^
(it's not so late here, west coast USA)
Good/interesting work FreeportMainePoker!
Some fyi's and impressions (but note, take this with a grain of salt because I'm not a full poker pro, mainly a 2/5 and 1/3 NL weekend grinder the past 10+ years, but profitable every year, and I did play 1000+ hours many of those years):
1) Looks like you have the Mac-Expert version of this app, so you've probably noticed the "Ranking" selector which has 8 well-known hand rankings. But your ordering is different in that it's a flop-playability ordering so is a good new ordering to look at, whereas these 8 are mainly hot-cold equity going to the river. Tutorial section:
https://www.pokercruncher.com/ipPoke...RankingSystems
2) You had mentioned your playability ordering is for 2/5 NL cash games, or even 1/3. A note here is that these low limit cash games are usually shallow effective stacks, so flop playability (ability to flop/turn draws) won't be as important as hot-cold equity. Because stacks are often in by the flop or turn if both players pick up something, so you can't use your ability to flop and turn draws and combination hands to bet people off their hand as often. I think your playability ordering would be most useful in deep effective stack games, and/or maybe at higher limits.
3) Occam's Razor - The 8 well-known orderings mentioned above have one thing in common - they are all defined by very basic rules, e.g. "PokerStove - preflop all-in equity vs. 3 random hands". So I think that for your playability ordering, I'd make your stats/rules simpler: for made hands I'd go with what you have: TwoPair+, TopPair, OverPair. But I'd simplify the draws side: just FlushDraw, OpenEndStraightDraw, DoubleGutStraightDraw, and that's it!
That is, I wouldn't go crazy detailed with what stats I include, and wouldn't even consider the combination stats. Not just for Occam Razor simplicity principles, but because the hands that will flop e.g. a BackdoorFlushDraw best are prob. the same ones that will flop a regular FlushDraw best. Same for the combination stats. So there may not be much additional value in including more detailed stats. I think the ordering with this super-simple set of stats would work out well because e.g. T9s would still clearly be ranked above 65s because T9s will flop TopPair more often, and wouldn't be counterfeited as often.
4) A final/complete ordering should prob. have both playability characteristics as you're investigating, and hot/cold equity characteristics, say a weighted average of these. Because in the end / final ranking we definitely want AKo to move up quite a bit.
Just some 2c, take with grain of salt!
Interesting work by you.
-RJ
Last edited by rj999; 09-27-2019 at 12:50 AM.
Reason: Added 4)