Be Thankful for “Fortune’s Formula”

Boris Schlossberg

There is perhaps no greater skill in trading than losing. To be successful you need to be comfortable with the idea of either losing small amounts frequently or large amounts rarely. In either case, the psychological impact is less than pleasant and almost every single trading error ever made stems from not being able to lose well.

Losing well is the fine art of not giving a f-. This is exceedingly difficult to do. Markets have novel and unexpected ways of torturing you, no matter how well prepared you think you are.

One very good trick to try is to walk back through any backtest you have and look at losing trades only. Quite often you will find that they can be clustered -- a sure sign of regime change in the market. However, the point of that exercise is not to tweak or curve fit the trades, but rather to imagine your reaction in the midst of the losing streak.

Athletes, gurus and all self-improvement books always teach you to visualize your winners. That may work in sport or real business, but in the Alice-in-Wonderland world of the markets, it’s much better to visualize your losers because that is where the real strength lies. If you can walk through the losing periods in your head, it makes it a lot easier to walk through them in a real trading environment.

Of course, The only ultimate defense against losses is small size -- mainly because small size make you honor your stops and therefore control risk. I’ve mentioned in the past the great benefits of trading any system on 0.01 lot for a good period of time, just so that you can become comfortable with its quirks.

Impatience is the enemy of all traders -- but there are two types of impatience that I generally see. The first is getting into trades at unfavorable prices as FOMO (fear of missing out) kicks in. That’s very common but can be overcome with strict entry and exit rules.

The second is far more insidious -- which is the impatience to make big money. That always causes traders to size up right away, which puts double pressure on them when the trade goes wrong -- the pressure of size and the pressure of losing. The pressure of loss is bad enough so it’s little wonder that the standard cycle of trading is -- lever up, get stuck in a losing trade, get margined out or close out in dejection losing more than 50% of equity and walk away with head bowed.

Trading tiny is a great way to start, but even when you are ready to trade bigger size, there is a great tool that all traders should use called the Kelly Criterion. Dubbed, “the fortune’s formula” in popular trading literature, it actually has some serious weaknesses that I will discuss in next week’s column, but its central thesis is pure gold -- and should be seared into every trader’s mind.

The principle idea behind Kelly Criterion is that a trader should never bet a fixed amount of capital per trade, but rather some percent of equity. This way when your equity declines you will bet less and when your equity increases you will bet more.

I’ll go through the Kelly Criterion in more detail next week, but for now, this is my holiday gift to you -- adopt a proportional trade size to whatever setup you are using, and your risk of ruin will decline markedly.

Boris’s Formula For Trading Success

Boris Schlossberg

Some of you may be familiar with the Kelly criterion which is a formula used to calculate an optimal bet size allowing gamblers and traders to maximize their winnings. The proof for the concept can be complicated, especially if you are mathematically challenged like me, but the net takeaway of the Kelly formula is that on 1 to 1 payout you bet 2 times the probability percentage minus 100% of your capital. So that if probability is 60% you bet (2X60% -100%) =20% of your capital on any given trade to maximize your profits.

In real world of trading betting 20% of your capital on any one idea is insane and Kelly has been criticised for creating wildly inappropriate bet sizes for traders enamoured with the math but utterly unfamiliar with the non standard distribution properties of capital markets. (In plain English -- there is no such thing as fixed odds when it comes to trading.The game could literally morph from roulette with a tiny negative 49%-51% edge to the 1 out 1000 odds of a daily scratch lottery -- and this is the key point -- all on the exact same trading strategy.)

So given the fact that odds in trading are essentially an illusion, I have my own rather crude but I think more effective formula designed not to optimize gains but rather to contain losses to a manageable level.

Anyone who has ever traded for more than a month quickly realises that to survive in the market the key is to control losses, as gains pretty much take care of themselves. The simple math is that it takes 200% profit to get back to even from 50% loss, so the key to long term winning is to never, ever get to that -50% level in the first place.

So here is my day traders formula for controlling size.
Ask yourself the following questions:
What is the maximum amount I am willing to lose in a day?
What is my stop on my day trading strategy?
How many trades will I make per day?
What is my expected losing percentage?
How much money will I trade with?

On a hypothetical example let’s say I never want to lose more that 1.5% in a day (I am assuming a worst case scenario of 10 straight days of losses would equal to -15% of my account. Make your own assumptions on this- there is no wrong answers -- as long as you assume at least 5 losing days in a row minimum)
My stop is 25 pips
I will make 10 trades/day
My expected loss is only 20% ( 2 losers out of 10)
I have 10000 dollars to trade with.

In stress testing these variables I assume that my expected losses will be three times what they should be ( The good old rule of thumb that in any business plan you must double your expected costs and half your expected profits comes in very handy in trading. In other words ALWAYS assume that things will be at least twice as bad as you initially think. In my case I assume that they will be three times as bad)

So when you put all of these ideas together I basically trade at 10,000 units (one mini-lot) per 10,000 dollars of capital. This way if I lose six times out of ten I lose 150 dollars or about 1.5% of equity. (Yes I know I still have 4 more trades left, that could reduce my loses, but as I said I am assuming the worst, ALWAYS).

Now for those of you with a mathematical mind or an engineering degree this “formula” is laughably imprecise and full of utterly subjective assumptions. But that is exactly the point. This a formula is produced by the school of hard knocks rather than crafted through the elegance of Big Data algorithms. Its designed to be as robust as possible against a non deterministic distribution ( simple English -- I have no f-ing idea what will happen) rather than a beautifully written model that may ultimately bankrupt you.

Get a Ridiculous Discount on Our Trade Signal Service

In any case I hope you find it useful.