you don't have to take me seriously, it's the internet

Are you going to respond to my points about how many players are remaining in those solutions you posted?

Can you even provide that information? Is it even available? Do you even know the details of the ICM model they've applied there?

If not then you're making the same mistake that bankers made at the height of the '08 financial crisis. Model risk is a serious issue in environments with imperfect information.

well I think it's lazy to bust out the ICM laser gun (Zap! Pow!) because there's 75 players remaining and not seriously consider whether that's reasonable or not, and whether the tools you're using have even been configured to solve for that consideration

and I think it's lazy to not even post assumptions about how people react to all your decisions (what exactly are your assumptions, anyway?), and instead blindly copy Nash equilibria as though there's no capacity to exploit against $500 MTT players

The model I posted was 27 left with 30ITM. So it is a low ICM spot, so would a 75 players left be a low ICM ITM spot. However, the 27 left is closer to the 75 left than cEV would be. You have shoves at 40BBs but not at 50. I have posted my evidence. You have not posted your evidence why you think a cEV 50BBs solution is better than a just inside the money ICM solution. You cannot just argue against my evidence without posting counter evidence. This is a logical flaw. Post a counter ICM solution in a similar just inside the money spot to prove me wrong.

You can say anything you want. Especially if you bring up player tendencies at Philly Live. Because to me that is more of argument to not shove AQ here, since the CO range is tighther than GTO. See you are just cherry picking the parts you want. You belittled another poster for not wanting to shove 80BBs here, when you are clearly wrong. that is mostly why I responded. When you talk down to another trying to seem superior to them with your cEV chart knowledge, but you were just wrong.

An ITM 3 table strategy posted on GTO Wizard is not 75 left. Yes, we can agree on that. But my point is that it is closer to this spot than you want to give it credit for, because you do not want to be wrong. ICM matters, even if it is low ICM pressure, it matters. I know for many years players have sort of ignored ICM for a bit once inside the money. I did that too for 15years. But I have learned by studying that I was wrong to do that.

GTO Wizard runs their sims off of Monker for cEV and HRC for their preflop ICM solutions. Their ICM solutions are set for a 200 player MTT. The relevant ICM charts starts at 50% of the field. So they have run solutions for 100 players left with 30 getting paid.

And as I said, running a solution for 75BBs with this exact chip structure in HRC is better. But the 27 left solution of 200 is better for this spot than cEV would be. You are still ITM in the small pay jumps portion of the MTT.

This is the parameters for the GTO W ICM solutions:


30 get paid in their solutions.



Final 3 tables at 8 players is 24 left, ITM with small pay jumps till FT bubble.

"...he said, without evidence..."


Look its a game of imperfect information and that includes the fact that we can't quantify a lot of spots because the calculations are too complex. Again, you need to simulate the tournament 75! times to do that here.

I don't have the capacity to do that--and neither do you.

Do I need a disclaimer? "All posts are the sole opinion of Eggs and no one else."

I'll say this: if you are correct that ICM is that impactful here then congrats, you've collected a lot of EV because no one else seems to be heavily considering it if at all; no one else mentioned it and 6 distinct people including OP responded before you did. You're also risking significant EV if you're wrong by distorting so heavily. And that's great! I genuinely mean that. There is no reward without massive risk.

I applaud your commitment to different considerations than others' in this spot, I truly do.

We'll just have to agree to disagree regarding out INTUITION in this spot. Let's be very clear, the application of ICM distortions here is entirely a matter of intuition. There is no closed form solution. I personally think it's silly to conflate 27 man ICM with 75 man ICM and you think I'm silly for thinking that. Fine.


"...he said, without evidence..." (and I looked at solution that's tighter than CO Nash opening range--though apologies, I didn't specify)

Opening range isn't the only consideration either. Still waiting for your response about what CO does relative to equilibrium facing a ship. Do you think people tend to defend well vs that? I can certainly produce evidence that doesn't happen (though at this point I wont).

OK now I'm a little upset honestly that this is your perception. I didn't belittle anyone. JK is a long time poster and we have a lot of back-and-forth. I don't think anyone except you has taken any offense to anything I've said and nor should they. This is the internet. And the white knight routine--swooping in to save the day from me and me hidebound ways regarding early post-bubble ICM--is probably neither needed nor desired by anyone. People can fight their own battles if they wish. Or not, that's fine too--it's the internet.

The OP is also an investor in my WSOP action this year. So if what I've written came across to you as belittling then that's entirely on you.

