I am new to the forum and have not yet read all the previous messages. If what I say is old hat, please ignore it.
In comparing systems, I note that some systems administrators use constant trading quantities whilst other vary their position sizes. In every case where trade size varies, when I look at the returns and divide by the lots and calculate the geometric mean which means the compound return, there is a significant difference in the results. On one system I looked at the result varied 50% higher and the other 300% higher. This on the same entries and exits. Not possible you may think. I can show the numbers and will respond to any challenge.
A second consideration in this context is that the largest loss will come suddenly and without warning. A stop loss is kicked or there is a gap. You cannot control the point count but the lot size can be kept in line with the rest of the trading.
The simple truth is that if you can bring down the standard deviation of returns, the geometric mean shoots up. If trading systems are optimized on a geometric mean the results can be much improved. Getting the smoother equity curve contributes as well.
My problem is that a system I want to trade carries this burden and would be much improved by trading constant lots.
I don’t understand what you’re trying to say in your second paragraph. Do you mean that the coumpounded and the uncompounded annual return are different? We just discussed that in another thread.
