excel would allow you to say

“IF weekly consolidation is less than monthly consolidation then add 20 points, if not take away 5 points.” This could be expressed in a format like this

=IF(AY2<AZ2,20,-5)+[other parts of formula that add or subtract a score]

where AY2 is the cell that lists the weekly consolidation, AZ2 is the cell that lists the monthly consolidation, and the brackets just represents a placeholder for other parts of the consolidation score formula.

You continue to apply certain rules that reward or punish, but not eliminate stocks. Once you are done, you can filter out stocks with a total score below a particular threshold.

Rewarding the monthly, weekly, and daily performance above negative absolute thresholds and below positive absolute thresholds as well as ATR or beta adjusted thresholds would be another way to punish stocks that are acting too fast and loose over a particular time frame. You could use quarterly and yearly if you wanted to as well, but that should be devalued or excluded to a long term consolidation score that doesn’t apply very much if at all to the overall score.

I welcome any challenge on my ideas or alternative ideas on how best to convert data into a system that rewards stocks that consolidate to try to filter out stocks that are definitely not consolidating, and rewarding stocks that probably are.

A lot of the ATR vs daily score, absolute daily scores above a particular negative and below a particular positive will apply towards the short term data. Possibly if weekly move is above a particular amount, if the daily movement is greater than weekly, apply a penalty. A lot of the weekly data and weekly vs monthly and ATR (14 day) can apply to the intermediate term. The long term can be mostly monthly data and later.

Comments »

Reviving OABOT

cpr

cpr

As many of you know, I made the decision to undergo a difficult project to try to convert some of OA’s teachings into code.
Along the way, I abandoned the project after I could no longer update the data with a single push of the button due to finviz automatically redirecting the page which would export to a sign-in/register/pay us money page.

The main focus was on volatility contraction and scoring a stock’s classification according to it’s own movement.
The next stage was to use industry metrics as well as location, sector & market cap size metrics to apply a bonus to those in an industry or theme that was “working”.
I also applied a bonus to stocks that were undergoing significant volatility contraction.
An additional bonus was layered for “themes” that were all showing significant scores (and likely saw significant volatility contraction as a group as well)
—-
The next stage which I had begun working on was reclassifying stock according to how it performed relative to the industry.

After taking much time away from the market followed by some time  trading without this spreadsheet functioning, I have made the decision to find a work around and get back to it.

With a fresh set of eyes comes a new perspective.

First,
The formulas got too taxing to make adjustments, especially if I want to make quick, temporary ones. I had so many formulas on top of formulas that I lost track at times of what the numbers in the formula meant. That made things very difficult to change.

The solution to that will be to create a tab separately for adjusting the amounts in the formula and explaining in English what the formula does. This way you can tweak a number on the formula page and it will automatically adjust without having to re-copy and paste the formulas and adjust the entire formulas.

Secondly,
While determining how stocks in a particular “theme” are doing relative to each other is useful, I need to do more to classify the “risk” level of a stock. Certainly determining whether a stock is consolidating off it’s 52 week high or 52 week low makes a difference in the “look” you are going for and what type of qualities you are looking for. However, I need to better fine tune definitions with seperate metrics that score the “risk factor” of a stock, it’s industry average and how it relates to it’s industry average (as well as other areas it may have in common with other stocks), to come up with an overall weghted average that converts to a 1-5 score.

I also want to use earnings data to sort of be able to X out stocks that have earnings coming up in X days.

Finally,
I want to either possibly run some kind of beta test and invite all the IBCers to try it free for some time, and/or possibly even build it live, explaining the details of what I’m doing and effectively making it open source.

———————

Less importantly, I will no longer need the sort of time sensitive data since I won’t be able to use it as efficiently as I would have hoped.
I had worked on breadth statistics of groups to try to find the stock in a group that wasn’t moving when the group was, and also try to classify where the group was in the “risk cycle”.
This was most useful for stocks in larger industries, but I’m going to focus on being productive and ignore the more taxing parts of the projects for now.

———————————

image
If I happen to finish these steps, the next priority will be to try to “normalize” the scores from -100-100 or -1000 to 1000 or something, and then try to make a very rough price projection.
The price projection is not expected to be accurate, but instead useful to help determine the larger rotation of where the capital is headed now, and what those movements will mean for the setups next.
The idea will be to be able to have ONE dimension of metrics based on the here and now, and another dimension of metrics based upon the anticipated movements….

