"Attacks" welcome; I used to be a lawyer so I don't view analysis as "attacks", but an opportunity to learn.

I wasn't suggesting anyone only use any particular screen. I have screens using 1 to 20 stocks, each backtested in detail.
My intention is to see about over-weighting the narrower ones so when a 1 or 2 stock strategy holds up on its own, and a 5 stock strategy picks it, I could double the 1 or 2 in dollars and test that.

You can just pick all Nasdaq100 stocks and see on gtr that 325 days beats 200 for a timing test on SPY.

I appreciate your earlier responses; it helps me then go to AI to examine things. As you noted PHL is particularly good and screens using it that use momentum are looking at similar universes.

My experience has been that the backtested results are real.

I realize the difference; years ago I hand-tested Zacks' claims on some of their touted screens and discovered that in real life you got 1/2 their supposed returns but in their case it was because their data is updated on Sundays but their backtests pretend you had it in your head to act on at Friday's close, i.e. 20-20 hindsight. That was probably 20 years ago and they still haven't fixed their backtester to account for this dishonesty.

The benefit of AI and Norgate data is that I can see every single trade; I can examine the news on each when there is a big move to see why, I can run tests removing the best performers to see how much that matters, and run any number of other tests gtr can't, and don't have to trust the blind data of gtr. In general tests in gtr match my AI tests; I can't always get AI to interpret gtr but my goal is to make money not use gtr's limitations and defintions, and I was able to quit being a lawyer 28 years ago due to mechanical investing despite people telling me I was insane since I didn't have much money. I subscribe to Warren Buffett's mantra "put all your eggs in e basket, but WATCH THAT BASKET". I have been willing to change strategies over the years as evidence dictates, which is my attitude toward life in general: be a Bayesian.

I realize not many people are willing to keep going through big drawdowns, but my goal wasn't emotional satisfaction but making money, and decades ago I read the study that said staying invested beat timing and was skeptical of timing until now. But now I have data and timing improves 85% of my screens and by a big margin, cutting max drawdowns by 1/3rd to 2/3rds. The strategies I shared had better beta, Sharpe , UI, mdd, recovery time from mdd, average recovery times from -10% drawdowns, rolling 1,3,5,10,15 year time periods and more.

The only thing they don't improve on from an index fund is they have more -10% drawdowns (not a shock when you have few stocks).

Trailing stop limit sell orders did not pay off in my tests.

I am a big believer in: if you make money your way I am happy for you. Most people shouldn't do what I do; I tell my friends to buy index funds generally and be willing to put up with market-level mdds and 15 year long recoveries. But I fully expect to do better than that, and have. My friends doing what I am doing with the strategies discussed this year are up well over 100%.