Option FanaticOptions, stock, futures, and system trading, backtesting, money management, and much more!

Statistical Manipulation

Peter Berezin, chief strategist for BCA Research, wrote an article for the September 2016 issue of AAII Journal. Near the end he made some comments about statistics:

     > Statisticians like to say that if you torture
     > the data long enough, it will confess to
     > anything. This old adage is especially relevant
     > to the study of stock market anomalies. First,
     > there is the risk that any anomaly that is
     > unearthed will simply end up being the product
     > of data mining. Second, even if an anomaly
     > turns out to be genuine, there is a risk that
     > it will be arbitraged away once the investment
     > community becomes aware of it.

I argued for an increased use of inferential statistics here and I later relayed the opinion of a financial adviser as to why inferential statistics may be relatively uncommon.

I still believe inferential statistics are useful to offer an apples-to-apples comparison but Berezin makes a good point that statistics may be used to manipulate. We can never be sure of an investigator’s underlying motives and unless we do the research ourselves, we also cannot be sure the statistics were correctly computed.

I do believe we can do a couple things to avoid these statistical issues. Data mining involves searching a large collection of data with the purpose of finding significant results. This should be avoided. Give me an indicator and enough data and I can find a snippet of price action for which the indicator works fabulously (fallacy of the well-chosen example). This is unlikely to be profitable in live trading, however. One way to avoid this involves searching the surrounding parameter space for a high-plateau rather than a spike region of profit.

With regard to market edge being arbitraged out over time, I need to monitor my system and have criteria indicating when it might be broken. Walk-forward analysis can help to keep a strategy current thereby increasing the probability it will work with live trading. I may also monitor total profit/loss and stop trading the system when this value falls below the equity moving average. This should be developed through proper validation methodology.

Market Measures Mistake (Part 2)

Last time I discussed my first criticism of the Tasty Trade show “Market Measures” (MM): reporting of incomplete information. My second critique is an occasional mistake in tabulating the results. A good example of this is the MM episode from November 8, 2016.

MM claims taking assignment on the short put and selling a call against it improves the overall trade. Slide 5 says this increases both the overall winning percentage (from 92% to 96%) and the average P/L (from +$46 to +$53).

Pay close attention to slides 3 and 4. Slide 3 says 92% of trades were winners. The average P/L was $46 and the average length of trade was 17 days. Slide 4 says losing trades lost an average of $769 in 42 days or $682 in 64 days if rolled (assignment followed by writing of a call). Consider case 1 where a losing trade is followed by a winning trade (likely due to the 92% win rate) vs. case 2 where the losing trade is rolled.

In case 1, we have average P/L’s of -$769 (average losing trade) and +$117 (average winning trade), which sum to -$652. The average trade length is 56.8 days (based on information from slides 3 and 4).

In case 2 we have an average loss of $682 in 64 days.

So despite the fact that rolling (case 2) turned 45% of the losing trades into winners (slide 5), simply realizing the loss and putting on a subsequent trade (case 1) did better (in 10% fewer days). While they don’t give enough information to calculate the average winning and losing trades, even if the winners only averaged +$1, the losers averaged -$1,241. This is over 10 times the average winning trade: a difficult number to stomach.

Yes, the improvement in average P/L on slide 5 is good but the implication of slides 3 and 4 is equally bad. I therefore see no justification for the all-positive takeaways shown in slide 7. I also walk away wondering if the numbers add up at all and I can’t know for sure because they provided incomplete information. I have sent a follow-up e-mail asking for another response two months after my initial e-mail received none.

In summary, I enjoy the MM show and I support the Tasty Trade efforts to boost financial literacy. The sloppiness does not set a disciplined example, though. Presenting incomplete information may be okay for people trading as a hobby. It cannot be taken seriously until sufficiently rigorous for a trading business, though, which is probably just what the disclaimers aim to suggest.

Market Measures Mistake (Part 1)

I am a regular viewer of Market Measures (MM) on the Tasty Trade network and I am a big fan of what Sosnoff, Battista, et. al are trying to do on the show. Unfortunately, I sometimes think they do a poor job of doing it.

Although categorized “optionScam.com,” I do not consider Tasty Trade fraudulent despite some conflicts of interest. I do believe their approach is sometimes flawed. I sometimes wonder whether presenters are intending to deceive or simply victims of the “don’t know what they don’t know” syndrome. I prefer to give the benefit of the doubt wherever possible.

I believe the audience has an opportunity to apply critical thinking in order to see through the mistakes. Doing this can prevent unexpected losses of capital. In this blog, I label this due diligence optionScam.com.

My criticism of MM is two-fold with the first being a failure to report key information.

