If someone talks up a trading strategy/signal/indicator/whatever by talking about it’s win rate (% of profitable trades) then huge alarm bells should be going off.
-
Show this thread
-
Expected profit from any strategy is E(pnl) = P(win) x E(pnl|win) + P(loss) x E(pnl|loss) — the win rate P(win) is one factor but the expected PnL conditional on a winning vs losing trade is equally if not more important (also much harder to estimate).
1 reply 0 retweets 4 likesShow this thread -
It’s trivially easy to construct a strategy with a high win rate. Just trade randomly and set a tight take profit and a loose stop loss. Your win rate will be high (at least 70%) but your losses on the 30% losing trades will more than cancel out the gains on the winning trades.
2 replies 0 retweets 6 likesShow this thread -
The corollary is that someone who only talks about the win rate of the strategy/indicator is probably hiding something about the distribution of PnL.
2 replies 0 retweets 3 likesShow this thread -
-
Replying to @NewRiverInvest @macrocephalopod
Good reminder. What is meaningful time interval to look at PnL. Monthly? Daily? Per trade?
1 reply 0 retweets 0 likes
As granular as you can get, ideally. Per trade is best if the holding period is < 1 day, otherwise daily is probably fine.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.