Thursday, December 22, 2011


Example Of A Leak

If you're hoping to hear about technical things I'm changing to improve my game, so you can implement them into your own, I'm sorry to disappoint. The fact is when ElRupert, for example, tweets a link to my blog post (which I do appreciate, thanks mate) and then next time I play he's on three of my tables, it would be cutting my own throat to talk about a lot of the changes I have made.

The reason balance is not that much of an issue in online MTTs is that you just don't build up that much of a history with other regs. That means it's difficult for them to spot patterns in your play (and vice versa of course), especially when you're only taking particular actions with a small subset of your hands. If I blab about them on here and people are reading it though, that cuts out the process of picking them up through piecemeal observation entirely!

I can give you an example of one leak though, as the solution is (unusually) making my play more balanced rather than less, and it comes up quite rarely anyway. Say you're at a final table with a high bubble factor (in English, you don't want to get knocked out right now) and you pick up a hand like AJ or 88 with 18 blinds or so. My original idea was that if I raise small with these hands then I will have odds, even allowing for bubble factor, to call a jam reraise. But AJ, 88 etc is not a sufficient favourite over the reraising range to want to get it in, so, assuming villains will call a jam less often than they reraise (generally a reasonable assumption), I'm better off jamming.

On the surface it sounds plausible, or it did to me, but it's sloppy thinking and it's wrong. DUCY? Quite probably. The problem is that any better hand than AJ is going to get it in with you whatever you do. The fact that AJ isn't in good enough shape against the whole reraising range isn't the issue, what you need to consider is how it does against the extra hands that reraise but don't call a jam. Against those extra hands you're in very good shape because, depending exactly on ranges, there are a bunch of KJ, A9, JTs, 98s and so on that you're way ahead of. Even allowing for 66-44 etc that you would rather force out, you're in good enough shape against the "extra" range that you're happy to get it in against that range allowing for bubble factors.

So the solution is not to jam these "middle" hands but just raise-call them instead, which means I'm raise-deciding with everything I play. It's not a big leak because it's such a specific scenario and both raiser and caller have to have a narrow subset of hands to bring it into play, so not that big a deal. But it illustrates quite nicely some of the ways in which other leaks have developed, particularly :

- Changing something that was working perfectly well before through trying to be too clever
- Fixating on being unexploitable and so
- missing out on opportunities to exploit others
- worrying too much about being exploited by people who mostly aren't capable of it

It's just an example as I say, but it's the kind of wrong turning that I've taken in the last year and it's not the only one by any means!

Addendum : By the way, it's unrelated but I was flicking through the old Full Tilt Tournament Strategy book and while some of it's pretty lol, re-reading the Gavin Smith section I was really impressed. A lot of what he said is standard online aggressive play now and he was way ahead of his time really.

Tuesday, December 20, 2011


Bounce (2)

The real message Matthew Syed is trying to put across in Bounce [1], as I understand it, is how we under-rate the benefits of focused, motivated practice. Hard work. Grind. I have come to realise that for about a year, from September last year to this, I was trying to cruise in poker, making no effort to improve and blaming lack of results on bad luck or the game getting tougher.

I was (and still am if I slip back) in danger of becoming like one of the live pros who were overtaken by the "internet generation". I always felt that they had no one to blame but themselves. Young players came in and devoured the game, they lived for it, players like Durr and Galfond, or Mercier and Elky in tournaments. Starting from scratch they were able to overtake the "old school" remarkably quickly, partly because online poker allows you to play so many more hands per hour, per day, per month.

The thing is though, what was stopping the old school from working just as hard, starting off a long way forward of scratch? Taking the experience they had and building on it with the same focus and determination? Taking what they knew and playing 200 hands an hour online from that base? Complacency and laziness. In other words, nothing. There was no reason why any of them couldn't do what Ivey did, but it was easier to cash the sponsorship cheque and cruise. It's also very easy to say "Oh, Ivey's some kind of supernatural genius, no one else could do that." I wonder who else tried?

That may sound overly critical but I apply it to myself over the last year too. Now I'm making an effort to find and fix leaks in my own game and I've been shocked how many I have found (and am still finding). The problem is it is very hard, in tournaments, to gauge progress because actual $ results are so random. I am trying to find ways to measure how effective my play is outside of just the bottom line. It's not easy but HEM is a big help and there are some tools in there that you can use if you're very careful with them - all-in adjusted EV for example.