They begin with an outline of the study, which I sometimes find to be incomplete. In other words, the description is not enough to allow me to replicate their study. I believe this is essential in any scientific research because a failure to do so is carte blanche to report numbers regardless of accuracy.

MM rarely mentions transaction fees, which I consider a major fault. I have demonstrated transaction fees to be capable of making or breaking a backtrade. A few years ago Sosnoff talked about “resort fees” as something they included. On December 12 of last year Sosnoff said “and this takes into account all the resort fees, all the commissions, and everything else.” Exactly what did they use for resort fees and commissions? And why associate them with a vacation? Get serious and specific because transaction fees need to be mentioned whenever trade statistics are presented.

Most episodes leave me wondering about key metrics. They present things like average credit, win rate, and average PnL. Sometimes they present largest loss, which I consider mandatory reporting for every single study (drawdown is risk). Rarely do they report number of occurrences: is it a large enough sample to be valid? Seldom do they report buying power reduction: are the results large enough in terms of return on capital to be meaningful?

I will continue next time.

Performance Update

I am very past due for a performance update.

I will focus discussion on the following table:


My first full-time year was 2008, which means I now have nine years of trading history. Through that time I have tried a few different things, backtested a lot, and learned tons. Hopefully I have learned most from my mistakes. Only the future can reveal whether that is true.

The table includes three sets of data. I start with my yearly performance and the compounded total return. I then repeated these calculations for the small-cap and large-cap indices. Green (red) numbers indicate where I outperformed (underperformed) the benchmark. Standard deviation is a measure of risk (as discussed here and here) along with max drawdown (MDD) (as discussed here). Risk-adjusted return is total return divided by standard deviation. CAR is compound annualized return, which makes CAR/MDD another risk-adjusted metric.

I have outperformed the benchmark in five out of nine years.

I have generated profit in seven out of nine years.

My average return significantly outpaces the indices. Mostly for that reason, the risk-adjusted returns are much better too.

My biggest disappointment is the relatively high standard deviation. To this end, my 2012 return of +52.39% hurts. I can’t say exactly what was going on with my trading that year without looking back and scrutinizing the records. Yes it’s a great number but my preference would be to have stable returns like I have the last few years.

I very much like the fact that my worst year was limited to just over a 10% loss. This is the kind of stability somewhat lacking to the upside. I experienced three catastrophic losses over the last nine years and the overall performance suggests I have bounced back quite well.

Graphically, the comparison looks like this:


The outperformance is clear.

Although I was in negative territory and underperforming after four years, my preference is to have a relatively flat equity curve in volatile markets as opposed to a curve more jagged than the coastline of Buzzards Bay, MA. This is something I have managed to accomplish thus far.

On the Need for Improved Financial Literacy

The need for improved financial literacy nationwide is conventional wisdom: a simple internet search will bear that out. I challenged this in my last post because people seem to have little interest.

Although I criticized this challenge based on limited sample size, it may have some merit. People are generally uneducated about investing and they seem quite willing to let professionals do the job for them. In terms of value, financial literacy differs from functional literacy. Many people who cannot read or write have felt the squeeze over their lifetime from those around them including prospective employers. Hiring financial advisers is much more socially acceptable than functional illiterates asking others to read/write for them.

This pertains to a blog post I wrote in January where I decided it wasn’t the result of a brainwashing perpetrated on the American public by the financial industry. Rather, the decision to hire investment advisers is a delegation of duty. The cost of this delegation includes management fees and lower investment returns.

We could ask whether the real issue regards a need for improved financial literacy or a choice about how people wish to invest. I don’t think the average person has enough education to decide on the latter so perhaps it does come back to financial literacy. I could also argue that most financial professionals don’t know as evidenced by the fact that so many of them do not employ options.

In the last post I pointed out that trader education is a subset of financial literacy. One can know a lot about finance, understand the role of investment advisers, and know how to interview/select a knowledgeable adviser. Even someone educated in finance may elect not to take that next step and manage his/her own investments.

I think the basics of financial literacy aim to keep people out of a “paycheck-to-paycheck” struggle. This involves how debt works, proper budgeting, savings/interest, etc. Investment management pertains to savings above and beyond that needed for annual living expenses. Getting a large proportion of the working class to establish and maintain a rainy day fund would represent a significant move higher in terms of financial standing. Having surplus capital available for trading and investing, though, is still a whole other level.

For those in possession of surplus investment capital, financial literacy may be channeled into a business. This is what I have done in order to retire from Corporate America. The pharmacist in me would point out a similarity to the way some have turned “medical literacy” about dietary supplements into a business. I have currently have no customers in my trading business, however, while many claims regarding dietary supplements are baseless and invalid.

