Monday, March 30, 2009

Top 2, Part 2

Hef's response to my post about playing top 2 pair in PLO got me thinking and made me realize that I barely scratched the surface of a topic that could be expounded on in pretty good depth. No, I'm not going to write a thesis on 2-pair on the flop in PLO, but I did want to respond and mention some other considerations.

Hef's contends that it is a better idea to not lead out of the flop with top 2, instead choosing a check/call line and re-evaluating on the turn. While this may be the correct approach in some situations, I think it is a flaw to always (or even frequently) take this line when flopping top 2.

Certainly, one's tendency to lead out with top-2 should depend mostly on the aggressiveness of other players in the pot. Basically, I would agree that we'll have a hard time continuing against a raise (without any draws of our own), so we need to evaluate that likelihood. Conversely, if we lead out and get one or two callers but no raises, we will almost always have the best hand right now. This is because low-stakes PLO is typically being played in a very straightforward manner right now. Players tend to push hard with top or middle set, and solid players will also push hard with big multi-way draws (and thus my suggestion that we play much more cautiously on draw-heavy boards). But with simple draws, like OESDs or naked flush draws, these players will just call, and they will also call with weaker 2-pair hands (top+bottom or bottom 2).

We create a number of problems for ourselves when we check into multiple opponents after flopping top 2:

1) We do nothing to define our opponents' hands. If a player in position leads after it checks around on the flop, does he have a strong hand, or is he just trying to steal the pot after no one else seemed interested? We could easily end up in a position where we check-call multiple streets and get taken direct to Valuetown.

2) By checking we potentially allow it to check around, giving a free card that is very unlikely to help us, but is likely to help someone else, either by generating a backdoor draw or spiking a set for someone who would have folded their pair to a bet on the flop.

3) By checking we forego extracting value from draws that are unlikely to pay us off on later streets. By which I mean that drawing hands will only call bets when there are still cards to draw to. If you check the flop and then bet the turn, you will be getting at most one street of value from a drawing hand. But if you bet the flop and then also bet a blank turn, you may get 2 streets of value, and you might also get a third street of value by taking a check-call line on a brickish river in order to induce a bluff from a missed draw.

For these reasons, I prefer leading out when flopping 2-pair from OOP. But, not always. One other consideration is how big your top 2-pair is. Getting value from a top 2-pair hand is of course contingent upon obvious straight and flush draws not coming in, but it is also contingent upon someone else not making a bigger 2-pair. So, when your top 2 is 9-5 on a 9-5-3 flop, or 7-8 on a 7-8-2 flop, I would be less likely to fire out, since your top 2 will sometimes be beaten when an overcard hits on the turn, making your opponent a bigger 2-pair. Generally, these hands, even though they are top 2-pair, have to be considered weaker hands, simply because, by the river, they fall behind a better 2-pair much more easily than hitting top 2 with K-Q on a K-Q-6 flop, or A-J on an A-J-5 board.

And now, after all of this blabbering, I plan to try to collect some hands as evidence. Unfortunately, almost all of my PLO play to date has come on Bodog, where it is a royal pain to collect hand history (some sort of "hand grabber" tool is required). Therefore, I plan on playing some more PLO on FTP, and at some point soon I will hopefully have some evidence that the ideas presented here are sensible and effective. Or evidence that I'm a spewy donk! Either way, it should be interesting, so stay tuned.

5 comments:

Anonymous said...

Danny,
I wish you luck, but if you play the way you describe I'm afraid you're going to find evidence that you're a "spewy donk", as you so eloquently put it. Especially if you do that in a 9-handed game. Your post doesn't mention 9-handed vs. 6-handed, but I'm assuming you will be playing 6-handed online. But since your original post was from a live game, I assumed those hands were 8 or 9 handed.

From your examples, I MIGHT lead out with KQ on a K-Q-6 board, but only if there are 2 or less other people in the hand and it's a rainbow board. But even then you risk getting re-raised by someone with an A-J-10 or a J-10-9, both of which would have called pre-flop. You're ahead, but still have to fold to their re-raise. Also, a lot of donkeys will re-pot you with K-6, and now you're again folding the best hand. And leading out with A-J on a board of A-J-5??! Recipe for disaster. You're going to get called (or raised) by anyone with AK, AQ, K-Q-10, or 2-3-4. And then what are you going to do if a K, Q, 10, 2, 3, or 4 comes on the turn? That's 16 possible scare cards on the turn. Which is why I'd much rather check-call the flop and then re-evaluate. Check-calling has a few advantages: a) cheap way to see the turn, b) your call might slow down your opponent and let you see a free river, and c) the dream scenario: you check/called with KQ on a flop of K-Q-6 and a donk with K-6 bet the pot. You called and a King falls on the turn. Now you take his whole stack.

I'm been playing a ton of PLO lately, and I speak from experience. Top 2 in PLO is waaaayy more marginal than it seems. But I look forward to hearing the results of your experiment. Good luck!

Hef

noldmax said...

Hef, a few points:
1) I do play much more 6-max than full ring, and of course full ring play requires tighter play. I would generally say the same thing for live vs. online, as you will see more boneheaded/loose calls (e.g., TPTK) in a live game than you would online, and you will also face more aggression online.

2) Your point about scare cards is important. There are very few flops which don't leave any immediate flush or straight draws. In fact, there are only 4:
K-8-2, K-8-3, K-7-2, Q-7-2. So yes, we will almost always be facing a turn card that could fill a straight. This is why we should only lead flops on which the draws are limited and obvious. On the K-Q-6 board, far more than half the deck will be a blank. The A-J-5 board is worse, but we don't need to sweat a 4 or 3 as much as the broadway straight cards, since fewer players play PLO hands with 2-3 or 3-4. And still, 16 scare cards is still much less than half the deck.

3) Most donkeys make the mistake of calling too much, not raising too much. Although aggro NLHE players moving to PLO may overplay top-and-bottom and bet pot from in position, in general I think more bad players will call from behind with top-and-bottom and you will be able to extract value from them.

4) I agree Top 2 is marginal, but is still usually the best hand. I think the best way to get max value from it is to bet the flop. And, since it is marginal, it is easy to fold to a big raise. Your approach seems overly fearful of being raised, while mine admits that betting and then folding to a raise is perfectly fine. Meanwhile, in most of the hands where I don't get raised, I'll be extracting value that you leave behind.

Anonymous said...

In a normal 6 handed PLO game, I would neither lead (Danny) nor follow (Hef).

Typically, I will check/raise! It quickly defines the hands - many times that takes down the pot (villian will put you on a set), plus I can fire if the turn blanks and almost always win the hand.

Also has some table image benefits, my oop check doesn't always mean weakness or medium strength.
-ron

Anonymous said...

Hmmmmm....I like Ron's strategy, which I will now implement into my game. In order, I like Ron's play, then my play, then Danny's play. Leading out on the flop and building a pot and then creating a really tough spot for yourself on the turn when you're first to act again is just a losing long-term proposition, imho.

Hef

noldmax said...

Ron,

I don't mind mixing in the check-raise, since it will improve your image, define opponents' hands, and possibly even induce hands like bottom set to fold (if followed up with aggression on the turn).

But, on the downside, you bloat the pot with a marginal holding, and since you are OOP, it is difficult to play later streets. It also makes decisions on marginally non-blank cards (e.g., on the A-J-5 flop, a 2 on the turn) more difficult and more costly.

I like it for balance, but would not use it very often at all, as I prefer to not play a smaller pot with 2-pair from OOP.