As I mentioned in a thread on 2+2 a few days ago, this has also helped with my motivation. It's a positive feedback loop. I have new lines to try and new things to remember - it's very easy to think of something you should be doing, do it for a couple of days, but even if it goes well you forget and move on to the next new trick. I keep a record of the points I should be remembering and mistakes I shouldn't be making, as recommended by Jared Tendler in his book.

As for the bottom line it has been going OK but nothing spectacular. Then again, that's partly due to two or three bad mistakes at final tables, spots where I just did what I always do and what I thought everyone else did, but when I analysed them later I could see how they were wrong. And those are the mistakes that can really cost you $$$. If I can fix those then I'm confident that I can go on an uptick over the next couple of months.

[1] In the first section anyway, the second and third are also interesting but have different themes.

Saturday, December 10, 2011


Bounce (1)

First bounce, perhaps I should say. I've just finished the book Bounce by Matthew Syed and I thoroughly recommend that you read this book. Don't even finish reading this until you do it. Download it now from the link (install Kindle for PC if necessary, it's great).

I picked this up on a recommendation from everyone's favourite Twittering footballer Joey Barton. The central thesis of the book is not new, it's the basic idea that Malcolm Gladwell put forward in Outliers. Bounce is a much better book though IMO, much less anecdotal and has the advantage of being written by someone who, as Britain's top table-tennis player for several years, freely admits that he used to believe that this was largely due to his own natural "talent" rather than the circumstances of his upbringing combined with extraordinary amounts of practice.

In some ways this is only tangentially related to poker but there's not really anywhere else I can put this now so here is going to have to do :). It does, however, have some relevance to a snippet on a Late Night Poker episode I caught up with today. Players were asked (for some reason) what sporting figures they would like to be/have been. They rattled off more or less what you would expect - Tiger Woods [1], Roger Federer, Michael Jordan.

I know it was just a flip soundbite question but I wonder if the answer "well I'm not sure I'd like to be a world-class sportsman at all" actually occurred to anyone as a possibility. Because you can be assured that those three people have devoted their entire lives to their respective sports. How they will cope with retirement remains to be seen.

Thing is though, I think we really know whether the players actually want to be Tiger Woods, for example. Because they have their own field of excellence in which they already have a platform and an opportunity to excel. Poker. So are they practising with the intensity of Tiger Woods? Every day, like Michael Jordan? Desperate to improve and build on every setback, like Roger Federer? Well, we don't know. But I reckon they probably aren't.

There's one poker player I can think of who might have done. He reputedly played 18 hours a day in Atlantic City for 2 years straight when starting up. He's one of the few "live" players who saw online poker as a tremendous opportunity to learn and improve, rather than just exploit by association. And funnily enough, albeit for completely the wrong reasons, he is sometimes known as the Tiger Woods of poker...

[1] Tiger Woods was chosen by Vanessa Selbst on the grounds that "golf doesn't look like you have to train for it very much" - LOL. That depends how good you want to be. To be Tiger Woods, you have to train incredibly fucking hard for your entire life, is all.

Tuesday, December 06, 2011


Late Night Poker (Finish)

I caught up with the Late Night Poker final today - look away now if you haven't seen it!

In the event, it turned into a rather disappointing crapshoot. The combination of an 8-handed final and only 5 being paid anything at all was always going to tend to produce cagey play in the middle, which leads to blind pressure at the end if the structure isn't set very carefully, and that's what happened.

From what we saw, I think the last 3 players were probably the 3 who played best on the day, and from there it was all about the flips. Sam Holden won 2/2 FTW, fair play to him. I'm sure Luke Schwartz is the best player in the line-up with any depth of stack, but he may have failed to adjust to playing 3-handed with 20bb stacks. Qualifier Rob Okell played very well from what I saw and would have been a big favourite to win if TT > KQ, which it didn't, so gg.

If people really think that calling it "Late Night Poker" makes it more significant than any other TV tournament then good luck to both the people thinking that and the people who made them think it :). That seemed to be the impression they were trying to create.

Anyway, it wouldn't be a TV blog from me without picking someone's play apart in hindsight, so let's do that. Roberto Romanello made what I think is a mistake on his exit hand which (result oriented klaxon) cost him dearly. In truth I don't blame Roberto for what he did, I'd have done exactly the same 3 months ago. From a stack of 13-14 bb he made a small open raise with AQo. The idea is, presumably, to induce some shoves (which you call LDO) from weaker hands and provide some balance for the times you want to raise-fold off that stack.