Giving Back (Part 3)

Before continuing forward, I want to clean up a couple things—the first being the need versus achievement debate. I’m not really in a position to assess achievement. You could also make the case that I’m not really in a position to assess need. I therefore will not be deciding whom to teach based on those criteria.

I have discussed two monetary factors with regard to a trader education program. I have the previously stated reasons for charging a per-session fee. I would also recommend having ample savings to eventually open a real account. Both of these are included to try and prevent students from dropping out, which would result in time wasted for me [preparation of presentation material] and for them [education never applied].

I have thought about giving back by taking an entirely different avenue: teaching high school students. Kids are generally means-challenged so I would not charge a per-session fee. Neither are they likely capable of opening live trading accounts. They do have a solid potential for future income, however, which is wealth they could later manage on their own. Because option pricing models and considerations of probability and statistics all fall under the “advanced/theoretical mathematics” category, I would target advanced math students. This would also get me academically-disciplined students to work with who would be more likely to complete the program.

On a totally different note, I was tempted to argue against the “improve financial literacy” battle cry because people simply did not seem interested based on my recent exercise of getting 11 responses from 54 messages sent. Cost could have been a confounding variable; people may have been unwilling to pay a stranger. I also have no way of knowing how many of my 54 messages were actually received. Over the years, I have gotten poor response rates over the Meetup.com website. For all I know, only 11 people were even aware they received a message.

I do have other reasons for thinking people may not have much interest in learning to trade. Over the years I have approached a few different libraries about conducting a trader/investor program. This was met with lukewarm response because they had not found investing programs to be well-attended in the past. This was the same reason the Ann Arbor District Library gave for discontinuing their subscription to Value Line a few years ago.

Learning to trade, though, is only one subset of financial literacy. I cannot conclude from this that people have little interest in the latter.

Giving Back (Part 2)

The current topic under debate is whether I should give back by trying to teach those without the resources to pay or those who have demonstrated achievement.

I think a significant discount qualifies as “giving back” even if it is not free. I thought about charging per meeting as motivation to stick with the program (it’s harder to abort once we have begun to commit). $20 per monthly meeting would be $240 for a year, which is far less than programs costing thousands of dollars. I would also encourage people in the group to study and practice (paper trade) together. Anytime they have questions I would be happy to answer. This would be a dynamite training package for a steal of a deal.

I am quite convinced that no matter how small, participants must have some skin in the game. I can’t force them to trade and I don’t want to do anything that might put me in an “investment advisor” role because I am not a registered investment advisor. A per-meeting fee helps them—by providing motivation to get through—and it helps me by lowering the probability of dropout. I would be extremely disappointed if I were to compose presentations only to later be deserted by my audience.

People who cannot afford a nominal fee face an additional problem. One must have savings in order to trade. I would probably suggest opening at least a $10,000 account to learn. I would expect interest to wane for someone unable to open a real account. Discouragement would build when trader education could not be converted into actual profits.

This is strike two against giving back to those without means. First, no fee means no front-loaded motivation to get through the course. Second, no savings means no application for the education itself.

I recently messaged people from Meetup asking if they would be interested in a trader education group. I suggested monthly meetings with a charge of up to $20 per meeting to cover expenses and to establish some accountability. From 54 messages sent, two said they would be interested, one person was a definite maybe, and eight declined.

Of the eight who declined, two said they wouldn’t pay $20 per meeting. This could also be a reason more people did not respond. While it may be healthy skepticism toward a stranger, I doubt anybody could find a complete curriculum delivered by a full-time trader for less. And what they don’t know they don’t know is that offering this for free would be doing them a greater disservice.

Giving Back (Part 1)

Over the years I have found little interest among people in learning to trade options. I am thankful for what success I have had with my trading and for the freedom my entrepreneurial trading business affords me. I have given significant thought to how I might be able to “give back” as an expression of my gratitude.

I want to begin by addressing a potential contradiction between giving back, which implies free, and teaching people to undertake a for-profit enterprise. As an inextricable component of the tremendous business opportunity I have cultivated for over nine years, I believe trader education is extremely valuable. Maybe this is not something to be given for free lest its value be undercut. With the domain being strictly financial, an up-front investment seems more fitting with the goal of consistent profit over time. No investment is free and therein lies the contradiction.

I perceive a significant difference in subject matter between offering people what is essentially business training (trader education) and volunteering to tutor after school kids. I also believe the story would be different if the trader education were used to generate profits for charity.

