How to Determine Your Position Sizing

01/20/2010 12:01 am EST


Trading, like most things in life, is part science and part art form. You must be willing to accept that there are certain strategies that work, and those that do not. You must also learn the art of trading, develop the intuition required to read the market, and the experience on which to anchor your confidence.

The most crucial systemic or scientific aspect of my trading lies in the way I use position sizing. It is the fundamental basis for success in my trading. Without the discipline to stick to my system, I would be lost in the market. Every good trader has their own view on position sizing to fit their style of trading; the key is that they have one. If you do not have a system that you follow rigorously, I implore you to find one…now! Nothing—and I mean nothing—will improve your trading results more than having a position sizing and stop loss system, period.

I use a position-sizing algorithm based on the average true range (ATR). For the lay person, the average true range is just another word for volatility. It represents the average price swing from high to low that an asset experiences on an average day. I use a 20-period moving average to calculate the ATR, but some people use 14.

Why do I use the ATR? Well, for one, it gives us a good way to measure how much a position is likely to move in percentage terms on an average day. Trading with technical analysis is all about the pattern. Many times we can “guesstimate” the expected move an asset will make based on the pattern we see in the chart. We are not trying to capture pure price movement, but we are trying to capture a pattern. It makes no difference whether we expect the asset to move in our favor by 3% or 30%, what matters is the pattern. By using volatility-weighted positions, I am making sure that I’m not exposing myself to any one trade to a major extent above the others.

Here’s how it works.

Let’s say that my account value is $100,000, the ATR of the asset I’m trading is $3, and the asset trades at $100. Now, here is where you must decide how much risk you are willing to take on each trade, and how much of your capital you are willing to lose by being wrong. Let’s say that you are willing to lose 2% of your capital on each trade, equivalent to $2,000 in this case. You also must chose how many ATR units you are willing to let the asset go against you before you are wrong and stop out of the trade. I have read a good deal of literature on this subject, and have had good experience using a two-ATR stop. Using our risk parameters, we are willing to allow the asset to fall by 6%, or in this case, equal to 6% before we stop out, on a long side trade. We come to this by multiplying the 20-period ATR of the asset by our risk parameter of a two-ATR stop. We now have the absolute value of our stop, which is $6. We divide our 2% risk capital ($2,000) by $6 and come out with the number of units to buy, or shares in the case of an equity. In this example, it would be 333 shares.

Put more elegantly, position size in units = ((portfolio size x % of capital to risk)/(ATR x 2))     

NOTE: % of capital to risk is represented as a decimal (.02) in this case.

Now, in my case, I will add units of risk at times as the position moves in my direction. I will take another lot 1/2 ATR in my direction. That second lot gets the same risk management treatment as the first, but think of it as it’s own separate trade. Some systems, like mine, allow up to four units of risk in one trade. In this case, you would be putting 8% of your capital at risk. It takes a lot of guts to put that much on the line.

I don’t use 2% of portfolio value as my risk tolerance level. I am far more conservative because of the fact that I often add several units of risk to a trade. I use .5% of the portfolio, maxing out at 2% per trade with the max four units of risk. How much you are willing to risk should be dependent on two things: For one, your ability to handle big losses and stay disciplined with your strategy in the face of defeat; and two, how often you trade. Everyone finds their own happy medium, so find yours and stick to it!

By Leigh Drogen, trader and blogger at

Related Articles on STRATEGIES

Keyword Image
Yield Curve Hype
04/04/2019 2:32 pm EST

Despite what you heard, there has not been an official (10-year/two-year Treasury spread) yield curv...