Subject: Re: Rankings for 1Dec2025
I also wonder how this screen would do using the Russell 1000 instead of VL 1500.

Probably just fine. Since the screen leans strongly towards large caps, the overlap among picks will be very high. The differences will come mainly from which particular method is used to calculate ROE and what counts in long term debt.

The only thing I'd want to do is to make sure you're using an out-of-date list of Russell 1000 members--use one that's at least several months old. The distortions around the reconstitution dates are legendary. New entries are big underperformers on average, and demoted ones are big outperformers on average. Most of that is due to simple mean reversion--recent good performers get added--but it's a particularly big thing with the Russell indexes because they are extremely mechanical and easy for pros to front run.

Nobody tries to front run additions to the VL database because no money tracks it.

Jim