forex-scalping.com

Resize every trade or once a month? 240 trades at 1% risk, simulated both ways

Two ways to apply 1% risk. One recalculates it from current equity before every trade. The other fixes the dollar risk at the start of each month, using that month's opening equity, for all 20 trades in it.

I simulated both on the same sequences of trades: $10,000, 240 trades (12 months of 20), each winning 1.4R half the time and losing 1R otherwise, 20,000 runs.

Average final equity:
Resize every trade: $16,201
Resize monthly: $16,056

Median: $15,877 against $15,797. The 10th percentile outcome was $12,495 against $12,552, so the monthly method was a touch better on the downside, and at the 90th percentile it was $20,175 against $19,909.

Per run, the difference averaged about $146 with a spread of $182, so it is small next to the spread of results between runs. The choice matters less than the risk percentage. Anything frequent enough to follow the account works, and a spreadsheet built for either takes an afternoon. These are results from my own simulation with round numbers, so treat them as a size of effect, not as a forecast for any account.

The difference comes from compounding within the month. After a few wins, resizing every trade makes the next trade slightly larger, and after losses, slightly smaller. That pushes the average up a little and widens the spread a little. The monthly method behaves like a fixed dollar risk for 20 trades. At +0.2R per trade, 20 trades make about +4R, which is +4% at the start size, and the compounding effect on top is about +0.1% a month, so around 1% over a year. That fits the $146 gap on $16,000. It is real but nowhere near the difference between a 1% and a 2% risk.

A practical reason to prefer the slower method is fewer mistakes. A position size recalculated on every trade needs the right equity figure at the right moment, and with open trades or fees pending, which equity is it? A monthly figure is a number you write on a card once. It is also easier on a prop account, where the rules often refer to the initial balance anyway. If sizing from a stale balance, the chance of error is in the other direction: after a 15% drawdown, risk is 1% of the old, larger balance, which is 1.18% of the actual equity. A weekly rebase is a middle path that handles most of that.

The tails are close too. In the monthly method a losing run inside the month keeps the dollar risk fixed, so after a 10% drop 1% of the opening equity is 1.11% of what is left, a small effect over 20 trades. The 10th percentile differed by $57 in the simulation, which is about the size of the sampling error for a percentile from 20,000 runs, so I would not read anything into it. If you add a rule that cuts size by half after a 10% drawdown from peak, that would matter far more than either way of rebasing. I did not simulate that variant, so it is for you to test.

You can run the comparison on your own numbers. In a spreadsheet, put 240 rows of simulated R results, with =IF(RAND()<0.5,1.4,-1), then a column for equity under the per trade rule, equity times (1 + 0.01 x R), and another for the monthly rule that uses the equity at the start of each block of 20 rows. Press recalculate a few dozen times and watch the two final values. They stay close together almost every time. The sheet takes ten minutes to build and gives you the answer for your own win rate, payoff and risk, which may differ from my example.

Small difference in the averages, as expected. Monthly resizing is simpler to manage by hand and only slightly less efficient. The main advantage of per trade resizing is that it reduces size faster during a drawdown, which protects the account more in bad periods.