You have 2 strategies for predicting the market direction. One has a success rate of 60% and the other strategy has only 10%. Which one is better?
If both equal, then 60%, no brainer. I developed long term futures system and it does steady 4-10% winning trades, but the winners are huge and can take up to five years to complete, so it is then all about managing the losses through hedging.
More background info: The given success rates are not that for just the next day, they rather are generic properties of the strategies. Ie. strategy #1 is correct in about 60% of the prediction cases, strategy #2 is correct in about 10% of the prediction cases. These values were found by doing countless trials in simulations, or by backtesting historical data. Each strategy analyses all available data and predicts whether the market (or just a single stock) will close the next day up or down relative to today's close. For the sake of simplicity let's assume the stock can never close at the same price like yesterday, so there are only 2 possibilities to predict: up or down.
1-P(correct.direction) = P(wrong.direction) 1 - 0.1 = 0.9 just flip prediction sign of strategy 2 0.9 >> 0.6 I'll take the 10% strategy any day; where can I get one?
I kind of hate these simple questions -- the market and things to consider and all its variables are so dynamic at any one given moment.
10% can be better than 60% because at 10% you have a good chance of actually getting more than 10% winners with little chance of getting less. But if you have to get 60% winners to make money there's a good chance you'll go lower than 60% and hence lose money. But risk/reward does of course come into play so the question is sort of hard to answer.