iBankCoin
Home / statistics

statistics

Nasdaq 100 TICK Stats

Ahoy mates, I saved you the rigmarole of running standard deviation studies on the NASDAQ 100 TICK. Using high quality IQ Feed data, I ran a study on 1-minute bars to determine 1st-3rd standard deviation for the indicator.

I set audio alerts for the 2nd and 3rd thresholds using Multicharts. We’ll see if those last or end up in the junkyard with the other algo scraps.

Here’s how the data looks on my side:

Positive TICK 1st, 2nd, 3rd sigma: 36, 54, 70

Negative TICK 1st, 2nd, 3rd sigma: -37, -52, -68

NQTICK_02222015

Why does any of this matter? Have you ever looked at something and thought, “hmm, that’s different”? Would you want to know if something you’re seeing is in fact abnormal? I do. When I am in a trade I accept that my limbic system will be active and a big part of my job is being aware of the waves of emotion that sometimes get in the way of trading. One of my boulders to channeling that creative energy into objective trading is basing my observations and decisions on cold-dead numbers, logic.

Are they Holy Grail trading signals? No, but they do offer a peek inside the engine of our good friend /NQ_F.

Comments »

Sunday Morning Stat Mash

Today we are looking at the range statistics of the Nasdaq regular trading session.  Although Nasdaq futures trade around the clock on globex, our focus is primarily to trade the index when the equities underlying its value are also active.  Thus, this range analysis is conducted using the trading hours from 9:30-16:15 eastern time.  This data was pulled from the IQ feed servers and compiled using about five years of trade (3/16/2009 start date).

The data has been split into two categories, up days and down days.  This information is helpful when preparing your hypothesis on the day because you have a sense of what price levels are relevant and actionable, and which ones are beyond the average reach of the marketplace.

Some notable stats include:

  • Average down day range: 35 points
  • Average up day range: 29 points
  • 2014 Average down day range: 45.5 points
  • 2014 Average up day range: 32.5 points

Usually I will use a histogram to calculate a “normal” range, considering about 68% of occurrences as normal and anything else an outlier.  What appears to more relevant in this instance is to look at the data over time to see how ranges are trending.  I applied a 10-day moving average to the daily up and down ranges.  The average daily up range is at about a 5-year low at just under 20 points.  The average daily down range is just less than 40 points.  Thus recent trading has provided about double the daily range when trading lower than higher.  See below:

09062014_UP_rangestats 09062014_DOWN_rangestats

Comments »

Labor Day Morning Stat Session

During the week long market profile webinar we talked about ‘the story’ and how when we choose to focus on the market activity in the context of an auction we react in planned ways.  As the market moves we can constantly return to reading context by asking ourselves a few questions:

  • What has the market done?
  • What is it trying to do?
  • How good of a job is it doing?

These questions help us answer the final question which determines how we react, if at all—what is the market likely to do from here?

You truly need to see this in action, like seeing a big rotation and running through the questions real time, to see the effect it has on your mental vision.  You return to these questions, you use market profile as a tool for seeing the auction, and the process provides a logical decision making process.

Probabilities can be just as logical a basis for decision making.  They are statistics derived from past market behavior and it is reasonable to include them as part of a decision process.  For example, if the overnight low breaks 89.52 % of the time and you have entered a short position which is working in your favor and is within a few points of the overnight low, then pressing for at least a 1-tick break of the overnight low makes sense, especially if the session has matured a bit, increasing the probability of a break.  These little inches we fight for add up to miles when it comes time to calculate expectancy.

Relying on these foundations (logos, as the Greeks called it) for trading will yield better results and a more objective eye.  Imagine your statistic does not fulfill because that is the simple nature of the markets.  This resistance to the laws of large numbers will speak to the context too.

