I have to say, I'm tired of the timing debates. Seems like some people like to tickle GTR1 to get it to show a big number. I don't trust it when it comes to timing, mostly because I don't have a view if what it is doing internally.

I've evaluated timing schemes in Excel spreadsheets for 10+ years. The data is taken from finance/Yahoo. So you can look at it and _know_ what it is. No question that ^GSPC is the S&P500 index and ^IXIC is the Nasdaq index.
In the spreadsheet you can control the parameters and you can see exactly what's going on. It got to the point where the spreadsheet you can change the parameters -- the index, the thresholds, the lookback, etc. easily and immediately see the results. Then importing those signals and applying them to the historical price data of a screen.

To evaluate the timing scheme I don't look at the absolute numbers, I look at the difference from no timing. First the risk factors, drawdown, volatility, Sortino. Lastly, look at the CAGR to see if it hasn't degraded much.

I have a hard time believing any timing scheme that shows an increase in CAGR. I have very low confidence that is going to happen. It feels like "torture the numbers until they say what you want".

Regardless of what GTR1 shows different between 200 days and 325 days, I do not see it when using the Yahoo data. I give more credence to that. "The answer you like the most is the one you should trust the least."

(Also, in 2006-2026, 52 week beat 65 week in both CAGR, drawdown, and Sortino.)

IMHO, the _only_ useful bit of information that came out of what was presented in the original post was: timing helps and the exact parameters DO NOT MATTER.

::whew:: Now I'm gonna have a little lie down.