The final principle

Sounds like a good movie title (or perhaps, a good title for a bad movie) doesn't it? Anyhow, I was all set to regale you with tales of my SNG exploits last night (two table turbos in deference to the fact that I was trying to keep my eyes on the cricket as well), but when I reviewed the hand records they were a real yawnfest. Just one question from there before moving on:

On the bubble (5 remaining, payoff 4:3:2:1:0), blinds 200/400 (25 or 50 ante, I can never remember), you (2K) are UTG with K9o. Remaining stacks are all roughly 4K-5K, except for the BB who has about 10K. The medium stacks have been (probably incorrectly) soft-pedalling, waiting for you to bust. Do you push?

So, on to the final principle from Lessons in Play and its purported relevance to poker. This one is called the “Don't burn your bridges” principle. In the theoretical setting, it is again intended as a tool to simplify analysis. Basically it says that, between two moves, if one leaves a strictly larger set of future options available, then you don't need to consider the other one as part of a winning strategy.

Of course, we don't expect such hard and fast rules to apply at poker. But consider: you are in position, possibly with the best hand, but, regardless, with a strong draw; you can check or bet, but if you bet, and are check-raised you'll have to fold. Here, based on the principle above, it's often correct to check.

Tinfoil hat time

I had a new podcast of Fighting Talk to listen to last night, so I was able to play a regular single table SNG (if you don't think that's a non sequitur then you spend far too much time reading this blog).

I saw my first ace (don't even talk to me about pairs, or suited connectors) when the blinds were at 75/150, and my stack was down to about 900. There were six players left at the table. Flushed with excitement, I pushed A9o from late position, got called by KQ and doubled up. Then it was back to card dead city for a while.

Fortunately, on the bubble, I got a little rush of reasonable hands and connecting flops, with the result that when we reached the money, I was actually chip leader (which just goes to show, that the primary requirement for success in non-turbo SNG's is patience -- it helped that it was a really good episode of Fighting Talk, which, at this point, was winding down). I lost a couple of 60/40's, so when we reached heads up I was actually slightly behind.

I thought I had been card dead before? No worries, the dealer proved to me it could get worse (I don't know how many times I saw J4 off). Fortunately, my opponent was very passive, so I was usually able to escape for, at worst, the loss of the BB. He invariably limped from the SB. Unfortunately, he caught on to my card deadness and started calling everything. I had to resort to running big bluffs with hands like 62o on an AQ9 flop.

But I was confident ... I still had about 40% of the chips in play, and felt that with any big hand I'd likely be able to double up. After 50 hands heads up, I got AA in the big blind.

And he folded preflop.

One hand tied

The next principle from combinatorial game theory that I'd like to look at is the “one hand tied” principle, whose name comes of course from the schoolyard taunt:

I could beat you with one hand tied behind my back.

In combinatorial game theory (I'm really testing my aversion to acronyms here), we use this principle to prove that one player has a winning strategy. By restricting that player's available moves, if we can show that she still has a winning strategy, then she certainly must have one in the full game. The point is that the restriction may well simplify the analysis (and keep in mind that all we care about in such an argument is who wins -- not by how much). So, does this one have any relevance to poker?

It's a bit of a stretch, but I'd claim that there are some games (ultra low limit cash games, low buy in MTT's, but most especially sit and go tournaments at odd times), which are so good, that it's actually a mistake to do anything but “play with one hand tied”. By simplifying your game, you can avoid out thinking yourself, avoid having yourself marked as a tricky, or just as bad skilful, player, and generally stay under the radar. It's a matter of establishing and maintaining an image. When the time comes, you might make a “lucky” move to take someone out, but ideally the remaining competitors will still just think you're one of the crowd.

Give them enough rope

I promised, implicitly at least, to provide some illustrations of how concepts from combinatorial game theory (which, need I remind you, you can learn from the marvelous new book, Lessons in Play) can be related to poker, despite the fact that the description of combinatorial games (two player, perfect information, no chance elements) scores 0/3 as a description of most common poker games.

Of course these illustrations would work perfectly well, and in much the same way, if the source were the I Ching, Tarot cards, or your favourite religious tract. The benefit is largely in thinking about it -- I'd include advice from the magic eight ball, except it doesn't allow for much in the way of reflection. Anyhow, back to the story.

