Tuesday, November 09, 2010


Premier League IV

The Premier League is back, if you didn't catch last night's episode you can see it on Channel 4 On Demand (UK only). Now I look at that I can see it's been on for 5 weeks already, but that was online qualifiers and Party Pros playing off for the last two seats, and even my life's too short. Now, these are not the best 12 players in the world (even discounting qualifiers), to be fair I don't think anyone involved actually believes that and you can allow them some hype. But it is an interesting line-up with a lot of different styles of play.

So, before we get into hands, we have the commentary. Phil Hellmuth is first up on the mic and that means we are treated to an hour of self-aggrandisement and highly result-oriented analysis, to say the least. First of all Luke Schwartz five-bet shoves 64 on Yevgeniy Timoshenko, who can't call, and Phil is "standing up and saying Wow". Amusingly, he doesn't actually say "Wow", he says "I'm saying Wow". 10 minutes later Phil Laak 4-bet shoves 88 on Ian Frazer, who has Kings, and now Laak has just made the worst play ever, what a donk, etc etc. Naturally Frazer concurs with this analysis in his interview :).

The thing is, in Hellmuth's world, and Frazer's too I suspect, the concept of "range" hasn't yet come to light. When you think about the hand in terms of ranges, of course, Laak is making a fairly standard play against Frazer's three-betting range, in that he expects Frazer to fold some of it and he expects to have reasonable equity against Frazer's calling range, which might be something like [AK, TT], against which 88 is 31%. Now, you can argue that stacks are a bit too deep (Frazer has around 50bb I think), or that Frazer isn't folding much of his 3-bet range (impossible to tell without seeing more gameflow). Conversely, you can argue that Laak has ICM on his side, as he has Frazer well covered and the points system means that ICM is in play here (unlike most TV SNGs). Even so, you could convince me that the play is somewhat -EV because of stack sizes, but the point is that Hellmuth seems to have no idea why Laak made the play.

This comes up throughout, as Hellmuth repeatedly says things like "well JC isn't going to call here .. oh he does call". He's really quite bad at predicting what the players will do in various spots. And my point is that, even if, for the sake of argument, we agree that Phil is right and the other players are all donks, someone who fancies himself so much as a great hand reader (working out someone's hand from their betting action) ought to be a lot better at predicting what players are going to do (working out someone's betting action from their hand).

Well at least we were spared Phil's WSOP commentary where he was, by all accounts, even worse. Going back to the Schwartz-Timoshenko hand, I thought it illustrated an interesting point. Briefly, Vanessa Rousso limps, Timoshenko raises KJ to isolate in position, and Schwartz 3-bets 64o in the blinds. At this point Timoshenko has a decision. IIRC, Schwartz has made it 35k and stacks were about 280k at the start of the hand. With a stack of around 4-5 times the pot, shoving is definitely an option. You can weigh up the merits of shoving vs 4 bet-folding in various scenarios and it's probably quite close until you reach this one : if Schwartz has nothing but is prepared to bluff-jam over a 4-bet, then jamming >>>>>>>>>>>>> 4 bet-folding. That's probably not even enough >s. This kind of spot comes up on 2+2 every now and then and the super-confident regs will generally say "of course you shouldn't shove because 4-betting small achieves the same result if he has nothing ZOMG LDO you donks". I like to call this Devilfish syndrome. If your assumptions are correct then one play is somewhat better, however it could be massively worse if your assumptions are not correct. You can then take your choice between factoring in this uncertainty or calling your opponent a fucking idiot if he does what he shouldn't have.

Then we come on to ICM. The points structure for the heats is 16 to the winner, then 11-8-6-4-3-2-0. Based on these points alone, when play starts each player has a bubble factor of almost exactly 2. In simple terms, if you get all your chips in on hand 1, you need to be a 2-1 favourite to make it +EV. Just to illustrate how huge this is, suppose we're playing, on hand 1 it's passed round to me in the small blind, and I shove without looking. You, in the big blind, know I haven't looked. What hands can you call me with? Have a guess, it's in the footnote at the end [1].