Perhaps academic scholarship can help determine whether trader education should be free or for sale. Scholarships are generally awarded based on need or merit (i.e. achievement). The latter often includes people who would not otherwise qualify based on need. If I want to give back, should I teach financially-challenged or high-achieving individuals?

I believe consumers of expensive trader education services are more likely to stick with the program than those without obligation. I have previously discussed how traders can be a very fickle lot (myself occasionally included). Over the years I have seen people come and go through various Meetups and trading groups. This stands in stark contrast to many students I have seen motivated to complete programs for which they paid a hefty bill up front.

With many trader education programs costing thousands of dollars, people who attend are generally not those in need.

To whom should I “give back?” Score one point for merit-based because those with nothing to pay are lacking a key motivator to get through.

I will continue next time.

The Risk of Going Naked (Part 2)

I recently presented some excerpts from an online forum regarding the risk of trading naked options.

This is the kind of sobering talk that makes me uncomfortable with leverage. Regardless of the extent to which market turmoil occurred in the backtesting period, a severe enough market crash could always bring to fruition something worse.

Without leverage, drawdowns tend to be minimal, risk of Ruin is relatively small, and quality of sleep is restful throughout. While I am tempted to take advantage by boosting position size for better total return while maintaining equal or lesser drawdowns, doing so means adding leverage right back into the equation.

If I want to hedge myself against the most extreme moves then I can buy [cheap] insurance. This will increase return on capital. This will also save me if the catastrophic market crash actually takes place, which has anecdotally happened every 5-7 years throughout the current century. The market will more commonly fall several percentage points and level off or reverse higher. This magnitude of correction is nowhere near that required to realize max loss on the hedged naked puts (vertical spreads). The breakeven point would require an even larger drop since the insurance is not free.

And while that max loss is much lower than the “undefined risk” of naked puts, the loss is probably catastrophic either way. I’m tempted to backtest this and look at different position sizing but the sample size would be too small to allow for any meaningful conclusions.

I think it would be nice to insure the extremes and be able to claim “were the market ever to crash to zero, you would be covered [or even profit].” A total market crash is what doomsday forecasters love to prognosticate. Hedging against this is always possible with options but it comes at the cost of lower profitability during normal market action. My preferred solution is to allocate no more than X% of one’s total portfolio to this sort of “unlimited risk” trading.

Allowing for the possibility of catastrophic loss and managing risk by position sizing is, in my mind, what has relegated a strategy like this to accredited investors and hedge funds.

Then again, this argument could be easily contested. Common stocks—purchasable with leverage to investors who are not accredited—also go to zero (e.g. bankruptcy). While baskets of stocks rarely go to zero and indices have never gone to zero, all of the above have suffered catastrophic losses: pick any stock market crash. Limiting total portfolio allocation, therefore, is probably smart whether dealing with naked puts or long stock.

In the eyes of the regulators, naked puts would probably be okay for anyone as long as position sizing is based on Reg T margin requirements (i.e. cash-secured).

Fooled by Randomness

William Bernstein, co-founder of Efficient Frontier Advisors and author of several books, was interviewed in the September 2016 issue of the AAII Journal. He made an interesting point that reminded me of Nassim Taleb’s book Fooled by Randomness:

     > …never confuse outcome with process. By that
     > I mean that there is a lot of noise in investing,
     > and it is perfectly possible to have a bad
     > strategy or be a stupid investor… and do very
     > well. People who buy lottery tickets occasionally
     > win. Having a good strategy and not having a good
     > result happens all too frequently as well. The
     > essential thing people really don’t understand
     > about finance is that there is an enormous
     > amount of noise, particularly over the short term.

The take-home message for me is the first line: never confuse outcome with process. The stock market goes up most of the time. It may even go up for years. When it corrects, large sums of money can be lost in short periods of time. One must be careful not to lose all that has been gained or more.

While the market is going up, people are happy with their strategies and impressed with themselves. Many think they have skills and have attained a level of expertise. Remember the old saying “a rising tide lifts all boats,” though, which suggests most stocks rally when the market moves higher. Anyone participating during these bull market periods is likely to make money regardless of strategy. Ego can really come into play when people dramatically increase position size. This inconveniently happens all too often just in time for the next market correction or crash.

Trading system development methodology is designed to determine whether a profitable system is attributable to an effective strategy or simply a matter of luck. The process is designed to screen out the luck factor by studying large sample sizes, maximum drawdowns, employing Monte Carlo simulation, and walk-forward analysis. This is challenging work because it requires coding skills and software know-how in addition to theoretical understanding.

The cost of mistaking random luck for skill worthy of confidence is an increased likelihood of catastrophic loss when the next bear market rears its ugly head: something none of us look forward to.