Greetings, Traders! In today’s post, an example of the Socratic method of teaching. Socrates was known for posing questions to his pupils in a way so as to allow them to arrive at their own conclusions. I have such a question today and it will pretty much tell you whether you will succeed as a trader… or NOT.

Here’s the question: Are YOU practicing the single most important key to your trading success… or are you leaving this one, most important factor… up to chance?

First I guess it would pay for you to know what I think that single most important key really is. That key might surprise you: it’s NOT an uncanny skill for timing the market, or a secret formula that allows you to always pick winning stocks. It’s not an infallible entry signal. It’s not some secret valuation formula or any kind of fundamental analysis.

In fact, the single most important key to trading success is *none* of those things that you are likely to find whole books written about.

Naw, the most important key to trading is within the grasp of anyone that cares to learn it and surprisingly… it has nothing to do with most of the things traders seem to concern themselves with. So what is this ‘holy’ GRAIL of trading? Read on 😎

In a very interesting experiment, legendary trading coach Ralph Vince took 40 Ph.D’s and gave them a WINNING game. Each participant was assigned $1,000 of virtual money and full freedom as to how much of it they could bet. Then the players sat in on a game that was fixed… *IN* their favor… 100 betting trials in which the payout was even money and each try had a 60% chance of winning.

Get the picture so far? Bet $1, win $1… and you are guaranteed to win 60 of the 100 trials. You just don’t know when the wins will happen.

Wanna be surprised? I sure was when I heard this one: only TWO of the FORTY Ph.D’s made money with this *winning* game.

Um, that means the rest lost. 38 out of 40 means 95% losers.

That’s scary. Now, I was relieved to hear that none of those Ph.D’s had a degree in applied mathematics, statistics or probability. But still..! We are talking very smart people, failing to win at a very easy game that is fixed in their favor. Yikes.

The reason behind the failure was this: *arbitrary bet sizing*. Remember, only two of forty Ph.D’s won that game. Of the two, I’d be willing to bet that one got lucky and the other used a system.

Bet sizing (or as it is called in trading, POSITION sizing) is the single most important factor to one’s success, and the single most overlooked… even ignored… skill. Now it seems a little more plausible that these bright folks couldn’t win a wagering game that’s fixed in their favor. They just didn’t know that the single most important factor to their success was knowing how much to bet.

**Great Expectations**

It’s easy to determine if a game or a trading system has what’s called a positive expectancy. Expectancy is how much you can expect to win… or lose… over time in a number of trials if you use a proper betting system. Remember you can still LOSE at a winning game if you don’t bet properly, so read on.

The formula for finding and beating the right kind of game is expressed below. Ready for it?

**E = ****[****W X P] – [L X (1-P)]**

Where:

*E* is the expectancy of a trading system;

W is the size of a win

P is the probability of a win, expressed as a fraction: i.e. 60% chance of winning is .6

L is the size of a loss

Plugging a $1 bet into this system reveals an expectancy of 20 cents. Your $1 win, which you will get 60% of the time, minus the $1 loss you will suffer the other 40% of the time, yields a net 20 cents for every dollar you bet.

That is, *IF* you don’t go broke first by betting too much!

With a positive expectancy game like I’ve described, how is it that 95% of the people playing it lost? Easy: I imagine that a great many of those Ph.D.’s began by betting $200 or $250 or more, being overconfident because they knew the game was fixed in their favor.

Once they experienced a string of three or four losses… not unreasonable with a 40% chance of a losing play each time… they found themselves with a very small amount of capital to work their way out of the hold they had dug.

This is how hard it is to make money with a game that’s stacked in your favor. What happens when you jump into the stock market, where everyone wants to take your head off? You guessed it… it’s war out there.

**Three Bears: This Bet is Too Much… This Bet is Too Little… THIS Bet is JUST RIGHT!**

How much SHOULD one bet? Actually, the ideal amount to wager is determined by the game itself. First, never place a bet on the side of the game that has negative expectancy. Whatever the potential payout is, likelihood of your winning is too low for you to risk your hard-earned capital. If possible, take the other side of the bet if you come up against a game with negative expectancy.

Once you have computed expectancy (look at your trading record, or a back-test of a system for an idea of this), you have a basis for sizing your bets ideally.

In the positive expectancy game I described above, with .20 cents won for each wagered dollar IF you don’t go bankrupt first, there is an ideal bet to place. The ideal bet size is not related to what you started with, but rather it’s a dynamically changing amount.

If in our .20 cent expectancy game above, we chose to play a fixed amount of $100 with each of 100 trials, we could expect to make $2,000. It’s highly unlikely… though POSSIBLE… that we would have 10 losses in a row and be out of the game form betting $100 in the first ten trials. After 100 goes, we have won $100 sixty times, and lost $100 forty times. That makes an account size of $3,000, or $2,000 in winnings!

**Fixed Betting is Safe… but Limited**

The above fixed bet of $100 is very likely to win a positive expectancy game. $2000 for $1000 seed money is a 200% gain, nothing to sniff at.

But contrast this with the IDEAL: The ideal amount to bet with a .20 expectancy is 20% of one’s bankroll with each play. This dynamically changes with the flow so that you take full advantage of strings of winners. Each win will increase the amount that you put into the next play. Likewise, a string of losing plays won’t hurt as badly as a fixed bet might. Each loss will decrease the amount of the subsequent bet.

By betting this way, we arrive at an anti-Martingale kind of effect. With 60 wins and 40 losses, adjusting the bet continually to equal 20% of one’s bankroll, the end result is an account value of $7,490… winnings of $6,490!!

In the above examples there was no talk about charts or historical trends. There were no news events, no entry and exit signals. We didn’t even talk about what Jim Cramer likes this week 😉 Rather, we illustrated that the single most important factor in trading success is determining beforehand *how much* to wager. Seeing as this is the single most important factor or success, it’s time to roll up our sleeves and figure out what that number is for us.

Okay gang! Startin to ramble a bit again so I’ll cut ‘er short. Sound off below, wouldja?

Happy Trading,

Kurt

fooled me! I would have guessed that risk management would have topped the list but this also makes sense.

Thanks for sharing.

Yer welcome, Eddy! In a sense, this IS risk management. Bet sizing is aimed at maximizing potential. Now, it is possible to take on too much risk, that is true. But it is ALSO possible to take on too LITTLE risk. For example…

Me: “Wow! I just got a 100% return on one of my trades!”

My Wife: “That’s great, dear. HOW MUCH did you have in that trade?”

Me: “Er, uh… thirteen dollars?”

My Wife: “You’re cute honey.”

If there isn’t ENOUGH being risked in a trade, that is also far from optimal. IN the blog’s example of the $1,000 pot… 60% even-money wins… and 100 tries… If I bet just a dollar 100 times I would end up with (60 wins X $1) – (40 losses X $1) = $20 in profits. So my account of $1000 would have grown to $1020. That’s playing it TOO conservatively.

Short answer: yeah, risk management DOES top the list. Bet sizing IS risk management. But ‘managing’ risk doesn’t just mean looking for the lowest possible risk but rather the most appropriate AMOUNT to risk, given one’s account size.

HT,

K

Happy Trading,

Kurt

Great stuff Kurt. How do you calculate the optimal position percentage?

For example, I’ve been studying bear/bull option spreads over at power options (thanks for pointing me that way btw) and there are always several spreads in the weekly options that have an expectancy of 5-10%. What would be the optimal size for those trades?

Thanks for asking, Lyle! The optimum percent to wager is supplied by so-called Kelly Criterion, which takes into account probability, payout, and account size. Look up Kelly Criterion on Wikipedia… kinda cool!

HT,

K