Enough emphasis on why statistics matter, yes?  Without further adieu, I have performed a study on five years of trade in the Nasdaq futures.  The raw data has been pulled from the IQ Feed servers via their symbol @NQ# which is the continuous contract.  Some key points:

  • Overnight high/low break occurred 89.52% of the time, with 73.28% of breaks occurring before noon
  • The normal volume on an overnight session is between 17-39k contracts
  • The normal range of an overnight session is between 16-46 points

09012014_ON_breakSTATS

09012014_ON_range_histogram

09012014_ON_volume_histogram

Comments »

Expectancy Shows The Obstacles

The numbers are the truth and the numbers can be your friend if you are on the proper side of the expectancy equation.  After reviewing all trades taken over the last two months, my best strategy per the numbers has been buying out of the money call options.  The ‘when’ question has been a prickly one, because my worst strategy has been the always sexy YOLO.  Coming in second place has been good old stock trading, which is toeing the zero line of expectancy.

My current win rate on normal options trades is 34.5% and 18.9% on YOLOs.  My stock trading win rate is at a frumpy 50% aka coin flip.  Yes, think about that next time you blindly follow me into a trade.

My average win on a stock trade is just a bit greater than the average loss.  This is a very thin edge at the moment, almost not worth the intense time commitment and effort.  My average win with options is about 3.5x the size of the average loss.  Notable big winners contributing to the large average win size were X, C, CELG, TWTR, and MBLY.  My average win trading YOLOs is only 2x my average loss.  This could be because a solid win has not occurred to skew the data set.

There are a few areas for improvement.  First is becoming more selective with stock trades.  These type of positions need considerable amount of thought and planning and might be better as longer-term trades.  Next is throttling back to YOLOs.  The best strategy for me appears to be sticking with a YOLO for about 3 weeks, willingly purchasing another week of time as best is possible, and cutting bait if the trade never pans out.  Two YOLO positions at a maximum with no more than 1.5% of risk committed.

Overall I have been using about 10% of my risk capital in normal option plays.  After reviewing the statistics it seems reasonable to increase this overall exposure to 15 percent, slowly of course, with a penchant for only the best chart setups.  Remember, this is the tendency with expectancy, to increase activity where expectations are highest.  This in turn can lead to lower quality entries, very much something to watch out for.

The rest of my risk capital must either sit in cash or longer term investments.  Current investments are TWTR, LO, GPRO, TLSA, and XON.

Thus my risk book looks something like this at the present:

Cash – 35%

Investments – 42%

Stock Trades – 14%

Option Trades – 9%

As you can see, I am in the uncomfortable position of having a heavy cash position into the start of a new month.  I will be seeking to correct this as trade ensues next week.

Comments »

Sunday Morning Stat Crunch

During the week long Market Profile webinar I hosted for the After Hours with Option Addict crew, I defined and discussed initial balance (IB) quite extensively.  This simple concept of noting the price range printed during the first hour of trade can be a huge help in determining market conditions early in the day.  The same can be said for the opening swing, only the opening swing interprets market activity even sooner, sometimes being established in less than 5 minutes.

One of the primary reasons for tracking the initial balance is its use in determining what sort of “day type” is occurring.  Another primary reason, the reason which is more actionable intraday, is that we rarely see a full day of trade without breaking either IB price extreme.  Therefore, if we manage to enter a trade inside of initial balance which is working in our favor, we can press that day trade and ride that winner a bit further.  Traders always emphasize the importance of letting your winners run because those few extra ticks you gain from a well managed trade can make a huge difference to your overall expectancy.

Nothing builds confidence in an idea like statistics and probabilities.  Therefore as an addendum to the weekly course, I have built out the relevant IB statistics for my product, the Nasdaq E-mini future contract.  I used five years of pure IQ Feed data to compile the following stats.  Some highlights:

  • We break initial balance 94.75% of the time
    • By 11:30 – 73.03% of the time
    • By 12:00 – 81.13% of the time
  • Normal IB range (69.87% frequency) is 11 – 24 points
  • Normal IB volume (66% frequency) is 40k – 75k contracts

And without further adieu, I present the data (looks familiar, yes?) in its entirety below.  Enjoy:

08242014_IB_breakSTATS

08242014_IB_range_histogram

 

08242014_IB_volume_histogram

Comments »

Expectancy

Expectancy is one of these trading concepts that rarely receives a thorough explanation on the web because it is not very sexy.  However trade expectancy and position sizing are two of the most important pieces to successful trading.

