Merge pull request #3203 from jpribyl/update_expectancy_docs

Update wording in expectancy docs and add example
This commit is contained in:
Matthias 2020-04-23 19:44:02 +02:00 committed by GitHub
commit e6d3e2e7d3
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -79,7 +79,7 @@ So lets say your Win rate is 28% and your Risk Reward Ratio is 5:
Expectancy = (5 X 0.28) 0.72 = 0.68
```
Superficially, this means that on average you expect this strategys trades to return .68 times the size of your loses. This is important for two reasons: First, it may seem obvious, but you know right away that you have a positive return. Second, you now have a number you can compare to other candidate systems to make decisions about which ones you employ.
Superficially, this means that on average you expect this strategys trades to return 1.68 times the size of your loses. Said another way, you can expect to win $1.68 for every $1 you lose. This is important for two reasons: First, it may seem obvious, but you know right away that you have a positive return. Second, you now have a number you can compare to other candidate systems to make decisions about which ones you employ.
It is important to remember that any system with an expectancy greater than 0 is profitable using past data. The key is finding one that will be profitable in the future.