In other words, what happens IF the price action occurs as projected? How does that change the scores of all of the stocks, and what’s the result of the next projection and the next one.

If a stock is consolidating, a 1-3 price move influenced by capital flows into the industry or related elements could be enough to give it the boost to understand when to expect a breakout, and a second iteration of that combined with the 5-10 day look would allow you to look forward and finetune the “YOLO” trades a bit more accurately as well as get a better picture of which 5-10 day moves are likely to fizzle, and which will do well.

The 1-3 strengthens the 5-10 day projection and narrows the timeframe, as well as helps filter the 2nd and 3rd iteration to eventually a 30 day projection

Without the time sensitive data being timely available–that influences the 1-3 day projections–in this next version, I’m probably going to instead focus on ~5-10 day projection as the shorter term and ~10-20 as the longer term and have a couple iterations forward to give me about a monthly projection, and maybe a few monthly projections to give me quarterly, and a few quarterly projections to give me a year projection.

The idea of being able to project some of the best OTM LEAP option purchases and quickly identify the run away trending stocks before they happen is very intriguing to me, but I wish I had some of the shorter term “character” movements that help anticipate sort of the breakout and fizzle, break and chase, fake out and break out, and the false moves to fast moves that really better help define what to expect and how to handle it.

Unfortunately it takes a very long time to set up the calculations of how breadth within an industry and volume within an industry compares to the individual stocks and then looking at which moves are lagging on the days and weeks and determining which stocks are working in the cycle and which are next… and it can only really be used for industries with the greatest number of publicly traded stocks; which is about 10% of the market.

So this time around, I’m probably not using it.

Comments »

Topping Pattern

dow

In the post titled The Unconfirmed Top, I went into why I thought the risk/reward favored the bears.

I sitll believe that to be the case, and now am thinking there’s a real good entry right here.

Take a look at the dow.

dow

Major markets as a whole
toppy

The silver lining is in the optimistic line drawn in the dow and that this is a giant fakeout combined without having a euphoric, huge volume top.

However, with such thin volume, it also wouldn’t take a lot for the stock market to move upwards or downwards quickly with significant buying or selling pressure.

acwi

The all world index looks bearish as well…

However, again, there’s a silver lining. In this case, it looks like the market swiftly rejected head and shoulder breakdowns across the board.

The failure to breakdown can be very bullish IF the buying pressure allows us to take out highs. This could trap a lot of shorts and provides fuel for an explosive rally over the coming several months.

The alternative is that we fail to get past highs, bears reinitiate and eventually the buyers that bought the breakdown give up buying and capitulate which leads to swift selling.

So it’s about risk/reward, and I believe for the time being the entry can be managed most effectively to the downside here.

Comments »

Trading Marubozu Candles

marubozu

If you are leaning too strongly bearish or bullish and want to find a quick hedge in the opposite direction once things begin going against you, you could do a lot worse than a marubozu pattern. A Marubozu pattern is candlestick pattern where a stock trends from it’s open to the close without reversing directions long enough to take out the low or leave a “wick”. If it’s a white candle, and without taking out the high if it’s a red candle.

White Marubozu – bullish.
Although more than 50% of the time the stock continues upwards, the real trade works when it actually fails to follow through and then breaks down. This is because the strong move upward indicates a lot of capital tied up chasing the stock higher, and once this capital is underwater, stops begin triggering and selling continues. It is a confirmed “breakdown” once it takes out the low, which is also the open the day of the marubozu. It also often creates a measured move downwards pattern on the intraday chart. Look to trade it within the last 5 minutes before the close if it’s below the marubozu candle’s low rather than waiting until the next open where a gap down may cause you to miss out.

marubozu

White Marubozu stats:
white marubozo

You can see the stats show an edge even when you’re trading against the current market environment, making it a good hedge in bull markets. It also is very common to find this “bullish candle” in bull markets and likely you’ll be able to grab a put on a big down day intraday or at close if there’s a selloff.

Black (red) Marubozu – Bearish
The black marubozu indicates a bearish breakdown that likely continues, but when it takes out the highs, you potentially have a short squeeze and sellers who are under exposed and want back in. Also, the strong trend down and movement required to reverse both requires a change in sentiment as well as it tends to create a measured move on the intraday chart.

