There’s all kinds of discussion and advice about how much risk should be taken on a trade-by-trade basis. Figures like 1%-2% are often tossed out in the active trader community.Ă‚Â William OĂ˘â‚¬â„˘Neil suggested 8%Ă‚Â in his book How to Make Money in Stocks. If you really want to do it right, and you have the appropriate statistics, you can come up with a risk level based on your strategy’s/system’s win rate and the relative sizes of the winners to the losers. We don’t always have a really good set of performance metrics to work with, though, so here are a couple of quick and dirty ways to get to per trade risk.

**Fixed Risk**

If you’re planning on taking the same nominal risk (say $100) then it’s very simple. You take the maximum amount you’re willing to risk over a given period and divide that by the number of trades you expect to make in that period.

Let’s say you’re maximum risk for the week is $500 and you expect to make 10 trades. Take $500, divide that by 10, and you come up with $50 risk per trade.

Alternately, if you want to think in percentage terms, let’s say you want to set your max risk at 5% for the week and will do 10 trades. In this case you’re per trade risk would be 0.5% (5%/10).

**Fixed Ratio**

The math gets a bit more complicated when you start talking about setting your risk as a percentage of your account equity. This means, for example, that if you decide to risk 5% and your first trade was a loss, your second trade would mean risking 5% of the now 5% smaller account equity (95% of the initial value). That would be 4.8% of the initial account balance (95% x 5%). This sort of strategy means as your account balance falls from a series of losing trades you’ll actually take smaller and smaller risks through aĂ‚Â negative compounding process.

Let’s bring back the 5% risk for the week and the 10 expected trades.

We can figure out what the per trade risk would be by using this base formula in Excel:

=1-((1-r)^(1/n))

Where r is the total risk for the period we’re looking at and n is the number of trades expected for that period.

Plugging in our 5% total risk and 10 trades we get:

=1-((1-0.05)^(1/10))

or

= 1 – (0.95^0.1)

Which works out to 0.512% for the per trade risk.

Note that this is actually slightly larger than what we got in the fixed risk calculation above. That’s because of the negative compounding involved.

**Scale it Up!**

Note that you can use this same sort of process for setting up your per period risk limits. For example, you could figure out what your per week risk limit should be given where you want your per month risk cap to be. It’s the same math, just scaled up from trades to time periods.

**Very Conservative**

Now, in both of the sets of calculations above we’re going on the assumption of a worst case scenario where every one of our trades in the period in question is a loser (or in the case of time periods, that each of the smaller ones sees a maximum loss). That makes it inherently very conservative in nature. To that end, it’s good for setting the floor in terms of how small you should trade. As such, new traders can make very good use of these quick and dirty calculations to keep them trading conservatively during the learning processĂ‚Â (so long as they are pretty close in their estimates of how many trades they’ll do). More experienced traders will want to adjust up based on rational assessment of performance.

**How do you set your risk?**

I like to use the quick and dirty stuff above when looking at more position-oriented trades. They are inherently less frequent, so as such more subject to strings of losses because the law of large numbers cannot necessarily come in to play.

What about you? How do you decide what to risk on your trades?