Quote:
Originally Posted by Foucault
Is Cruncher capable of "reverse-engineering" a range? Like, could I ask it for a list of all hands that have 35% equity against a range of {KK,AA, AKs}?
Sounds like a great program regardless, but this is the feature that would really make it stand out for me over something like Pover Stove.
Yes, PokerCruncher can already do this, with the range equity breakdown heat map feature. Good question, such use cases are the reason I (and customer M.S.) wanted the heat map feature last year.
Your example range {KK+, AKs} is very strong; I ran this range vs. {all hands} in Cruncher and only AA has an equity higher than 35% against this range (AKs is around 33%, KK is around 32%).
So I'll use a weaker range here as an example: {top 10% of hands}. To run this in Cruncher:
Player1: range:{top 10%}
Player2: range:{all hands}
The heat map and detailed stats of {all hands} are below. The heat map has a slider that lets you set the upper/lower bounds of the color spectrum. You want to see all hands that have >= 35% equity against {top 10%}. Breakeven equity is 50% so the diff is 15% so I set the slider to 15%. This makes all cells that have <= 35% equity bright red, and all cells that have >= 65% equity are pinned to bright green. So all cells that aren't bright red are the answer to your question: 22+, A9s+, AJo+.
The heat map gives a quick visual picture; you can look at the detailed text stats to dig down. However the Monte Carlo error margin for each cell of a range is higher than the error margin for the entire range because each cell gets only a small fraction of the total Monte Carlo simulations, so if you want accurate heat maps then simulate a lot of hands. I sim'ed 5 million for the below picture. The recent 4x calc speedup helps with this.
I've considered adding a "Sort" button to the detailed stats but postponed it. Maybe I should add this, now that the calc is faster and the detailed cell equities are more reliable.
Is this the kind of info. you wanted to see? This is one way Cruncher goes beyond PokerStove; I think the Deal-To-Flop/Turn/River/HoleCards and the many additional stats (odds for flopping draws and combination draws) are also useful additional features over Stove. Do send suggestions for improvements; I'm not done with this app, thanks.
{top 10%} vs. {all hands}, heat map of {all hands}:
Detailed breakdown of {all hands} (exported out of app; abbreviated):
Range Breakdown: Player 2
22+, A2s+, K2s+, Q2s+, J2s+, T2s+, 92s+, 82s+, 72s+, 62s+, 52s+, 42s+, 32s, A2o+, K2o+, Q2o+, J2o+, T2o+, 92o+, 82o+, 72o+, 62o+, 52o+, 42o+, 32o
Range's Equity: 32.0%
Breakeven Equity: 50.0%
Num hand combs in range: 1326
Num hand combs possible: 1326
CombsInRange/Possible: 100.0%
Hand, (%age of range, number of hand combinations), Equity
AA (0.45%, 6 combs) 85.2%
KK (0.45%, 6 combs) 72.9%
QQ (0.45%, 6 combs) 64.6%
JJ (0.45%, 6 combs) 56.7%
TT (0.45%, 6 combs) 51.7%
99 (0.45%, 6 combs) 46.8%
88 (0.45%, 6 combs) 43.0%
77 (0.45%, 6 combs) 42.0%
66 (0.45%, 6 combs) 41.9%
55 (0.45%, 6 combs) 42.0%
44 (0.45%, 6 combs) 41.5%
33 (0.45%, 6 combs) 40.4%
22 (0.45%, 6 combs) 40.4%
AKs (0.30%, 4 combs) 59.9%
AQs (0.30%, 4 combs) 53.4%
AJs (0.30%, 4 combs) 46.2%
ATs (0.30%, 4 combs) 42.8%
A9s (0.30%, 4 combs) 40.0%
A8s (0.30%, 4 combs) 38.0%
... truncated ...
AKo (0.90%, 12 combs) 57.9%
AQo (0.90%, 12 combs) 50.7%
AJo (0.90%, 12 combs) 43.5%
ATo (0.90%, 12 combs) 39.1%
A9o (0.90%, 12 combs) 36.0%
A8o (0.90%, 12 combs) 35.2%
... truncated ...