ICM is fucking huge here and it's something that a lot of regs on the internet don't allow for correctly, never mind old-school live pros [2]. Consult Kill Everyone for a lot more detail on the subject. Anyway, in high bubble factor situations there are basically two ways you can approach a hand, either keep the pot small or be the first one to put all your chips in with good fold equity. The problem with the second approach is that you are relying on your opponent to be aware of the situation too ; if he just calls anyway then the two of you are spraying equity to the rest of the field like champagne off an F1 podium. I have decided that you're not generally going to get rich by trusting in opponents to do the right thing, and so I tend to try to keep the pot small in ICM-heavy spots unless I have a sizeable edge.

This comes up in the first elimination hand where Daniel Negreanu is "coolered" by Phil Laak. On the face of it, Negreanu's flippant "what could I do" interview seems fair enough ; he flops bottom two vs middle set about 50 blinds deep. Basically, Frazer raises, Negreanu calls with 65s and Laak calls behind with 66. It comes T65cc, Negreanu bets and now Laak raises, Frazer folds. Negreanu gurble gurbles a bit as per usual and then sticks it in, Laak calls LDO and Daniel gets "the bagel".

Once again we don't have the benefit of knowing gameflow, but Laak is generally on the T side of TAG and I think it's safe enough to say that he isn't bluff-raising this flop, but he could certainly have a big draw. Cutting short the Pokerstove analysis (DIY if you want to know), if we give Laak A-x of clubs (questionably IMO but let's be generous) as well as the sets and the combo draws then Negreanu's equity inches up to 45% which is just about good in chip terms. However, accounting for ICM, it's way short. Way way way short. Certainly I'd have been inclined to have a look at a turn, insta-check-fold a club and then make a decision on anything else. Daniel though is happy enough to say "cooler, what can I do" and walk off whistling. Anything rather than actually admit being, or even ask yourself if you were, wrong.

It's kind of funny that I'm advocating the old "take off a safe card" line compared to Daniel Negreanu's "I go arrr in", but this is what happens when ICM considerations are ignored in favour of the kind of "can't get pushed around, must be table captain" dick-measuring that we can expect from this line-up over the next few weeks. Even if you don't play poker (though why anyone who doesn't play poker would be reading this is beyond me), armchair psychologists can have lots of fun playing "spot the projection". Just as long as they get Hellmuth off the commentary ...

Update: The more I think about that last hand, the more I think Negreanu can actually fold the flop. I know, it sounds insane. However, if you posit that Laak is not bluffing ; has no worse made hand (eg an overpair) ; is never folding the flop or the turn and doesn't have A9cc-A2cc (except A5cc) then we see that Negreanu is 36% on the flop and only about 42% on a safe turn. Put it this way ; to make this even close to a call then you have to overturn those assumptions and say that Laak can be bluffing, or have an overpair, or some weaker flush draws. What it most definitely is not is a shrug-your-shoulders-what-a-cooler-what-could-I-do hand.

Another thing that occurred to me is that starting off with 150bbs and a bubble factor of 2 is a unique situation. You're just never that deep with a high BF in a normal final table situation. This could make for some more interesting spots in the shows to come.

[1] You can call with [88+, AKs]. Yes, AKo is a fold.

[2] To some extent the live pro attitude of "waiting for a better spot" is actually better than the internet style "I only play for first" in high bubble-factor spots, albeit for the wrong reason. The spot in consideration is often simply bad (in terms of $ EV) and so waiting for a better one doesn't even come into it. You could argue (and FredTitmus just did on ICM) argue that this is only semantics, but this falls down when the same live player folds for a better spot with 40 players left, because he just likes folding for a better spot.

Cracking post Andy - the bit about the 4 bet shove/smaller raise is something that I had never thought about before.

I must add an obligatory "lol Hellmuth". So here it is: lol Hellmuth.
Good to have you back.

To be fair, I would rather 5 bet shove with 64 against Jovial Gent than 4 bet shove with 88 against Frazer.

That is all.
I agree, but I wouldn't stand up and say WOW in the first case and then tell Laak to shut up because you're not allowed to talk during a hand when you get it in bad in the second!

Maybe he didn't stand up and say "Wow". Maybe he just *said* "I'm standing up saying "Wow"".

Who knows what relation Hellmuth's contant first-person narrative has to what's he's actually doing.
I have been really glad after reading this blog as the knowledge which has been given via this blog is simply tremendous. I would congratulate and appreciate the blogger for doing this much hard work.
Play Bingo Games Online, Bingo Free Play, Play Bingo Game Online.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?