SuicideSpree

Fair enough, maybe intellectually dishonest is taking a bit far, especially since I do agree with the point in this example, although the % value of that 22k isn't negligible and a few % points will matter quite often in PLO.

I think a better term would have been sloppy which I stand by. It still just felt very jarring to see one number rounded down and another one up, both in a direction that would emphasize your point.

I do agree it doesn't really become intellectually dishonest, but I wonder why given how little effort it would have taken to just get them right.

I've edited my comment to remove some of the overly judgmental language.

It seems quite sloppy to round up and round down numbers to suit your point.

It just cannot be that -288k is rounded up to -300k, +310k is rounded down to 300k and then you blithely go on making your point with exact EV calculations.

I understand the baseline point, but surely you can be accurate with numbers in your favour to make it?

I know I'm asking a few months after the video is posted, but I'm very interested in one issue, so I'm sharing a huge number of questions/thoughts in the context of what one would think is the rather straightforward hand at 7:50 or so. To me it seems very strange and possibly related to group-think, that we use software to analyze a spot, the software provides an answer that none of the players actually use, and in fact would get you laughed at and un-backed in many cases, and yet we take parts of the answers provided by ICM software as gospel truth, if they say fold, even within a margin of -.5BB. This seems actually mind-boggling to me. I have some questions/observations.

1) In the hand with 77 vs callmuckty, if you were in Watnlos' place, would you shove K5s there?

It's well above what the sim showed as a profitable shove. I'm quite deliberate about choosing this as a question, rather than say "would you shove 54s against this open" so that it's not possible to answer that it's more profitable as a flat etc. The main point of my question here is that I think you (or vast majority of MTT pros) internalize, don't truly believe ICM recommendations, but largely select ones that conform to pre-existing biases (not wanting to commit stack without nutted hand is the one I have in mind here).

2) Follow-up Number 1. If you are in Watnlos spot, you aren't shoving K5s or even K8s. You assert at 7:53 "Small blind shoving which he can't actually do". Why can't he? The math impeccably demonstrates that he can. I imagine both you and he would fold a lot of hands that you can shove according to ICM, and I'm assuming you wouldn't do so, even against tourney regs that understand ICM (as opposed to a cash guy who doesn't value it as much). Is there some cognitive bias preventing this play? Is it that you don't want to look stupid?

If it's +EV to shove, is the ICM wrong or is it that absolutely no pro does this, solely for the reason that they are all afraid of looking stupid? Or does everyone without exception think they have such a big edge that they can fold hands even 3 or 4 levels higher than what they can shove with +EV?

If it's that people don't believe that ICM is actually providing the right answer with some of these recommendations, then why present other recommendations of the same software as gospel?

Your comment is actually a little ignorant and massively disrespectful of MTT players. I have several observations in response.

1) This isn't easy. I have 10 years of experience playing MTTs and even I don't do short-stack perfectly. It's not like we are all stupid monkeys and cash game guys are geniuses with 190 IQ who can work out in two days what we've done over long periods. This guy is playing a $2k tourney and showing skill worse than many$15 regs in a vital department of the game. If playing in $2k and above won't motivate him to study this, I really doubt comments on a forum will.

2) Tournaments have guarantees. Top tourney players put up their buyins without even being sure who the field is. None of the SHR MTT guys is worried about cash guys. Best MTT players won't be scared of playing cash and will do relatively better at cash than cash guys will at MTT. Anyone who has played $2k online tourneys or above (myself included), knows perfectly well how to play deep-stack cash. You can't play big tourneys well without knowing cash well. Because big tourneys are well structured, cash guys can play tourneys and make big stacks, which they usually don't handle perfectly deeper because their game, study and skill-set isn't based on that.

3) No-one has a big edge post-flop 16bb deep. That's totally absurd. Any player who is even semi-decent will not make enough post-flop errors to give anyone even 2BB/100 edge at this stack depth, and his preflop errors are much bigger than this.

4) His post-flop expertise is at 100 BB or deeper mostly. I very much guys who specialize in cash and play very little tourneys would play better than tourney regs at shallow stack depths.

5) Studying 100 BB poker is actually much more doable than studying MTTs a genuinely high level. It requires much fewer situations to be learnt. Just on a solver-based approach, an MTT guy has to plug in way more hands because the different stack sizes drastically affect ranges and solutions. Being good at MTTs requires you to know cash on top of all the skills cash guys tend to lack.

6) Is it easy to learn short-stack to basic competence or minimize a few glaring errors by looking at Snapshove? To an extent, yes, and you would think so, but when pros playing $2k online can't or don't do this, it's just staggering to claim that this is just so easy and effortless.

7) I not only watch, I've occasionally played in high-stakes (albeit ultra-soft private games) cash, the vast majority of play and study is at 100bb or deeper.

Oh yes, I agree with the first point, as I noted Elias isn't over-folding, but it's hardly optimal what he is doing. Imagine he sees me play some hand at 6max NL 1k where I stack of 150bb pre with AJo, what will he think of my game? That's actually how sub-optimal his overall strategy is here, and I definitely won't be making such an enormous error in deep-stack poker, because MTT guys must learn deep-stack play, but the reverse isn't actually true for cash guys.

