Analizing push or fold situations, should i use reset strategies and reset regret?
Should i use this options when analyzing a simple raised and 3 bet all in situation with HRC? CI of 10 is alright? i listened to bencb saying that CI in 5 is more accurate for this push and fold situations.
3 Replies
You only need to reset regret if you lock something and therefore want to start the solve fresh. If it's a simple push fold spot it should run down even to CI 5-1 very quickly - remember you only need to run the subtree you're interested in, not the whole tree.
I don't understand OP and several of the terms in it; can someone explain them to me?
CI is HRC's measurement of accuracy - the lower the more accurate, kind of like Pio's exploitability.
Reset strategies and regret is basically restarting the solve, getting rid of whatever has run so far. This is important if you run an equilibrium solve and then decide to nodelock a range to see the adjustments. If you don't reset, the new 'exploit' samples will run on top of the millions of GTO samples that have already run, and you'll get a weird merged strategy of the two - or won't see the exploitative adustments as it's hard to overturn the millions of samples where they played perfectly.