Black Marubozu
black marubozu

This is actually a trading system that can be very straight forward, and is especially useful if you focus on weekly options for a 3 day move or so on stocks with historically cheap premium.
How it works:
Black Marubozu watch list:
1)Save a word document as “marubozu” and enter today’s date.
2)At the close of the day, use this screen and convert the list to tickers http://finviz.com/screener.ashx?v=111&f=sh_avgvol_o50,sh_opt_option,sh_price_o10,ta_candlestick_mw&ft=4
3a)Save the list of tickers under today’s date.
OR )Enter the tickers into finviz and save the link that links only to those tickers in your document
OR )Set a price alert at the low of the day for each stock
4)Each day, remove tickers that have followed through or already triggered the trade and repeat the process.
Using the watchlist
1)View: Either click on the link you created in the doc or copy and paste the tickers (or wait until you get a price alert.)
2)Once the order is triggered, evaluate for a trade, in the last 5 minutes in normal conditions. (potentially you may consider a system that trades intraday if there’s a correlated sell off or negative breadth, or market index just took out the days low or if you have some sort of additional confirmation)
3)Optional:If the close is back above the low, exit the trade.
4)Otherwise sell in ~3-5 days or when price target is met.
—————
5)Also consider anticipatory trades by running the scan 10 minutes before close and trading in the last 5 minutes, or taking an unconfirmed marubozu near the marubozu high. Stop on a close above the marubozu high. You will need to wait longer for pattern to confirm, but have a better risk/reward.
6)Also consider anticipatory trades IF the stock has failed to break out in 5 days with stop above the 5 day high.

———————————————

Comments »

Bond

tlt
On the “weeks to months” time frame, TLT may be making a topping pattern. You have a huge concentration of capital culminating in a enthusiastic type of move. You have an equal low made, and the next rip offers a manageable entry. You’ll know pretty quickly if you’re wrong, and it’s possible it’s just setting up for the next leg higher as you could interpret it as a wedge pattern as well. The enthusiastic move above makes me think lower for now.

However, the longer term time frame shows the makings of a developing parabolic market with a steepening trendline. Maybe that topped out in 2015 and simply retested that top in early 2016. Maybe not.

Meanwhile, corporate bonds are breaking out, seeing capital inflows

corp

The ratios of TLT to corporate bonds shows it may just be temporary, and pulling back to the trendline. If the trendline holds, you would see TLT outperform corporate bonds as the ratio trends higher.

tltratios

Nevertheless, there’s a huge amount of capital that is moving to and from bonds, depending both on capital flows as well as potentially indicating inflation of capital and leveraging up, or deleveraging, so it will be an important development to watch.

From time to time bonds flips correlations with stocks as well. Either both bonds and stocks trade up in a bull market of growing wealth and increased leverage in the system. Or bonds is the “flight to quality” or “risk off” trade during tough times, while stocks tend to be the “risk on” trade. If the treasury bond markets gets too saturated on this generational cycle, it’s possible that everything we know about the relationships between stocks and bonds will eventually flip for the first time since the early 1980s. The flip would be the first time bond yields reversed a downtrend and began an uptrend since the 1940 low. 1942 was the first higher low since the 1933 low. Both stocks and bond yields trended higher (bonds prices lower) from 1942 to around 1969. A bull market with rising interest rates may result. Corporate paper or just cash may be the flight to quality trade instead… But perhaps that doesn’t happen for longer still.

Comments »

3/21 chartblast

As usual, I start with a list of optionable stocks with available weeklies and look for some kind of setup. Usually I am looking for a 1-2-3 reversal.
A bullish 1-2-3 setup will form a low (1), break the longest downtrend from highest high (2) while ideally forming an equal high, and then set up by dipping to form some kind of dip or consolidation (3), which will also usually be consistent with an “aversion” type of setup.
A bearish 1-2-3 setup will form a high (1), break the longest uptrend from lowest low (2) while ideally forming the first equal low, and then set up by ripping or forming some kind of consolidation towards support (3), which may be consistent with a “subtle warning” on sentiment chart.
—————–
a few bullish setups

finviz link
3-21

Some of these bullish setups have not yet broken the longest term downtrend from the highest high, but are either consolidating some kind of triangle bottom, or at least have broken a shorter term downtrend and have set up.

I still have more to look through but for now here are some potentially bearish setups

link here
3-21bear1

Some of these are just potentially setting up for a secondary correction within a longer term bull market as they have not yet broken the trendline from the lowest low, and I haven’t yet looked at every shorter term chart to verify the setup, or looked at volume profiles just yet. I will try to look through a deeper list to come up with a better bearish list on Tuesday or Wednesday if I have time.

