Adrian Reid explains why testing a trading system on half your data then checking a few profitable trades on the other half doesn't predict real performance. The same handful of trades often dominate results, making you too optimistic. Applying a system to other markets requires much more than basic backtesting.
is if we take the system and we apply it to half the data, let's say, and out of the 600, 10 are profitable, right? We say, okay, let's take those 10 and let's validate out of sample that they're still profitable. And then we get 5 of those are profitable. It's the same 5 that we got when we tested it over the entire history. So we still end up with the same problem in some ways. I see. So it's a little bit tricky, right? It's— it might be better, but I think the key is you've got to take that in-sample data and you've got to do something more with it. And we'll get to what I'm suggesting in a moment. Okay, so I think the key is it's not trivial to take a single ETF system and then find all the other markets that you can trade it on. You've got to do a little bit of work here. If you want more trading tips from Enlightened Stock Trading, make sure you give this video a like and follow Enlightened Stock Trading so you don't miss the next video. And go to enlightenedstocktrading.com/free to get your free Trader Acceleration Bundle to start improving your trading results today.
