$1,000,000 One Drop Final Table (part 1)
Posted by Sauce123
Video Player is loading.
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Resume Video
Start from Beginning
Watch Video
10
Loading 22 Comments...
2 questions:
1. At 17:45 you say that in ICM spots that force players to play tight, an adjustment should be for players to use small bet sizes. I'm trying to understand the math behind this adjustment.
Is the idea here that a) the negative EV of losing X chips is greater than the EV of gaining X chips, so players should fold when given marginally +chipEV calls, thus b) to make villain indifferent, you have to bet a size that makes their call sufficiently +chipEV, rather than how we want to make a call 0 chipEV in a cash game?
2. At 39:25 you say that the best bluffs are lower cards and give the example that on the board of [6s 7d Qc As 2s], 3s 3x is a better river bluff than Kh Jh. Does this effect of wanting low cards hold independently of having a spade in your hand? It seems like Kh Jh would be a better bluff than 3h 3d because KJ block some strong Qx and Ax combos.
As always this was an excellent video, thank you for taking the time to make it.
re 17:45, Suppose a hand has a maximal value betting size S under normal conditions. With MCV (marginal chip value), continue ranges for opponent will contract, meaning the largest EV maximizing size will now be < S.
re39:25, I think big spade > spade > big card > small card. The distribution of Ax / Qx kickers is going to be wide in a big blind defense situation and gutters won't be XC turn; so we mostly want to block things that make flushes or can XR bluff representing flushes.
Thanks for the response. I'd never thought about blocking villain's x/r bluffing range before when bluffing IP on the river. Pretty interesting concept, since we don't usually think of villain's bluffs as a portion of their "continues" that we want to block.
Dan, more specifically, given we have a bluff our continue freq against XR is obv 0 (even if perhaps we beat some of their bluffs). The amount of their continue freq that's a bluff will be around 1-A, or in the case of a potsized XR the ratio of bluffs:value will be 1:2. So, with standard XR sizes it can be a pretty big effect.
Hi Ben,
39:20. You say you rather bluff lower cards on the river when Fedor holds KJs. Could you evaluate why? My thought is KJ doing a great job blocking some of Solomons stronger qx that may fold to a river bet.
If you’re bluffing KJs on the river, you want him to fold Qx, so having 33 would be better in that you unblock Solomon’s folding range.
Forzamora,
I think you're correct. In HU play, OOP is often XC turn with K, Q, and even J hi gutters hoping to win at showdown, and the expanded 3b range decreases the freq of TP with higher kickers in the flatting range; so lower cards are preferable for double barrel. In 6m play, even against the wide BB defense range, I agree that higher kickers are preferable because non FD gutters won't be XC turn under ICM; and the no folding MP/TP combinations will be weighted towards higher kickers.
RatedGTO,
This effect gets more important as betsize increases (And is also more important under MCV).
Overwhelming tournament video. First hand. Fedor is playing a extremely unusual game tree (pre and post) on this stacks setup. As far I checked (ICM wise) this board gets checked on the 96th percentile and OOP never uses pot on the turn when he decides to half pot flop. He is definitely exploiting something on this strategy. Pre, probably Salomon and Smith underdefending and post not really sure.
It should rely on one assumption that if IP always 3bet As[K-T] and fold the lower offsuit aces, then it makes Justin overdefend by 1.6% (considering he calls flop with 9d8d, Td9d, JdTd, QdTd, Jd9d, QdJd, Qd9d, KQ-KJ, KxJx-Kx9x, 77, 8x7x, As2s, As9s-As2s,QsJs-Qs8s,JsTs-Js8s,Ts9s-Ts8s,9s8s). If Justin 3bets As[K-T]@50, then he reaches MDF on the turn by calling just KQ-KJ, KxJx-Kx9x, 77, 8x7x.
I disagree on the argument that 22 should bet smaller than AK since 22 have no harmful card removal effects and having a K decreases Justin's calling frequency on the turn by 27% if you consider he continues KQ-KJ, KxJx-Kx9x, 77, 8x7x vs pot. Checking seems a weaker option than betting specially since Justin is not likely to put two bets in with his bluffs very often when OOP turn calling range should be reasonably strong and there are not a large amount of large equity pivots on the river vs that range on K727.
If Justin calls turn with KQ-KJ, KxJx-Kx9x, 77, 8x7x then he doesn't need to defend every Kx by the river and the turn large bet by Fedor is really well suited to his specific hand. Since it is a spot largely impacted by ICM what do you expect Justin's counter exploits to be on this situation (assuming Fedor is somewhat balanced on his strategy) ?
Raphael Nogueira
Thank you for the well thought out analysis. I agree with the observation that many typical players will tend to play unbalanced strategies on runouts like the K727 after opponents XC turn because of the lack of equity pivots (or, in other words, the well known bias that “if nothing has changed and you call the turn, you have to call the river”). I don't think this heuristic/bias is nearly as true in the high stakes tournament metagame. If Justin thinks Fedor's turn XC range is folding >alpha such that a bluff is winning $EV, then I expect Justin to bluff. If Justin doesn't have a strong read (as is most likely) then I expect him to play a GTO approximate strategy which includes some bluffs on turn, and some bluffs on river.
Mostly agree.
Ran the preflop spot on HRC (the program assumes the hand goes check down so there is some weird SB flat calls) using ICM.
Adjusting to the more "realistic" strategy play then 22 is losing significantly preflop.
33:00 I wanted to see the OOP response to the the smaller bet size. I think Fedor correctly assumed that he'll face a raise less than equilibrium on this bubble vs. a medium stack, and if you node lock that into the sim KQ is probably a pure bet.
That seems right.
On the other hand, in almost any situation where you node-lock a raise freq <1/2 equilibrium the result will be pouring tons of betting into that node, especially if the board is dynamic.
35:00 If you watch the video, Justin did raise T8o and they got the action wrong.
excellent vid as always
especially the analysis on the flop texture+pio
love it!
enjoyed the last cash game much more than this live FT tbh. the hands were much more interesting being so deep and not just raise/take it or raise/ c bet/fold like they were in this one.
Pt2 is a lot more interesting imo, or at least has more action
Ben Sulsky cool, looking forward to it. are we gonna get a part 2 for the cash game still though?
Nice Video! What software are you using to analyze ranges in the Video?
PioSolver
thx :)
yo ben loved the video , just want to know @15.44 945 hhh if solomon had to lead what sizing should he choose and also given the risk premium of solomon is pretty huge vs justin would that matter to for the lead?
Be the first to add a comment
You must upgrade your account to leave a comment.