Backtest results

Without getting into the religous discussion of wether backtest results are any good or not, Are these good results?

Time frame is 15 minutes, EURUSD, 2/1/2010 to 2/27/2010, 10000 start balance, lot size 0.01 (no money management), 100:1 leverage.
Profit: 365.11
Total Trades: 35
Profit Factor: 2.36
Expected Payoff: 10.43
DrawDown $: 114.9
DrawDown %: 1.11%

I've done a Google search for Optimization results and how to read them; there's more opinion out there than fact.

"Then why ask here"? Because I'm sure to get expierenced traders in here as opposed to those that are selling something.

TIA (Thanks In Advance).

You're asking such an open ended question, I don't think it's going to be possible to avoid dispensing yet more opinion, I wouldn't live trade anything off such a small sample size.

I see your point. I'll have to extend to a full year (2009).

I'll do that then post some results.

What I'm basically looking for is what Expected Payoff, Profit Facto and the 2 Drawdown's mean. I've done research and still basically don't understand them.

"drawdown" is simply the maximum consecutive loss. Your tester should also show you how many losses in a row occurred during your test to provide that number.

"Profit Factor" is the average winner divided by the average loser. Again these numbers are clearly labeled on the test results. It is the "math expectancy" of the system.

"Expected Payoff" is (I think) what the system is expected to make on average win or lose. Not really sure about that.

As Craig pointed out the data sample is extremely small and February was a month of large moves and tends that ended in large choppy markets. You'll need a much larger time frame to really show how your system worked.

I'm sure you've already heard it but I'll just put this out there just so you can be aware of your results...try to avoid at all cost "back test optimizing". That is by tweaking one or two little setting to get a better result on your testing. It seems like genious at the time but never works out in forward or real world testing. Each little "adjustment" you make get your system to "fit" the data will only make it worse. trust me on this.

If you system works well it will work well in ALL conditions...if it doesn't...well back to the drawing board.

My final point is that there is no program, no computer algorythm, no hard logic that can match the speed and accuracy of execution as the human brain. But I'm sure you'll keep trying, as I did. But just so I've said it...put your time to better use and improve your manual trading...the pay off is far greater the the frustration and headaches far less...

Thanks for the advice.

"put your time to better use and improve your manual trading". I have. I've got my manual system down to about a 96% success rate on trades. I'm now just seeing if I can code it.

The reason for coding it, is that I still have a day job; I can only watch about 3 pairs a day; my system only works (at the success rate I mentioned) on 5min and 15min timeframes (depending on pair). So, I want to see if I can code. And I want to be the one to code it, not someone else. I just like doing things myself.

Ok, I changed the dates I used. I used 1/1/2007-4/30/2010. Hopefully, that's enough time.

Tell me, in your opinion, which pass do you think would be a good one? Now, I know, that once a pass is chosen, I should actually forward test it (maybe along w/ backtesting instead of optimization).

I've attached the CSV file of the results. This way it can be loaded into whatever spreadsheet program you use.