I am not preaching this lesson from the pulpit, high atop my cloud of mastery and divinity, but instead from the ground, bare foot, with the purposeful intent of advancing our trading abilities.  That is my disclosure before you view my very humbling July trading expectancy.  See below:

July_Expectancy

As you can see, I operated a net losing operation trading futures in the month of July, and expectancy took a radical turn for the worse last week. I will not begin to interpret this data, but instead will use this very unsuccessful data in our lesson today on expectancy.  Let’s look at how expectancy is calculated:

EXPECTANCY = (Probability of Win * Average Win) – (Probability of Loss * Average Loss)

The very simple goal is to trade in a way that produces the largest expectancy number possible.  It only takes some basic algebra to understand how to affect the outcome of this formula—you need to make the product of the first set of parentheses larger than the second product.  We have three independent variables:

Win rate

Average Win

Average Loss

Therefore we have three areas available for improvement.  We can increase our win rate, increase our average win, or decrease our average loss.  (Simple, but not easy)  If you start trading too tight in an attempt to decrease the average loss, then your win rate will tank and you will miss out on otherwise good trades.  If you reach too far for bigger wins, then you are more susceptible to general market conditions, where price can stop working in your favor at any time—this is the nature of markets and why we spend much of our time studying the pillars of continuation, price and volume.  Finally you can add filters and qualifiers for taking a trade with the effort of only taking the most “sure” trades.

Once we have an actual expectancy, we can calculate how we are likely to perform.  Let’s run my numbers last month to demonstrate the concept:

Average July Expectancy = -1.16 ticks

Average Trades per day = 9.6

Trading Days in August = 21

Expected ticks earned (lost) = -233.86

As you can see, this is a tick losing enterprise, and without any adjustments this approach will lose about 11 ticks a day on average.

You can perform this basic math with any trading approach to see how your strategy fares.  And you should.  As a benchmark, the elite futures traders on my timeframe, the best of the best, have about a 1.5 tick expectancy.

If you are calculating expectancy on a stock or option trading strategy, it will usually make more sense to calculate your expectancy on a dollar basis as opposed to ticks because different stocks carry different daily ranges and your position sizing will adjust according to the dollar value of the instrument.  Once you determine the expectancy, you simply roll it out on the # of trades you take per month.  As enticing as it may be to take more trades using a positive expectancy method, BE SURE TO KEEP GOOD STATS, because you will likely see yourself deteriorating that expectation unless you keep your standards up.

Stay classy out there, and keep your expectancy on the uptick in August.

Comments »

Trade Statistics Session

Below you will find a brief summary of my trading statistics since we started the month.  Key data that jumps out of me are as follows:

  • Plan compliant trading carries an 86.2% win rate
  • Non compliant trades lost a net total of 183 ticks or $915.00 trading a 1 lot
  • Winning trades last twice as long as losing trades (let your winners run)
  • Almost half of my trades are whimsical, impulse, unplanned trades. WTF?
  • Eliminating half of my impulse trades only improves the win rate to 62.4% aka not good enough
  • There are 4-6 quality trades available per trading day

I failed to complete stage 1 of my trading plan during this set of trades.  The goal is simple, a 70% win rate over a set of 50 trades.  To do so requires supreme patience and complete avoidance of unplanned trades via impulse management.  I have just two trading pictures.  Technically 4, two shorts and two longs symmetrical in nature.  If you have been paying attention to my full disclosure trade entries then you know my pictures too.  The challenge is to consistently seek only my two quality pictures and avoid the other trades.

Finally, this is 1 lot trading and the next of all my trades is an egregious -1 tick loss.  Commissions equal about $280.00 bucks.  This is a cheap but necessary education if you ask me, a core building block before advancing to more size and employing trade management tactics to achieve consistent profitability.  The key is to actually learn because what I am currently wasting is the greatest resource of all—time.

STATS:

071002014_stats

VIDEO:

Comments »