Even in this back-and-forth we're now having I've never once called into question your competence, I've merely expressed skepticism that ICM is anywhere near as impactful as you're claiming (which is ultimately a futile argument for both of us--the only salient think to take away from this is that some people will go ham applying ICM here and others will ignore it entirely) and I've tried my best to get you to think critically about the precision and accuracy of solutions that purport to contain more simulations than there are atoms in the universe.

GTOW has every incentive to get you to believe that everything you see should be taken at face value; I shouldn't be flamed for suggesting that you be skeptical of that.

And for the record:



you haven't proven this one bit

This would be a quite tight opening range for CO relative to Nash while still taking into account the live poker silliness you'll come across like people opening A3o (I see this online occasionally too and honestly its not even that silly under reasonable assumptions regarding how often people reraise [exploitably infrequently] and in particular how often BB defends [exploitably infrequently]:




SBs calling range which is quite a reasonable real-world assumption in my opinion:





Here's our squeeze 80bb strategy:



Here's CO unexploitable calling range:





I believe CO will massively overfold and AQo just prints gobs of chips. To me the incentive to play a pure ship strategy--perhaps exploitatively ignoring ICM in the process, or perhaps not--is so compelling that no other considerations come close.

Just my opinion...

Ok I will give you that. I did not know you guys had a history. But in the past this is why I haven't posted here on 2+2 for years because I saw too much of that. So maybe I jumped the gun at his defense though.

My evidence of the Philly Live player pool and AQ not being a shove is I am in Philly. I play the Philly player pool 3 - 4 times a week. In the standard Philly pool of players CO is opening too tight and therefore, AQo is slightly downgraded from the solvers shoves at 50BBs in my mind. And as a player exploit, I think 3b small to consider a fold versus Philly population. I know it sounds nitty, but many in the Philly player pool doesn't find 4 bets with less than AK, and even at that many will only call with AK/QQ instead of 4b. However, they also will call hands from bottom of range that should be a fold, so I am perfectly ok with 3b this player, lowering the SPR to make it better for AQ when we hit versus his too wide range versus a 3bet.

Actually, (and I know this helps your argument more than mine) I think you are too tight in the SB compared to the Philly player pool. They call wider than that. But also, they have hands like KQs and AJs in their call range instead of a 3b range more often than not, and they will call 77 - JJ more often than that as well.

Well we'd love to have you back. The more the merrier. Network effects are very real in a forum setting. I think you'll find we have a small but deeply engaged pool of posters for whom insolence of any kind will not and should not be tolerated. You won't find that kind of attitude here anymore, I guarantee it--even if the terse and prosaic nature of online communication sometimes gives that impression

Can we at least agree on the following?

-even if CO has 50bb, ignoring ICM and under certain assumptions about how wide he opens we can unexploitably find 3b ships; if CO has 30bb we can find a lot of ships unexploitably
-sometimes an action can be computed to be so compellingly +cEV that ICM can be ignored
-ICM calculations are immensely complex and solutions beyond a handful of players should be viewed with a critical lens and in some cases just taken with a grain of salt

I didn't read responses. I dont play this big. My 2 cents:

I may 3 bet slightly larger preflop because of noted sb excessive cold calling.

Id fold vs flop check raise. Aq is now a pure bluffcatcher(Q outs are very dirty, A outs are slightly dirty) with little chance to beat the value range sb will almost always hold. revenue from aq vs flop check raise must come in significant proportion of winning unimproved, due to the small portion of revenue provided by improving to a winning hand.

The ev of winning unimproved vs flop check raise is thus heavily dependent on the sb bluff range(1 - avg non value equity)(non value check raise frequency) = a rough idea of your near pure bluffcatcher's pot share if all in, with no other ev sources contributing any significant showdown winnings.

Opponents vary their non value range construction. Some mix a ton of combos. Some semi bluff exclusively. Some check their real draws and bluff with junk. Some will try to out level you with gambit variation.

Without reads indicating how the opponent constructs his strategy, im folding the flop. The opponents that construct their strategies in a way that causes aq to perform well are the exception in my experience.

1. Yes at 50BBs with no ICM there are ships as deep as 50BBs here and AQ is one of the hands in that range.

2. Yes and no. Yes we can math out 50BBs being a profitable shove with AQ, even in ICM. But there are 2 other contingencies. Just because a hand is profitable to shove does not necessarily mean that it is the most profitable option. If shoving is +2 BBs but 3betting is +2.3BBs then shoving while still beign profitable is not the best choice. And secondly, when you move a hand from the 3b into shove how does that affect the rest of your range when playing your range matters? In a $500 at Philly live, maybe total range construction doesn't matter.