I'm going to cheat (another way in which poker is unlike combinatorial games) and start with a principle which is not about the theory of combinatorial games, but about their play.

Give them enough rope (to hang themselves with)

In combinatorial games, this is used to describe how to play when you, because you've read the book and become a combinatorial games guru, recognise that, somehow, you manage to find yourself in a losing position. Chess players know it as “playing for complication” and it's a particularly effective tactic there, especially if your opponent is under time pressure.

The basic idea, is that if you are losing, you need to make it as easy as possible for your opponent to make a critical mistake. So, you should play in a way that leaves the maximum number of possibilities open to your opponent -- or, at the very least, which presents a number of plausible but losing options. As I said, in competitive chess, this is a well recognised ploy. Before moving on to poker, a small digression (gosh, I really am settling back into academic mode).

The oriental game of go is, arguably, the most complex pure strategy game (a.k.a. combinatorial game) that is played competitively. It has a rich literary and philosophical tradition, which includes a significant amount of game etiquette. Central to this etiquette is the idea of showing respect to your opponent. For instance, the first move is traditionally played in the corner of the board closest to your opponent's right hand (where, should he choose to respond locally, it will be easiest for him to play). I never did sort out if you were supposed to adjust to left handed opponents! Anyhow, the enough rope principle is completely antithetical to this etiquette -- since it implies that you believe that your understanding of the board position is superior to that of your opponent (yes, I know, this is precisely what playing the game is supposed to prove -- but you're not supposed to be obvious about it!) The “proper” play in go, when you recognise that your position is lost, is to resign.

So, what has it got to do with hold 'em? A fairly common situation heads up on the river, is to suspect that you have the second best hand, but that your opponent's hand is not overly strong (for example, you have second pair, you suspect that your opponent has top pair). If you are in position and your opponent checks on the river, then a check by you can never make life difficult for your opponent. But, if a bet by you might plausibly represent a better hand (e.g. the river has completed a flush draw, present since the flop, or filled some sensible straights), then a bet might be just the rope that your opponent needs.

Of course, this is probably not the most common interpretation of the phrase “give them some rope” in poker. More obviously, it can be applied to the slow play. Here, with great strength, you represent weakness in an attempt to induce bets, or calls from your opponents. But really, this is just another view of the same principle: you wish to ensure that your opponents are faced with difficult decisions, not easy ones.

An opportunity not to be missed.

First of all, apologies for yet another episode of bloggus interruptus. All the usual excuses:

• uninspired (when did that stop me before?);
  
• too busy at work (As I'm sure you're aware, when an academic is given a workload that would be considered "normal" by 90% of the population, he melts, Wicked Witch of the West-like, into a puddle on the floor, with only a pair of eyes staring puppy-dog like remaining)
• taking a beating at the tables (What? You wanted to hear how I lost a buy in at a cash game with AA against T7 suited, all in pre-flop? Nice of you to say so, but I don't believe it for a minute.)

But I'm back, and I even have a plan for a few more posts, so there's some hope that this particular ball will get rolling again. Ok, what's the title all about?

In the best traditions of blogdom, I'm now a published author. Well, I've been a published author for a long time, but academic articles don't count -- this is a real book, with covers and everything -- that you have to pay for! And, needless to say, I hope that many of you will.

Lessons in Play, is a textbook or self-study guide in the area of combinatorial game theory. I've only got a second or two to explain that before the eyes glaze over, I know. So, it deals with the theory of two player games, in which there is perfect information, and no chance elements.

I know that it's going to be a best seller, and the first edition will become a valuable collector's item, so you should all rush out and buy your copies now. "Why should I?", you ask, "What does it have to do with poker?". Let's see. Poker is a multi-player game of imperfect information, with significant chance elements. The relevance should be obvious!

Follow the link above, or this one, and if you mention at the checkout that you know me .... ummm .... they'll say "That's nice."

Coming soon, what Lessons in Play really has to do with poker.

