Harrington on Holdem 3: The Workbook REVIEW

[ChrisV]

I just finished reading the new Harrington On Holdem book. Like the previous volumes, it’s generally good, with a lot of useful advice and analysis. There’s even some (pretty basic) discussion of SNG bubble play and the ICM this time. It definitely deserves a place on your shelf. That said, there are a number of errors scattered through the book. Some of the following are my opinion and some of them are incontrovertible (especially problems 46/47).


Problem 3: Early in a SNG, we raise on the button with 9 [img]/images/graemlins/spade.gif[/img] 8 [img]/images/graemlins/spade.gif[/img], the SB calls and the flop comes K [img]/images/graemlins/heart.gif[/img] 5 [img]/images/graemlins/diamond.gif[/img] 4 [img]/images/graemlins/spade.gif[/img]. SB check calls our continuation bet. Turn is the 8 [img]/images/graemlins/club.gif[/img] and SB checks. Harrington wants to check here. I much prefer betting again. Our hand is vulnerable and if SB was planning to checkraise, it’s probably with a hand we’re drawing dead against, and if we check here we are probably going to have to call a river bet. I’m not putting him on a king just because he check called the flop.

Problem 9: We’re playing as Hellmuth, in an MTT. Blinds 100/200, our stack 15K. 3 limpers to us in cutoff and we limp with 33. Button (“no particular read”, 18.5K) makes it 1,100 and it folds back to us and we loosely but reasonably call. Flop is K76 rainbow and when we check our opponent checks behind. Harrington now bizarrely claims that this check probably means a weak hand. Against typical players the check means at least a good king the vast majority of the time. Turn pairs the 6 and now Harrington and Hellmuth both advocate betting. Turns out the opponent has AK. What a shock.

Problem 12: SNG, 1500 start stacks, 25/50 blinds, the cutoff (T600) open limps and we have K9o on the button. Harrington limps, which I don’t like much but whatever. Point is though, in the intro to the problem we’re told that cutoff has played “good, solid poker”. At the end of this hand it turns out he limped 86o in the cutoff for 1/12th of his stack. His postflop play is questionable as well. At the start of Problem 13 (same setting) we’re again told this player is “tight, solid and dangerous”.

Problem 13: SNG, blinds 25/50, stacks 1500ish. One MP limper to you and you’re supposed to limp in the CO with ATo, which I can live with. Only BB comes, and flop is A [img]/images/graemlins/heart.gif[/img] J [img]/images/graemlins/heart.gif[/img] 5 [img]/images/graemlins/diamond.gif[/img]. BB checks, limper bets 100, and we’re supposed to just call because one of them might have us beat and you can’t be too careful. No thanks, I’m raising. If I can’t even play my hand strongly on a flop like this then I’d rather not limp it in the first place.

Problem 14: SNG, blinds 50/100, 7 handed, you 2,260 UTG. Blinds slightly cover you, other stacks range between 1,240 and 1,880. We’re supposed to raise ATo UTG here. Firstly no thanks, secondly how is this a better raising spot than the previous hand? BB calls, flop is 975 rainbow, it goes check/check, turn is another 5 and BB checks again. Now you’re supposed to check again because of some convoluted reasoning that makes no real sense. I’m a simple minded type of guy and often I like to bet when I think I have the best hand. Of course versus a very aggressive player checking might be right but BB is not described as such.

Problem 18: Live MTT, Blinds 1500/3000, Antes 200 (total 6300 in pot). Pro on button moves in for 10K, you have 30K in BB and 65o. Pro has a $5K cash bounty on his head. While Harrington gets the answer right (call), my jaw dropped when he simply added the $5K (cash) to the $T pot when calculating pot odds. HUH??? He also, while working out an opponent range, informs us that the correct play for the pro on the button here is to move in with any two cards, even though he’s in the process of proving that we’re going to call him with 100%. This doesn’t compute. Pushing with 72o can’t possibly be right if we’re up against 100% in the blind. How he doesn’t see this I’m not sure.

Problem 36: MTT final table, Blinds 3000/6000, antes 300. We are BB with 185,000 and 88. Folds to button (275,000, “tight and aggressive”) who makes it 22,000. The problem is to figure out the size of our reraise. Harrington wants to make it 80,000, which I have no real problem with, though it might leave you in an ugly spot on the flop. But he gives less credit to moving in than to making it 44,000 or 56,000, both awful raises which price in your opponent who has position on you and probably two high cards. He can then pick you off at leisure on the flop. This because a smaller raise “disguises your hand. After all, you wouldn’t push allin with a big pair”. This is backwards. Against perceptive players I would push allin with a big pair. It’s true I will make somewhat less money with big pairs this way, but I am dealt big pairs substantially less than other hands and I will do a lot better overall locking up the 50K pot and not leaving myself in bad situations postflop out of position. (If this is the “fear of flopping” thing that Harrington talks about in the intro, count me in).

Problem 42: This is part of the SNG bubble series. Blinds 100/200, 3 stacks including you have 1000 and the big stack is UTG. He pushes (range 100%) and it folds to you in the BB. Harrington says call only AA-88 and AKs, which I assume is what the ICM says to do. I would call a little looser here because otherwise my opponents can simply fold-war me into the ground. I will be forced to go allin at some point so if I have a decent hand I simply have to eat the small loss here.

Problem 46, 47: We come now to the real howler of the book. We’re heads up at the end of an SNG, limiting stack is 4,500, blinds 200/400, antes 25. The only difference between the two problems is whether we or the opponent have the limiting stack of 4,500, Harrington correctly making the point that it makes no difference. We’re in the BB and SB pushes. Lest I be accused of unfairness, I’ll quote the description in full here: “Your opponent has played a solid and perceptive game throughout the tournament. You’ve been playing heads-up for about 20 hands. So far the pots have been small.”. The problem is to decide which of the following to fold to the push: QQ, AKo, 99, ATs.

I giggled and flicked to the next page, confident the answer would be none of them. I stared aghast at the page as Harrington told me the correct answer was to FOLD 99 AND ATs. The justification? Well, our opponent is probably tight, since he is “solid” and the “pots have been small”. If he’s very tight here, naturally he only pushes with TT+, AQ+. If he’s only moderately tight though, he’ll be shoving a far looser range: 77+, AT+, KQ.

Folks, I am not making this up. The whole problem reads like it’s from another galaxy. It’s an embarrassment to 2+2 that it ever made it into print. I can only assume that Harrington has never played a SNG in his life.

Quite apart from the sheer insane tightness of the pushing ranges, I’m expected to believe that my opponent’s standard play with AA or KK is to push? Even though he apparently never pushes normally? Give me a break.


OK so my tone has been somewhat harsh in this review. Honestly I think this is a good book, I learnt things from it and would recommend it. I think it could have benefitted from a proof read from a good SNG player, especially the insane Problem 46. And I just wanted to highlight some of the questionable choices in the book, in case people out there are taking Harrington’s word as gospel.