3. These ICM solutions are not perfect for this situation. However, I believe that there are certain heuristics that can be derived and can be consistent across several runs of different ICM conditions. There is an ICM risk premium in all hands in a MTT. However, in the early stages it is sub 1% and can be ignored. In the sample I gave I think there would be about a 5% risk premium, which is low compared to the possible 15%+ on teh bubble or final table sims. I would estimate that 75 left in a larger field would also have a risk premium around 5%. So in that respect, while this is not the perfect ICM run for this specific hand, I believe it is close enough to derive heuristics from the solution given to give at least a solid foundation on which to derive answers from in this spot.

Is no one accounting for flush draws being played aggressive? or KQ/QT in range? This doesn't seem like a clear cut fold to me. Sure he has 2 pair, but doesn't have all of the sets and may have other weaker QX betting for value. He has KT and T8.

I'm not saying this is a slam dunk call either. It is close. I would basically decide based on any reads or tells I have on this opponent in game.

Reasonable chance it's not even an exploit to fold flop. Admittedly I did glance over what OP said about SB defending "a wicked % of his small blinds" but I'm still not convinced we should be assuming he has the entire file of 9xs or 8xs or the entire rank of Jxo or Txo

there's an additional wrinkle of irrelevance to the GTOW screenshots you posted: we have 80bb and are close to the tournament chip lead. This allays a ton of whatever ICM pressure may or may not be on us. Your solutions are if CO and BB have equal stacks. Makes a huge difference. SB is almost irrelevant when we're considering shipping it; he literally almost never calls, both in theory and in practice. If you start seeing people call-call with 77 for 80bb, I'm genuinely interested to hear about it. And we're in damn good shape when he does find a miracle call which probably happens no more than 3% of the time--we almost always have two clean overs

I don't rule out kq and qt, but we block both(16 combos total) and both have a bunch of outs vs aq(which has slim redraw). Including these hands at generous frequency probably doesnt sway the flop ev much. All 2 pairs and a portion of sets are well represented by this action, and should be included at high frequency. Praying he has kq, qt, or a flush draw that very well may blast us out on the turn, this doesn't seem like the pot that i want to risk half my stack on.

this unfortunately happened really early so no history w/V except he called abnormally wide from sb. i do think flush draws go for that nasty size check raise but i think a fold is in order- no need to play such a massive pot w/just top top- better spots to get paid in the great city of philadelphia

as far as the squeeze debate- i do think squeezing to 10ish bb was optimal and my raise pre was too small oop

yeah i think in this spot i just sigh pay them off. good point on them nearly never calling an 80bb shove too

If you're approaching the flop raise with "welp i have what seems like a strong hand in an absolute sense, ill call half my stack on flop and then mindlessly pay off future streets in the hopes that the opponent is doing something stupid without reads indicating such tendencies." I think its a very poor approach to poker strategy construction. Shrug and call is reserved for limit poker on the river. Even in limit poker, calling multiple streets with a bluffcatcher that has very little chance to win when behind, and significant chance of being outdrawn if ahead is going to lose money long term; You need alternate ev sources to justify continuing with more streets to play.

You blundered by betting the flop and blundered again calling the check raise.

I don't understand your reasoning. You get check raised to 350K and then on the club turn, you don't think a flush is a possibility? What exactly do you think the other player has? I think you are looking at a minimum JT/T9 and most likely something much stronger. Why can't the villain have KT? I think this deep, you need to let this go unless you have a specific read on this player that they will just go bananas with a flush draw.

I don't hate betting the flop on this board (I probably bet smaller) and those who are saying check it back are being results oriented (but I think it is the optimal play). OP, What would make you let this hand go?

I also think that you have to decided on the flop. You can't put in 1/2 your stack and then fold on a blank turn.

I would raise pre-flop to 100k. Its a sales thing (no discount).

We want CO to fold and we prefer SB to fold. Trying to convince SB to call is not a good plan. We miss the flop 67% of the time.

On the flop as played I would bet 61k. If we had made it 100k I would bet 80k on the flop.

This is an auto fold to the c/r with AQ on the flop. The 350k raise is never a semi-bluff (unless you know this player semi bluffs like this). There remains no fold equity for either of you. So as the 5th largest stack out of 75 players to risk it all here with TPTK is not good. If he was semi bluffing with a double draw he would have gone all in. Yes you now have lost 140k or 20% of your stack but move on.