A change of perspective

When I learned to play Go, as a graduate student, my mentor refused to allow me to play anything other than lightning games for the first few months. The reason for this was that, given that I was going to be making ludicrous mistakes regardless, it was better that I make them quickly (and analyse them later), rather than spending inordinate amounts of time thinking, and then making a ludicrous mistake anyway. It was much more important to develop a reasonable amount of experience in the initial stages of learning, rather than trying to develop ones skills while playing.

Particularly for an analytical (emphasis on the first two syllables) type like myself, this was extraordinarily good advice, and I've subsequently tended to apply it wherever it was practical. Of course, playing poker is one of those areas and almost all my play to this point has been of the "lightning Go" variety.

My New Year's poker resolution, though it's really too vague to be calling it that, is to move beyond the point where I'm simply playing for experience, and to begin taking my play more seriously. A particular part of this resolution is to fold when my instincts or reads tell me to fold -- rather than calling (which in the experience phase is frequently the only way to check whether the instinct to fold was correct!)

Here's an illustration from an MTT (or is it "a MTT", I can never remember the rules about articles before acronyms) last night:

Blinds still at 10/20, I (1450) get A♦A♠ on the button. There are three limpers in front, two who have me covered, and one who is all in. I raise to 80, the blinds fold, and the limpers call. The pot is 290, and the flop is 2♣5♥6♥. The first limper bets 200, and the other active limper raises to 400. Oh my fur and whiskers! I fold the aces (8♦6♠ and 6♦5♦ -- the first is clearly certifiable, since he continued to bet unimproved through to the river).

Of course this one was easy, particularly as the all in limper meant I'd be able to see the outcome regardless. Still, you don't fold aces all that often, and in any case, each journey begins ... (or choose your favourite cliche here).

The Endeavour or the Titanic?

Nothing quite like trying to complete a grant application under time pressure to sharpen up ones skills at procrastination ...

There's been a bit of recent talk in the poker blogiverse about "floating" (and I'm sorry, I neither have, nor can find the links). Like a lot of poker terms there seems to be some debate (not always civilised) over the precise meaning of the verb "to float", but certainly one of the accepted senses is a call made with the intention of bluffing on a later round.

Defending against a possible float is tough, particularly without reads. The difficulty comes from trying to distinguish between a float, a slow play, and sometimes a draw that has hit. That's precisely why the float, used judiciously, is such an effective move of course.

Here's an illustrative hand that happened in my first return to the hold'em world after the holiday season. First hand of a two table turbo SNG. Buy in small enough that you won't be short of money for the groceries if you bust out, and indeed the cost of your two minutes of entertainment won't be far above that of playing on an unfamiliar arcade game (speaking of which, does anybody actually play arcade games anymore?)

Anyhow, blinds are 10/20, and UTG you get T♦T♠. For no particularly good, nor particularly bad, reason, you open with a raise to 80. All fold, except the SB who calls.

The flop is a very ragged J♠7♦3♠. The SB checks, you bet 120 into the 180 pot, and the SB calls. This seems like a good point to formulate a plan for the rest of the hand. Whatever plan you might have formulated though will probably require some reconsideration, because the turn is 10♣ giving you a set.

Again the SB checks, and, despite the straight possibilities, you bet 200 into the 420 pot, and, to nobodies great surprise, the SB simply calls.

The river is the 8♦, putting J♠10♣8♦7♦3♠ on the board, and this time the SB varies his play by going all in for his last 1100 chips. Of course, calling this will put you all in too. Do you?

At the table, I thought it a tight decision between calling or folding. I folded, in order to maximise my "entertainment value", possibly at the cost of more traditional EV. The SB didn't do me the favour of showing his hand, so I can't tell you whether or not my decision was correct. But, as I did go on to second place, it wasn't too bad!

I'm back ...

Well sort of.

Had a great December in Brisbane, where the locals were marveling at the cool weather (27-30C, with very pleasant late afternoon breezes), while the locals in NZ were similarly marveling (12-15C, with persistent rain), but with a slightly less positive attitude.

I've got a very busy few weeks coming up, as I've got a grant proposal to write, two summer school classes to teach (don't cry for me -- they count as part of my regular teaching load here, and so make the load the rest of the year ludicrously light), and the annual Maths Olympiad camp to organise (the academic side of it anyhow).

So, since I try to keep this blog at least reasonably on topic, there's not likely to be much updating going on. Oh yes, Happy New Year.