Semi bluffing and slowplaying

[spino1i]

Yup this is the problem with only having a super-aggressive game where you always bet when you have draw/made hand, and always check when you have air. Works great 1/2 and below, but 2/4 and higher its going to get exploited by smart players.

[Troll_Inc]

[ QUOTE ]


See, this is just wrong. If I'm playing LHE, and I'm on the button with two limpers and AA, I'm raising 100% of the time.


The only point of randomizing your strategy is to make it so that it's not exploitable.

[/ QUOTE ]

A world class player would limp in the AA situation very occasionally.

The reason to do something is to increase your overal EV from a game. Deception is -EV, so you limping with AA would have to overcome that loss in EV for it to be profitable. Of course what you are looking for is the type of situation where you you limp on the button with AA and because you did so, someone overplays a lower set or lower pair thinking that they have the best hand (because you would _never_ limp on the button with AA).

[iggymcfly]

You're missing the point here. If it's -EV in the given hand, there has to be some larger situational consideration to make it a play you need to make occasionally. Just the fact that someone thinks that you'll "never have AA" in that spot isn't enough to make it a good play, even on occasion, because the extra 2 BB you'll get once in a blue moon when you set-over-set someone won't make up for the 1.5 BB you're losing on the flop every single hand.

It's only a good move to make when there's some way they could exploit your limp if you never have AA there. In a 5-way pot, your opponents will still have to contend with a large field, and even if you're opponents know you don't have a big pair, they'll still have to deal with the myriad hands that could be hit by other players. Over 1000 hands in that given situation, I don't think a world-class player would ever limp unless there was a really specific situational consideration (i.e. BB always raises when limped to, never 3-bets) that he was looking to exploit. It would never be a good play for simple deception however.

The point is that there are multiple ways to prevent something from being exploitable, and while randomizing every single hand is one of them, it's far from the most efficient, and will cost you much more EV than just determining if you're exploitable on one board.

In addition, this (being unexploitable) is such a secondary concern that you can ignore it against most players, and even against the most observant players, you only need to do it in certain situations. (i.e. as long as you were occasionally checking medium to strong hands, you'd have to play probably a minimum of 5000 hands with an observant player before he'd even start to notice that your checks on straight boards always meant you'd fold the turn or river were that true). Randomizing your whole play, even when sitting with a table of unkowns is costing you a very significant amount of EV and will lower your winrate severely. FWIW, I've only played with one player ever who I even had to worry about issues as specific as randomizing on this type of board. Trying to break it down further than that is ridiculous.

[LA_Price]

I was reading through mathematics of poker last night and reading the debate here between Iggy and Troll I think what the argument is about is whether to play an exploitive strategy, as if there were no hand besides the one you are playing, or to randomize your plays against some regular opponents.

So I think if you are playing online against a group of players with whom you play with all the time, or are playing in a live cash game that you will probably only play in once makes a difference.

Just one question for iggymcfly you said: [ QUOTE ]
Just the fact that someone thinks that you'll "never have AA" in that spot isn't enough to make it a good play, even on occasion, because the extra 2 BB you'll get once in a blue moon when you set-over-set someone won't make up for the 1.5 BB you're losing on the flop every single hand.

[/ QUOTE ]

Say you never make this play with AA or one like it in pot-limit omaha against regular thinking opponents. Might they not generalize not into the fact that you'll "never have AA" when you limp but that you will never limp with big hands. And if you did make a play like that might they not think now think that you sometimes play hands in random fashion, which would make it more valuable, no? They might now be less sure of their "reads" of what you could have in any situation.

Also would still like to hear from Dave and/or Pete on some good spots to slowplay.

Just one more thing. Limit hold'em? In the future lets try and use Omaha examples when trying to make a point. It's bad enough to have to play that game everytime I want decent stakes live here in Minnesota, but to discuss it strategically should be considered blasphemy in this forum.

[Big Dave D]

You're back in the US?

I guess an example of a slowplay might be a middle set or the obvious flush draw like you said. But in general, in 6 max at least, I tend to fast play almost everything and I'm not convinced you pick up enough equity in deceptive slowplays. But the fact I almost never bet the pot probably helps a lot in playing this way.

gl

bdd

[LA_Price]

Yes and I haven't played a live Pot-limit Omaha game in over a month. I had found a 2/5 1k max buy-in at the Horseshoe in Omaha, Nebraska but it morphed into 20/40 LHE/L08 which to me is still better than playing just LHE, but just not the same. There needs to be a PLO cash game TV show so people will actually play this game and realize how awesome it is.

