Day Trading & Forex Forum

meanreversion

It's generally assumed that you need to "run profits and cut losses".. but some casual backtesting shows that you can make money the other way round as well. For example, if you had operated a 1:2 ratio (take profit after 0.5 pct move or stop loss after 1 pct move) in EUR/GBP over the last few months, you would have made money using RSI to get back on the trend. Obviously the win ratio goes up nicely and more than compensates for the occasional large loser. I don't use this method myself, but I'd like to hear from anyone who does.

tenbobtrader

"Obviously the win ratio goes up nicely and more than compensates for the occasional large loser"

you no it will continue

You will ride the eventual multi lose period out for how long exactly

The Risk of Ruin

The Edge

meanreversion

It's just a simple percentage target. Spot goes 0.5 pct in my favour, I take profit. 1 pct against me, I stop out. So this is looking for small wins whilst tolerating losses which are twice as large, in direct contravention to the "golden rule" of trading. Except that it makes money if you know what the trend is and enter on pullbacks. This results in lots of 0.5 pct wins and some 1 pct losses. The win ratio goes up to 75 pct+ PROVIDED YOU KNOW THE TREND. I've never heard of anyone using this method, but it can be made to work, and psychologically is quite rewarding as you get lots of "wins".

meanreversion

The drawdown is not that bad. With a 2:1 profit/loss target, it's not uncommon to have 5-10 losers in a row. With a 1:2 method, that is much less likely.

Is there anything statistically wrong with this approach??

cloud321

i don't like anything less than a w/l ratio of 1

it seems to me that settling for a lower ratio, means a success% to breakeven goes up steeply
whereas settling on success%, means that w/l goes up shallow comparitvely

so running winners takes less effort

if that makes sense
there's probably some math word for it

tenbobtrader

"and psychologically is quite rewarding as you get lots of "wins".

agree and there imho lies the possible seed of ruin

you learn at deep level to perhaps do the wrong thing

think it is safer the other way round myself, add in slips and com and that bad run gets pretty psychologically hard to take

meanreversion

My long term win:loss ratio is around 35 pct, but that's because I've always been taught to run winners/stop out of losers. I suspect most trading robots operate a 1:2 w/l style in order to show very high pct win rates, maybe I should work on one of those and peddle it for £49.95 a time.

rawrschach

No there isn't, but everyone has read in a book not to do it so they'll tell you not to. You figured this out on your own, with your own work, so don't let someone that doesn't know what you're doing tell you it won't work.

IIRC one of the market wizards traded with a wider stop than target.