Saturday 3 May 2008

Lesson: starting hands

So we learned what hands are worth the wrong way round, right? Because you don't start with straights and flushes; you start with just two cards.

Well, no. We learn the other way round for two reasons. One, you need to be completely au fait with what beats what. You must not be confused about whether straights beat flushes (they don't) or full boats beat flushes (they do). This is because you need to be able to evaluate the board quickly.

You need to be able always to answer these questions: what hand can I make? what hand is the best? which hands could happen in the next card or two cards? and finally what is my opponent likely to have now, and what is he trying to make?

So you have AQ, a strong starting hand, and you flop QcTd7d, you bet and you are called, you need to know that you are probably best with TPTK, your opponent may have a worse queen or a flush draw, but he can be "trapping" with QT, TT, 77. He might also have 89 for the straight, maybe J9. You have to be aware of the possibilities.

It comes with time. You can't instantly read the board when you're a beginner. It's a skill that you learn (imperfectly: I'm still sometimes surprised by a hidden straight). But you can help yourself with study: save hand histories or ask for mine. Look at hands as they played out. Work out what was the best possible hand, and what people could be drawing to. The more you do it, the better you get at doing it. (If you want help, Father Luke, I will set you drills. Just ask. If you have the time, it's worthwhile to do. Reading the board correctly is a sine qua non of good poker, which is very hard for a novice, but does become easier. In time, we'll have you play some Omaha, because it gets you into the habit.) It is also important to understand, as a first principle, that hand strength is relative. This concept is fundamental in poker, and its importance in STTs cannot be overstated. There are spots in STTS where you cannot play less than aces! This is not because your opponent has a strong hand, but because yours needs to be relatively that strong.

Okay, so you can look at a list of starting hands, and that's fine, but it's very abstract (however, in STTs, having a grasp of the order of hands is important, because you need to be able to rank hands by percentiles). There are more concrete ways to understand starting hand strength.

There are basically three dimensions: high card strength, suitedness, connectedness. In most forms of poker played for high, you should select your starting hands according to these dimensions. The most important is high card strength, overwhelmingly so in STTs, but the other dimensions are also important. (High card strength means how big the cards are. Generally, how big both cards are matters, because if you pair, your other card will be your kicker. KQ is stronger than KJ among other reasons because a queen outkicks a jack. On a jack high board, KJ overtakes KQ though!)

We are going to talk a lot about equity. It's basically your "share" of the pot. It's how much you would expect to win against the range your opponent can have. Equity varies according to the cards on the board, and the range you think your opponent has, but when it comes to starting hands, we care a lot about "hot and cold equity". This is the chance you have of winning if you are all in with your opponent. So we say that AA has 83% equity hot and cold against 72. This means that if you are all in with 72 and you hold aces, you will win 83 times out of 100. (Actually, you may or may not. We are talking probability, so you expect to win 83 tims out of 100, but you can win 50/50 or 100/0 and that's quite possible. However, in poker we weigh our chances by strict probability, even though in principle we would need to play these hands out a million times to converge on our real expectation.

The point is that rankings of hands are less important to us than their equity against each other, and to a lesser extent how they play. Obviously, you can rank hands by their hot and cold equity. AA is the best hand, dominates all others and will win most of the time; 72 is the worst starting hand, will lose most of the time (but if the flop comes 752, the mighty aces are humbled). Learn that lesson: poker is situational. There is no best hand. AA is best preflop, but if you flop 772, AA is drawing to two outs against 72.

But we are thinking about preflop, and going back to our dimensions, high card strength is best. But you must remember that there is another dimension: hands can be made or drawing.

Pairs are by definition made hands. They have value purely by virtue of being pairs.

AK is a strong hand, but 22 beats it. (Hot and cold, 22 beats AKo, and the reason you rarely get it in with 22 is that you are uncertain that your opponent has AK or similar and not a pair.)

Suitedness helps. It adds a few percent to the equity for any hand. You shouldn't underrate the value of being suited, because many decisions will hinge on a percent or two either way, which makes a suited hand considerably stronger than its unsuited equivalent.

Connectedness also helps. 76 are connected. 76s are suited connectors. These hands are stronger than other rag hands because they can make straights with both cards together. Unconnected hands can only make straights with four on the board, which is not only rarer, but makes your hand obvious to opponents. If the board is showing 98652 and you're betting, you probably have a 7.

Heads up, most hands are won by the biggest pair. With experience, this will seem obvious, because you will play so many heads up hands in which no one has even a pair. But it means that AA is strong not only because it is the best starting hand, but also because it will "hold up" often and remain the best hand. (Indeed, like most pairs, it rather has to remain the best hand, because it only has two outs to improve).

So we are clear that AA, KK, QQ are excellent hands, and we can get it in with them fairly confidently.

And JJ and TT are very good starting hands. Sometimes, 99 is too. Remember, few hands start better, and even impressive-looking hands like AQ must catch up with it. At higher blinds, these pairs are all worth playing hard. Not so much early, for reasons I'll discuss when we talk about play at different blinds (in a nutshell, because if you get it in at low blinds, your opponent will tend to have a stronger hand than at higher blinds, because the pressure of blinds and the shorter table mean that he has "opened" his range somewhat--opening your range means to play more hands, and few players don't do that as the table gets "shorter"--has fewer players).