Balance though, as I said last week, can GIFAFI [1] in a TV single table. This is something I was doing for a couple of months online, but I found that on empirical analysis of a few thousand hands, it just wasn't as profitable as jamming. Inducing shoves from a few hands against which you're maybe 60% overall isn't that big a gain, especially in an ICM heavy spot like this one. The real downside of the play is that you give the BB in particular a very cheap option to call pre and stick it in if he flops anything, which is exactly what Luke does with KJ. Luke chooses to bet out on the J22 flop, where I would have check-raised, but it comes to much the same thing.

Turning it round, this is something you can easily do in the Big Blind when someone minraises and effective stacks are small, just call and checkraise jam any pair or draw. It's even better with antes in play. That's all for now, though I will probably talk soon about a book or two I read recently.

Addendum : @Standaman60 on Twitter mentioned the hand where Simon Trumper raise-calls a 3-bet from about 20bb with AK and check-folds a Jack high flop. One thing I'll say about that hand is that if you ask Simon about it, I'm very sure you'll hear the words "I put him on a middle pair". So this line is somewhat more profitable if your "read" is correct but, as I've said before, if it's wrong you could be burning chips. There's no way this line is beneficial against another AK, and against AQ or KQ it's just horrendous.

[1] Although it probably wouldn't FI, being so balanced and everything.

Friday, December 02, 2011


Protect Your Stack, Not Your Hand

I'm finding watching Late Night Poker a little easier to watch if I fast forward through all the dwell-ups. Quicker too. Especially the ones when we know you're going to fold. Seriously. It's better for your image to insta-fold anyway when you're caught bluffing, but I suppose then you don't get so much camera time.

Anyhoo, a moderately interesting hand came up in the last heat. I can't remember stacks exactly, but it doesn't matter for the point I want to make. James Bord opens K9o in the cut-off (fine) and Ram Vaswani flat calls Jacks (probably from around 25bb) in the small blind. I like this play a lot, you can induce a squeeze behind you (it didn't in this case but never mind) or check-raise a whole bunch of flops to leave your opponent with a nasty guess to make. Shoving all in pre lets an aggressive player off the hook here IMO, although 3-betting small to call a 4-bet has its merits. You might argue about balance but I feel balance is an over-rated concept in tournaments at the best of times, let alone in a one-off hundred hand TV single table. And anyway, there is a way I balance it, but I'm not telling you :).

So the flop comes King high, oh well, nothing comes for free and every play has its downside. Ram check-calls James' c-bet which is all standard. Then the turn comes a Queen, so its KxxQ with no flush draw that I remember, if there is it's back door. James now shoves something that looks like about 1.5x the pot [1]. This is a play that I really don't like. A lot of people will say "he's probably got the best hand" and so he can "protect his hand" with the allin bet. Yes, but...

Probably having the best hand is not in itself a reason to bet. If Ram is behind, he has one possible hand with 7 outs (AJ) and a few with 5 outs (AQ, second or third pair) - and he could easily fold some of those on the flop. The rest are 3 outs (worse King) or 2 (underpair like he has). Let's be generous and say 5 on average. James is shoving 1.5x the pot to "protect" the 5/44 of the pot that Ram has equity for if he's behind. Meanwhile if Ram is actually ahead, now James has 5 outs, or 3, or none for the whole lot.

The main reasons to bet are to make a better hand fold or a worse hand call. Now OK, you'd have to be pretty good to call with KT and fold K8 in Ram's seat, but that's not really the issue. If you were to either call both of those hands or fold both of them, then that would be pretty much a wash. With any significantly better or worse hands, Ram is not going to make the wrong decision. It's not the only time I've seen this in the episodes I've watched (so I don't mean to single James out) and I can't blame James and Vicky on commentary for not really explaining this (level one is best for the majority of viewers).

What really rang a bell for me though is that I'm re-reading Gus Hansen's book on Kindle and he does almost exactly the same thing by over-shoving I think QT on a Q high turn. His own justification is the line I use above, that you'd have to be a great player to call QJ and fold Q9, but that (as I suspect Gus knows very well) is not the point. I haven't even mentioned the other possible benefits of checking, viz. that your opponent (Ram in this case) might bluff the river. You are, I should say, usually calling the river after pot controlling in this kind of spot. If the villain bluffs the river then you make more money when you're good as well as not losing the lot when you're behind! And finally (unlikely but still a freeroll) you might be behind and improve to win on the river.

My own summary of this is that any time I catch myself thinking about betting to "protect my hand" in a tournament, I try to think a bit harder. Protect your STACK. That's what's important. Trying to win every pot is going to hurt you if it means that the pots you do lose are much bigger.

[1] I stand corrected if Ram has less than a pot bet left and so effective stacks aren't that high, but that's not how it looks.

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