 

edit 3/22: here’s a few more bearish setups.
Added panw,hig,dow,crm,vlo to link above

 

3-15bear2

Comments »

Chart Blast Ideas 3/17/16

Posting both weekly charts and daily charts to try to show you what I’m looking at.

Method: Manual scan of all stocks that have available weekly options. Using a list that I update 2-3 times a year.

Bullish setups:Usually looking for 1-2-3 reversals. Typically the first low following equal high.. Typically focusing on finding “aversion” sentiment, or in EW terms looking to buy before wave 3.

Bearish setups: Typically focused more on selling retest or resistance and/or with volume profile below and/or with some kind of bear flag or rising wedge.
Bullish list.
Tickers:AMBA,AMRN,BAC,BIIB,CTRP,CYH,DATA,DECK,FEYE,GALE,GILD,ICPT,JPM,MU,THC,TWTR,VIPS,WDC,YELP,
Bearish list.
Tickers:
COF,EA,INTC,NXPI,TRIP,WTW,
CREE,DD,DOW,UA,XLNX,
(not yet) ZTS,CYBR

Bullish (daily charts)
bullish daily
Bullish (weekly charts)
bullish weekly

Bearish (daily charts)
bearish daily
Bearish (weekly charts)
bearish weekly

Comments »

How Edge Influences Position Size/Strategy

volume profile

Volume profiles provide objective support/resistance levels that can measure points of possible future orders, with the psychology of whether or not we are above or below the price often determining who’s in control.

Lets take a look at a hypothetical volume profile and what zones of price mean for the stock.

volume profile

The upside target could also represent a selling point should prices draw near, as a move lower could re-enter the fast zone and enter “price discovery lower” until there’s more substantial orders coming in where support can be found. Similarly, the downside target could represent a buying point as a move higher would re-enter the fast zone and enter “price discovery higher”.

One of the key reasons these levels measure as good locations is there’s a clear risk and reward in which the reward outweighs the risk.

As a market gets closer and closer to a support/resistance level, the risk you take lessens, assuming your stop is a particular point beyond support/resistance and your target remains the next point of control. As such, assuming the same probability of outcome, your edge actually increases.

This is the one type of pattern in which you could actually add to a losing position, provided you still maintain the same stop.

This is supported by mathematics as you will see on a kelly criterion calculator.

Say for example, you have $5 of upside, $25 of downside. We can simplify this to a risk of 1 and a reward of 5 or a payout of 5:1. If we assume 40% chance of a “win”, we can use a kelly criterion calculator to get an expected gain on an equal risk basis.

http://www.albionresearch.com/kelly/

-According to the Kelly criterion your optimal bet is about 28% of your capital

-Your fortune will grow, on average, by about 15.31% on each bet.

Now let’s say the 5:1 payout changes as the price advances closer to your stop. Offering $3 risk to $27 reward or 9 to 1. If the odds stay the same at 40%,

-According to the Kelly criterion your optimal bet is about 33.33% of your capital.

-Your fortune will grow, on average, by about 31.12% on each bet.

Note: the position size can increase by about 20%, and over an infinite time horizon maintains the measurement of portfolio risk measured in this case as “one full kelly bet”.

Now let’s say the stock moves to the stop. I would suggest using a full ATR as the “risk” if you are placing the trade that day, even if you intend on closing out the trade if it doesn’t close above the stop. Let’s say this is a risk of $1 to $29 reward or 29:1. We have to at least adjust the odds of a win slightly since the order will stop out for a small loss at least a bit more often. But for the sake of demonstration, in this case we’ll say it won’t.

You can now risk 37.93% of your capital or about 35% more than the initial 28% position to reach one full kelly of risk. Assumptions are made incorrectly all the time. The infinite time horizon is an absurd assumption, as is the assumption baked into the kelly criterion of 100% volatility tolerance, as is the assumption of a perfectly, fixed, known edge. But these assumptions can still allow us to draw conclusions.

We can use this as a means for comparison and to draw some objective lessons. If you are increasing your portfolio size substantially (double and triple) because your edge improves, you are probably either under doing the first bet, or over doing the 2nd and 3rd, or both. If you are pyramiding higher position sizes, that certainly may make sense if the probability of the asset going higher outweighs the shifting risk/reward and/or if you have an undefined upside target that expands based upon the strength of the breakout.

Also, if you are adding to a loser, you still need a clearly defined exit that shouldn’t change just because you’ve lowered your cost basis.

