And you're the one who thinks a 1-minute timeframe daytrader needs 10 years to experience a regime shift.
I didn't say that. You are putting words in my mouth. You claim time span doesn't matter. So by your logic, it's okay to evaluate a 1-minute trader over 1 year of time span so long as he does, say, 10k trades? <- this is clearly amateur hour thinking.
Minimum number of trades required before a system can be trusted to be a good system should depend on trade frequency and hold times. So, if a system trades once a day, and it only fires a trade once in 2-3 days, it will have about 80-120 trades in a year. Take a 5 year period, and we are talking about 400-600 trades. 10 year period and we are talking about 800-1200 trades. However, there are good systems that trade even less like once a week. For such a system, a 10 year backtest will give only 500 trades and a 5 year backtest will give 250 trades. Then there are systems that trade maybe 10 times a year. These kinds of systems could be toughest to follow because of less number of trades.
You don't think 10k trades are a sufficient basis on which to evaluate a system regardless of time span. It's clear to almost everybody here but you who the real amateur is. Welcome to my ignore list, newby timewaster.
Not to take sides, but just use common sense. If a one yr time span had a very low volatility trending environment and you optimized 1000 1 minute trades that had a very inherent long bias, then used that as a proxy for your system performance, what would happen if the next year was a choppy high volatility downward trend? What about that backtest gives you any confidence of the next year? My own experience shows that the backtest should have some kind of validation method to show how robust it is to changing conditions. And a 1 yr window is way too short to draw conclusions. Maybe if someone is using video game style discretionary trades and demonstrated 1000 good trades, there is something to it. But, from a systematic perspective it's not too good. If you honestly want to see some examples to illustrate what I've been saying, and what some of the posters are trying to share, go look up fudancy presentations. You'll see some very nice smooth short term high frequency equity curves in the presentations (they'll even share code to reproduce these simple low parameter systems!). That alone shouldn't give too much confidence in the systems (without additional analysis) though. 2c on price It's a catch 22, because the price should somehow be proportional to the track record and longevity, but if that's missing, then you somehow have to rely on the backtest and methodologies. The problem is if they have to describe the methodology in enough detail to convince you, then you'll probably no longer need them.
You making assumptions based on one timeframe ("a very low volatility environment") and applying them to an entirely different timeframe. Okey dokey.
The first thing I look at for a system, is the ratio of its average year P&L to its max.DD ... but historical DD doesn't cut it for me, and the 2nd thing I do (pending having the distribution of trades outcome) is to run a MC simulation for 1-year worth of trades, and then use 5 std-dev away from the mean max.DD to re-evaluate the 1-year avg. P&L / max.DD ratio. I use that 5 std-dev away from the mean max.DD as my "system stop" - it is the capital that I am willing to put at risk if I start trading that system. A 1:1 ratio (1-year avg. P&L to max.acceptable DD) provides a "risk free" system at initial position size in 1-year (2:1 in 6 months, 4:1 in 3 months, etc). IMO, the value of a system should take that ratio into consideration. One aspect that has only be discussed indirectly here, is the expected lifespan of the system. In my view, positive & consistent backtest of the system over last 10 years (or more) is the best indication of the system's ability to survive market changes, but obviously no guarantee. And given that historical data is readily available, there is no excuse as a system designer (& seller) for not testing the system over as much historical data as can be purchased. Just for the fun of it, I am submitting the attached performance report for "valuation" This version of the system can (and do) take multiple concurrent trades. It has indeed 12,674 distinct entries in the course of the 6 years of backtest, but only ~3,300 exits.