[iggymcfly]

I had a reason for bringing up the LHE example, one I thought was clear, but which apparently the forum missed. OTB after two limpers in that spot, there are about 50 different hands you can have. Therefore, there (should be) no question that you can play one hand the same every time without losing anything in deception. Saying someone "doesn't have AA" means nothing in LHE, because after the flop, they could still have TPTK, top two, a set, a different overpair. It means nothing.

I figured that this point would be taken for granted, so I could move on to my main point, but Troll tried to argue that you should sometimes limp AA in a 5-way pot "for deception" (not just because of an easy LRR opportunity) which is 100% wrong. If I posted it as a poll in the High Stakes LHE forum, I'm guessing, I'd get > 90% agreement with the disagreements coming from people browsing the forum who don't actually play high stakes.

Anyway, the general point was that it's not how you play one hand that has to be randomized. It's how you play a certain situation that has to have more than one option. Staying with the LHE example, (if I can't get people to understand it in the simple context, then it's impossible to get them to understand it in PLO, where it's less obvious), something like "AA OTB with 2 limpers" does not have to be randomized. Just because they know you don't have AA doesn't necessarily mean they'll be able to steal the pot on any given flop with a decent percentage of success.

Now if for instance, you only limp JT, QT, QJ, or T9 in that spot, that would be exploitable since your opponents could take the pot away on any flop that didn't contain a 9 through J. The point is that a range that you make a given play with can't be exploitable. So, in this case, the situation is "on the button with two limpers" and you just have to make sure that neither you're limping range or raising range is exploitable in this spot. To make yourself unexploitable in this spot, you could occasionally limp with something like AJ, 77, or 67 and make it so that your opponents can't figure out what you're doing without actually ever limping with AA.

That's what finally brings me back to the PLO example. Just because you want to use an unexploitable strategy does not mean that every hand has to be open to randomization. You can still take your very best hand (in this case the nut straight) and bet it 100% of the time. Your opponents can only exploit this if they know that they can steal from you regularly on blank turns. As long as you sometimes treat your sets or two pairs as made hands and call them down to the river, you can still have an unexploitable strategy without sacrificing tons of equity on your absolute top hand.

Also, if you're going for even somewhat of an exploitive strategy (which a good PLO player should be) the best way to exploit this flop is to go for a pure steal much more often than you're betting with a hand that you're willing to get all-in with. Thus, your strategy is already unbalanced toward betting without a hand. Since this is our primary "leak" from a game theory strategy, we need to balance this by betting as much as possible when we're actually +EV against our opponent's "play back at us" range. Since most opponents have the nuts over 50% of the time when they raise us back, the only hands that are +EV against this range are our nut straights, and every time we check them, we're actually making ourselves even more exploitable.

Note that I'm not saying that it's wrong to use randomization and deceptive plays in some situations to make yourself less exploitable. I'd have no problem with someone saying that they're betting top set 35% of the time on this flop or even that they're checking top set on a dry board 10% of the time from MP. I often say things like "I'd call this river about 1/3 of the time." I'm just saying that this particular hand on this board is one that should be bet every single time, and doing anything else is a mistake that will lower your winrate.

[pete fabrizio]

I agree with almost all of what Iggy has said in this thread. The old poker saying of "nothing every time" is pretty much hogwash. However, I do think there are a few reasons to play in a way that makes me less exploitable, even if it would take my opponents many thousands of hands to figure out exactly how I play (if ever). First, it gives me peace of mind that I'm not being exploited. Second, it ensures that some idiot isn't just stumbling on the right strategy against me.

I generally play a mixed, less exploitable strategy the most when I'm playing people whose games I don't really know, and I just want to coast along until I figure out the most profitable exploitative strategy to use against them.

[RoundTower]

Good post iggy, I think you are right in general.

In particular I believe that you can have an unexploitable strategy without ever checking behind in the 68T example above. In the nut flush draw one, not so much. I think you should definitely check behind there sometimes not just for metagame but because it can be the best play.

[Troll_Inc]

[ QUOTE ]
Yes and I haven't played a live Pot-limit Omaha game in over a month. I had found a 2/5 1k max buy-in at the Horseshoe in Omaha, Nebraska but it morphed into 20/40 LHE/L08 which to me is still better than playing just LHE, but just not the same. There needs to be a PLO cash game TV show so people will actually play this game and realize how awesome it is.

[/ QUOTE ]

Sheeet, I was just there last night and only thing going was 4/8 O8b and tons of NL. I read the the BM forum and it didn't look like it got much 2+2 traffic. I guess I should have posted something about trying to get a PLO game going.

I'm surprised they don't mix in PLO into High Stakes...PLO really is a better cash game than NL. NL cash games are sooo boring.

And BTW, Omaha residents would be horrified if you got some place in Iowa confused with Nebraska. (Binions is in Council Bluffs)