Additionally, adding lower and reducing higher is best for entire asset classes where you are either maintaining set allocations or changing the allocations based upon the relative edge against all other alternatives on a risk adjusted basis (including cash and adjusting for personal risk tolerance). If you increase a stock ETF lower, you have very little risk that the entire market will make a huge overnight move, whereas stocks are at greater risk of gapping substantially up or down as well as far greater risk of going to zero. Even though you are setting clear stops, either that overnight risk should be reflected in your position size, or you should only consider this tactic when buying asset classes or broad funds.

So how does edge influence position size? Slightly. The difference from 0% allocation to some allocation is infinite, so there are instances where a slight change in edge from negative, zero, or a very small edge to a more significant one can make a huge difference.. but if you are talking about a single setup developing, that difference is probably going to have less impact than most people would position for.

The more complex question of how one should change overall allocations as an outlook ranges from very bullish to very bearish to everything in between is a more difficult one to answer, particularly when we are talking about multiple trades held simultaneously and at overlapping intervals of time. There are other aspects of game theory that could help answer this question perhaps.

It would depend upon holding period which would depend upon fees and market’s tendency to trend and the edge and magnitude of that edge.  Nevertheless, with all things equal, a portfolio should allocate according to the probability of outperformance in the asset class, with additional capital tied towards “risk off” and cash related allocation as a person’s tolerance for risk decreases.

In other words, if it was simply between cash and stock, 50% stock would represent 50/50 chance of market up or down. If not for additional risks of margin 50% short stock and 50% cash would be acceptable as well. 80% chance would represent 80% chance of a rally, and 20% of decline of equal amount (adjusted for risk which actually would suggest holding more than 20% cash due to costs from volatility). the inflection point of 50/50 would actually shift towards shorts, but assuming shorting isn’t worth the additional risk, 80% cash and 20% stock would represent 20% chance of stock outperformance.

Your holding period makes a huge difference. For example, say there’s a 52% chance of the market going up on a given day… what’s the odds of the market finishing with more days up than down on a given year?

It turns out to be around a 72.5% chance that there are more up days than down. So a strategy that would rebalance yearly would theoretically have to be positioned much more aggressively bullish than one that rebalances daily if you assume a bull market or slight bias higher each day.

——————

One final note: If you are aiming for targeted allocations using options, you have to also understand that a small movement in price can change the dollar amount allocated towards bullish ideas or bearish ideas substantially. So this is where finding a new bet or doubling down may make some sense, but it’s to maintain targets, or slightly increase them with your edge, not to drastically change them.

If you own 10% call options and 10% put options a small move upwards in the market can easily cause the balance to change to say 25% call options and 4% put options. This is where adding to a broad market hedge and/or holding broad market calls can be helpful in allowing you to quickly restore the balance of your intended allocation without selling individual positions short of their target or before your system dictates that you should, without having inconsistencies and over exposure to directional bias, beyond what you intend.

If you want to have 10% calls and 5% puts, mostly short dated options and the market moves, you now might have 20% calls and 2% puts. If you use index options on the dominant direction, you can quickly take off an IWM call and reverse it with an IWM put or find a put of something relatively correlated to the market (higher than .5) and have something like 15% call and 7.5% puts. This isn’t perfect, but it helps you stay a lot closer to your targets.

There is a substantial cost over time in paying additional premium for assets that perhaps have a lower edge just to normalize some of your volatility, ut it can be measured against the benefit of reducing the risk and even if you can break even on this tactic, if it normalizes your exposure you can see similar returns with lower drawdowns.

 

Keeping somewhat inversely correlated assets or low correlated assets can provide a substantial benefit, even if neither system is better than the other, or even if one system is break even as can be illustrated below.

normalized systems2

normalized systemsHypothetically, even a slightly unprofitable system that can virtually eliminate or substantially mitigate all risk of drawdowns can improve your strategy, as you can simply leverage up and see higher returns with the same or lower drawdown. Long term capital management failed to stress test their portfolio against a long enough history that included major credit risks and bond market collapses and/or “black swan” type of events, and didn’t protect their portfolio against the unlimited loss, and they used assumptions about liquidity that weren’t true but if they were able to adjust for these things, they probably would have been able to accomplish a better risk/reward and eliminate the big blow up. It’s the margin risk and unlimited loss potential and lack of liquidity without adequate time provided to raise additional funds that ruined their strategy, not the soundness of the strategy itself or the math involved. They also tried to keep up with a historically very strong market with enormous leverage, rather than just accepting a very profitable system without trying to benchmark their gains to an underlying market.

 

Comments »