I try it again, to make it simple:
- number of loosing trades
- number of winning trades
- avarage wins
- avarage losses.
Let's count it for one year.
These gives us in percentage of the portfolie winning.
There are also
- longest number of loosing trades in a row.
Or actually it is better to calculate as the
- maximum of (loosing trades - winning trades) in any given serious of trades.
This gives us the "risk".
There is one more thing, however:
What is the likelyhodd that the risk factor increases or decreeses? Can be counted by avarage and deviation. However, I would say that usually there is not enough data to count this.
Instead better to think it this way:
It is possible that the strategy turned bad. The market has changed. We detect it by noticing that the above maximum exceeds well our back testing. Let say by 20%.
Now, how much money are we willing to loose?
maximum sequence * 1.2 * avarage losses <>