Some random speculations on Ah3h hand.
I think because the line is so rare - 4% of overbets, most hands that will defend well, are more incentivized to block or bet themselves, not check-defend. Also, card removal is important when you decide to defend vs this sizing, but it can be better used to bluff with rather than bluff-catched with.
April 6, 2020 | 12:58 p.m.
Great post as always.
Just a comment to 4. if you go from 5 to 7.5bb it's not getting a lot better if you take into consideration rake. Pre-rake winrate just went from 12 to 14.5 (assuming 7bb rake) which is 20% increase in "skill". Granted it should result in 50% increase in winnings.
Dec. 21, 2019 | 1:14 a.m.
Thanks for pointing that out. Just got into a mini-downswing after reading your initial post :) and when I was looking at my punts in PIO I was thinking, ok bad - but it's like 2-3, max 10bb mistake which made me come to the same conclusion. So you are definitely right about that.
What methods do you use to quantify the magnitude of the errors? Do you node-lock or just take deviation from equilibrium play as a default?
Dec. 7, 2019 | 3:10 a.m.
Good post. Many people don't actually realize that you can be playing a -10bb/100 style very easily even if you are a "7bb winner". At least for a 1k/2k hands sample which can spiral into a number of BIs. I've yet to experience a downswing where I didn't spew at least 5-10 BIs.
Dec. 5, 2019 | 1:33 a.m.
Very well written post, kudos to you.
I think you did think in binary terms - "So the key to not tilting is to remove your expectations" - that's why I used the rather strong zero :)
Regarding the pain level, there is an image from the Vikings TV Series of one of the fiercest ones, Loki, enduring a water-drop torture. It broke him, although pain must have been 0.001/10.
I would say rather than try to avoid tilt which is I think we agree impossible, it's best to try to mitigate its negative effects by taking a break for example, or to engineer situations where tilt is unlikely to occur in the first place, i.e. playing properly rolled.
Sept. 30, 2019 | 11:53 p.m.
I don't think it's realistic for the human psyche to have zero expectations especially when effort has been invested.
Moreover, there is running bad and running bad. I am pretty interested in the reaction of a non-tilting guy after 35BI downswing.
Sept. 28, 2019 | 1:13 a.m.
In my experience, loss aversion is much more a part of bankroll management. I doubt that somebody would be fearful if they had 200BI for the level they are playing at.
IMO poker is a hard game because you never truly know if you are actually playing well or not. Sure, we have models and GTO solvers and what not but in the end of the day, we are not 100% sure we did the correct thing.
Sept. 12, 2019 | 1:12 p.m.
Thanks for editing your post. I appreciate your new insight.
July 20, 2019 | 3:40 a.m.
Thanks I guess, but the question wasn't regarding so much the 4bet and the validity of the ranges. More interesting to me was range distribution and how it affects betting frequencies.
July 19, 2019 | 6:36 p.m.
I rerun the sim with 66% OOP flop bet (clearly not a block bet) and now it bets even more often - 87%.
Some other analogies maybe?
Another point is that if OOP is forced to check, IP bets only 5% for the block 1/4 size which puzzled me too.
July 19, 2019 | 2 p.m.
Credit should go to Kevin Wang who made the conversion.
I glanced at the scientific paper because of the mismatch between results and ... results -> they have found a method AIVAT that can give you your "true winrate" based on a smaller sample but with a certain margin of error. So the true winrate is something like 4.8bb/100 when they have used this special methodology but within a margin of error of 1.3bb or something (cant remember exact numbers).
July 19, 2019 | 1:56 p.m.
I stumbled upon something today while playing: I 4bet OOP (COvsBTN) and the flop comes JJTr.
I did cbet almost out of habit 1/4 pot, however, this seems like a flop that certainly favor the IP 4bet cold-caller.
With the following PIO tree parameters: equity runs almost 50-50% but with very clear nut advantage for IP, both proportionally and as given hands.
However, PIO is still wanting to bet 75%! for the 1/4 sizing OOP.
Does anybody have an explanation?