Second, I said in my comment, Isildur's sizing is wrong if he's never getting folds here, but definitely, when BB shoves only 11% of hands, open-raising hands like this seems good. Maybe he needs to go full 3x or maybe even more, but it will be very good if his opponent is so reluctant to shove.

Also, there's no denying that a high-stakes cash guy will usually be better post-flop than most MTT regs. But that's not always true. In being an MTT pro for 10 years, I have to learn to play post-flop at 150bb deep also, albeit not as well as him. But converse is not true. Some guy who crushes 25/50 never ever has to play 25bb deep NLHE, or deal with always varying and sometimes extra-short stacks.

Especially nowadays, I think MTT is much more complex, not less complex than cash. A cash guy can just comfortably input his very constant ranges into PIO and all there is not much difference between 100bb deep and 150bb or 175bb deep. But there are enormous differences between even 40bb and 60bb deep. I am almost 100% sure, even a top cash player like Elias won't be better at 40bb deep post-flop play than top tourney regulars.

MTT and cash are fundamentally very different skills, but the thing with high-stakes cash is variance is lower, and there is much more potential for predatory behaviour so only elite players can actually survive at high-stakes cash, and that's why every reg at NL 2k or 5k is superb. It doesn't mean they are better at MTT post-flop spots than the guys who are playing highroller MTTs.

This video seems to me to be an object lesson in why cash-game guys struggle in MTTs. I will make some comments because I am actually very interested in preflop. Some of the range construction and thought process seemed to be from an alternate dimension, and frankly I think there's a basic non-comprehension of how many hands can or should profitably get all-in preflop with only 2 players, and an effective stack of 18bb.

Elias mentions a lot of hands as suffering vs his all-ins, that I would consider fist pump calls, and/or coolers or mandatory stack-offs. That's probably because 11% is a ridiculously small % to shove vs an SB open at this stack depth. Albeit he is protecting his weakness by calling 100% and not over-folding to this raise

3rd point, vs a nit-fish like this (and Elias is a fish at this stack depth), I think Isildur is right to have 93s in a raising range (although his sizing is very wrong if Hero never folds). 93s will benefit a lot from generating preflop folds and of course it can easily fold to all-in. It also seems absolutely insane to me, for a pro to see that his opponent has 93s in his open range and then reshove only 11% of hands. In fact, against a range that is so wide it contains 93s, it will probably be +EV to shove well in excess of 50% hands.

I would definitely have a decent open-shoving range against a player like this who is clearly very reluctant to get all-in preflop and then limp a pretty polar range.

This was quite interesting.

I'm a high-stakes reg catching up with some theory. I have an extremely unorthodox, very very aggressive and gambly style, so I make it a point to keep up with what most regs are doing.

My personal view is people (and by this I mean even people like yourself who know the theory) way under-defend these spots. I did a quick database study of my own play to check if I was under-valuing playability. You actually had me worried I was missing something. I got 561 hands in my sample, of defending the junkiest hands, the only difference from your parameters was that I left all positions in, because I defend 100% to min-raises even from UTG. I expect there's a lot of variance, but I got -100 BB/100, so I'm basically doing OK.

Posted this as food for thought, since overwhelming consensus seems to be that defending junk hands is bad (a consensus I totally disagree with).

I really have to point this out since I DO play DotA.

What the machines excelled at wasn't actual DotA, but a hugely stripped down, exponentially simplified version of 1 vs 1 DotA which is itself much much simpler than real DotA. As such it was much easier than chess or GO. They are deliberately conflating two things: A) DOTA is much more complicated than chess, poker, GO: Absolutely true and B) The Machine won at DOTA, absolutely false. What the machine did was about as impressive as a bot beating a human at a 25 BB push fold game.

If Mighty can really do this with impunity (because he has a big stack) then it's worth so much that others should be willing to take -$EV but +CEV spots in order to become the big stack.

There are flaws with treating ICM model as gospel. Notably that every hand is super +EV for the big stack when people play like this. If being the big stack is worth so much CEV and $EV it's worth taking -EV gambles in order to get that stack. With 8th and 9th paying the same and 7th being a tiny increment, I think everyone hero included was playing way too tight.
Especially with hero's huge skill advantage (imagine this FT if he and Mighty had switched places), he ought to have been taking +CEV spots to get a stack rather than trying to ladder up for $9000 more and putting himself in a situation where he is unable to leverage his extra skill, just as much at the mercy of the cards as if he had pushed in some spots, and additionally trying to out-cockroach a guy like RescueGroup which simply decimates hero's stack.

When everyone is desperate to sacrifice CEV to ladder up, it stands to reason that whoever is fine with gambling (at this table, nobody) just hoovers up the chips very often. It says a lot that at a table of good players, with this much money at stake, we are discussing clicking whether blindly clicking all-in like a clown is a good idea. The fact that this is a discussion underlines that this is because the responses to clownish shoving are bad in a way that makes it not so clownish.