Other pairs have some value, and in early levels can make sets if they flop a mate. That happens about one in nine times, and if you can play cheaply, makes a decent way to try to get a big hand.

This is an important idea when you look at starting hands: you need to think about what hand you're aiming to make with it. Playing 44, you are looking to hit a set. With AA, you are hoping to get it in and watch your opponent hit nothing. Play AK and you are trying to flop an A or a K, to get TPTK. It's the ultimate "top pair hand", that is, a starting hand whose aim is to hit top pair and have that hold up.

These hands are viable because top pair wins most hands heads up. A word of caution though. If you don't raise AK/AQ, and sometimes you don't, and there's a crowd to the flop, TPTK will often be beaten when there's heavy action.

There are lots of ways to rank the starting hands, because they do differently against different ranges, and as I noted, some play better than others--AK is weaker than 22 all in, but you'd rather have it to start with, because if the board comes Q73, you will never think 22 is good, even though it sometimes will be, whereas you might bet AK. The chart here is pretty decent. I don't suggest trying to memorise it or anything, but it will help you to get a feel for what is strong and what isn't. Don't worry about the figures. We'll come to that soon.

One reason not to memorise the chart (apart from, obviously, that it's very hard to do so, and isn't necessary at this point) is that it's better to think about what's good and why.

Let's take a hand as an example. KJs. It's a very decent hand but why?

1. It has two high cards. A pair of kings will win many pots, particularly heads up, and a pair of jacks is also a decentish hand. King high will win some pots all in, but you must take care: it is behind all aces, even the worst of them.
2. It is suited. Suited cards make a flush 6.5% of the time. That's huge because a flush is a very strong hand, which you will just about always be willing to "stack off" (get your whole stack in) with.
3. It is connected. There is one gap between K and J, so we call it a onegap connector, or a suited onegapper. KT is a twogapper. K9 is a threegapper, which is as wide a gap as you can have and still make a straight on the flop. The fewer gaps, the better. KQ is better than KJ not just because it has higher cards, but because they are closer together. I'd consider JT and K9 to have about the same value, because what JT lacks in high card strength, it makes up for in connectedness.

Let's look at another hand, which many low-limit players overrate: ATo

1. It has some high card strength. The A is obviously as high as it gets, but in the early levels, the T is often a liability not a plus. Hit your ace and you have TP with a T kicker. Given that all your opponents will play the big aces, you can easily make a secondbest hand with it. You are generally happier with a middling ace to hit your kicker than the ace. AT is a stronger hand on a T high board than it is on an A high board.
2. It is not suited, so it will not make many flushes.
3. It is a threegapper, and you will rarely make a straight with it, not only because you need three particular cards to make it (with, say, KQ, you can make straights with AJT and T98, as well as fourcard straights). Those cards are all in the "playing zone" (cards that will often feature in your opponents' hands, because they are playable and you can expect them to be out there if you are "multiway" (more than two players taking the flop)--pretty much T to A).

A final note though. AT is not strong enough to play UTG early in the tourney, and you do not want to call raises with it either, but later in the game it becomes strong enough to play. Remember, hand strength is situational. Against eight other hands, AT may not be strong, but against two, it figures to be the best anyone has.

1 comment:

Father Luke said...

Well, no. We learn the other way round for two reasons. One...

ahhh. okay.

You need to be able always to answer these questions: what hand can I make? what hand is the best? which hands could happen in the next card or two cards? and finally what is my opponent likely to have now, and what is he trying to make?

And, so, if I'm evaluating the
board and I see a player betting
big, and I can't see any conceivable hand, then there is a
bluff. I mean, provided I know what
possible hands might be made.

Hey?

Don't laugh... I've seen some
pretty bad bluffs in the STT's.
Fuck, man. I mean bad bluffs. And
then, usually late in the tourney.
Not so much in heads up, then,
too, but not so much in heads up.
Some maniac has a large pile, and
is trying to buy the blinds. When I
have the nuts, I just call, then
fuck them in the turn.

but yeah...

But you can help yourself with study: save hand histories or ask for mine.

The Harrington stuff you turned me
onto has these hands that work as
quizzes all through it. Like chess,
learning hands makes for a mindset
which lends itself to the task at
hand: becoming adept at poker.

So, I'm on it. Thanks... ;)

(If you want help, Father Luke, I will set you drills. Just ask. If you have the time, it's worthwhile to do. Reading the board correctly is a sine qua non of good poker, which is very hard for a novice, but does become easier. In time, we'll have you play some Omaha, because it gets you into the habit.)

I'll always take help. And, yes, it
is becoming easier. I'm sure my
actions, as sneaky as I thought I
was being, were literally
transparent to anyone else. Well,
almost anyone else.

So, yeah. But wut's Omaha?

The point is that rankings of hands are less important to us than their equity against each other, and to a lesser extent how they play.

Right. I'm getting that. I'm
understanding it...

but later in the game it becomes strong enough to play. Remember, hand strength is situational.

nail on the fuken head. Early on
plays so different than heads up.

- -
Okay,
Father Luke