Hello I'm beginner (in english too... google is my friend ...) and i try to build a mathematic trading system. I would like to know how much trades for validate a strategy ? Thanks for yours advices.
I usually require a min of 3,000 trades going back ten years for day trading and almost always use just June and Dec contracts for Indexes for methods that trade one to three minute bars. As far as longer term on weeklies/dailies, my min sample size is 10,000, but that easier to get by testing over S&P 500 stocks and going back twenty years plus. I want to know as much about what "might" happen as possible so I can make rules to possibly safeguard myself. Especially, for futures day trading, my style is often for chop for ES, so my methods stop when there is too much volatility. Or if there is much volatility and going in my direction, most likely will go to smaller timeframe of 2 tick range bars looking for first instance of weakness of trend to exit. But much depends on bars ones trades. I know many feel I am too extreme for backtesting, but out of doing it so long, history repeats itself and I am better prepared.
For exmple i test a system and the result is during one year, 50 trades, 49 wins 1 lost. If i need 10000 trades, i need 200 years...
Create a strategy, backtest it. Than forwardtest it for a couple of months. Amount of trades isn't that important imho. I trade a system that takes 20 trades/month, so I've gone live after 50-60 trades as the expectancy remained positive.
The absolute minimum for EOD is 20-30 trades. This is what some statistics books say. For ID there should be way more. Mike Harris has proposed a formula for the number of trades N = 20 x Np, Np is the number of bars in timeframe of operation. For 1-min stock data this comes to about 7800 trades. For 15-min data it goes down to 1950 trades. I think these numbers are reasonable.
If trades formed a Gaussian distribution, you could get by with as few as 36 trades. But the general consensus is that trades form a wide-tail distribution so you need a lot more. Wm Eckhardt insists on a minimum of 1800 trades. I personally think you can get by with about a third of that. Good luck.
It depends on your win rate. I'd be more comfortable with 50 trades at a 70% win rate than 100% trades at a 50% win rate, even if both made money. I higher win rate gives your more advance warning on system degradation. With the first scenario, a fall in the win rate will tell you something has changed far more quickly than with the second scenario. Runningbear
Give me any instrument, and I'll find a system that generates 20 winning trades in a row for that instrument. That's the magic of "curve-fitting". The more trades, the more difficult it is to curve-fit. What I'd suggest is calculating a system quality measure which includes the number of trades. For example: System Quality = sqrt(number of trades) * AveTrade / StandardDeviation(all trades)
Hi Handle, I'm glad to see that you're still sharing your thoughts here at ET and I wish you well. I backtest methods manually, primarily on daily charts, looking at one bar at a time over a period of years and recording entries and exits as I scroll the chart forward. I pretty much always start with January 1st of a year and end on December 31st of a future year. I test enough years to get 50-100 trades and then I calculate several standard performance metrics, such as net profit, profit factor, maxDD, and several others. I do this in one futures market that I trade and then I do it in 2 or 3 other unrelated futures markets that I also trade. A couple of my best recent strategies have had profit factors above 1.5 and win rates of 54-62% in each market backtested. So far, they're also doing well in real-time forward testing in those same markets, though it's only been a few months. I understand that you normally want 1000's of trades in your backtests and I only have a few hundred for these methods. In your experience, do you think it's possible to have confidence in data derived from less than 1000's of trades if they exhibit a certain high level of performance and it is observed in multiple uncorrelated markets? If the subsequent walk forward tests were also strong, though limited to only ~ 50 trades so far, would that be more convincing... or would you still need a lot more data to feel confident that the system is tradable?