How much to bet when you have an edge

60% win percentage does not make a profitable system. You can be right 60% of the time and still lose boatloads of money. To make the system profitable you would need to manage downside risk and let winners run. Incorporate risk/reward.

 

Trading isn't gambling -- I take issue with your use of the word "bet."

 

Well, the OP claims his system tries to determine the direction of the market and does that correctly 60% of the time. If he is talking about a trend following system, stop and reverse style, then 60% is more than enough for a good profit. If he is talking about intraday high frequency trading, then 60% is very low IMO.

 

Since the system only gives the direction not the magnitude of the move you'll have to have the same stop as your target.

If you want to make 100K per year you will have to risk $2500 per trade.

That is assuming 200 trades per year
120 winning trades and 80 lossing trades.

You would think someone who can develope a 60% winning system would be able to figure this out.

 

What do you mean that no system can "predict" anything ?

We can predict that >99.9% chance that the Sun will rise tomorrow, will it ?

 

Read my OP, our system can predict if market will go up or down 60% times.
We did not say that our system will be correct 60%.

 

I can say with some certainty that the market will go up or down more that 60% of the time.

Closer to 99.9% of the time.

The other option I beleive is to trade sideways all day, I predict will happen 0.1% of the time.

You can copy this sytem by flipping a coin.

 

You mean it predicts the close will be up or down?
Then your stop could be based on a bactest that could show by how much it retraces on average before clossing down. Maybe over a certain % there is no coming back on the red on average and you should stop out.

 

I think he probably meant that the confidence of your system's prediction is very low to be called a prediction due to the uncertainty in prior probabilities. In the case of the Sun you have very high confidence due to the extremely large sample and the fact that the event "Sun will rise tomorrow" is the outcome with 100% success rate.

In a strict sense your system attempts to forecast direction, not predict the direction.

 

I think your looking for you risk of ruin

this example is for a card game, but you can apply it to any wagering game(trading)

http://wiretap.area.com/Gopher/Library/Article/Gaming/ruin.txt