Metagame vs EV plays

I play a lot of STTs, and ever since I started taking them seriously, I've debated what the ideal style is for them, considering metagame and +EV plays.

I play a TAG style early on, and a LAG style later on for several reasons. One of them is metagame. I'll play super tight early to advertise the image of a nit who won't play anything but the nuts. I raise, they all get out of the way. Then of course I exploit this later on when the blinds are worth stealing, and just attack like crazy because everyone's so scared of me holding a big hand.

Now the issue here for me is that I sometimes play so tight that I think miss out on plays that could outweigh the value of having that tight image in the later stages.

This shows up in other games too, obviously. Someone might sit down at a ring game and play like a lagomaniac to advertise a wild image so he can use it to his advantage later. I've heard (and have done so myself) that people lose multiple buyins just so they can win even more later. But then you run the risk of the players at your table leaving (and your image being forgotten), or just plain losing too much in the LAG period of your play.

It's kind of hard to measure but where do you draw the line? I'm still varying between tighter and looser styles in my STTs, and haven't found a sweet spot yet, hence the thread...

I find that most of the people at the STTs at the stakes I play ($6, $11) don't pay attention to other players' images and focus mainly on their cards, so playing a tight +EV game is more profitable.

I agree with Vanquish. I play $11 ones and people don't really seem to pay fcuk-all to everyone elses play/image much. So I just play tight but open up a little on the bubble to steal some blinds n such like you said chuck.

Well that's the issue - whether or not people are observant enough to make the style worth it. I play the $16 turbos on stars, and see players from both ends of the spectrum. Multitabling, 5-figure winners, and absolute fish.

bump :/

No other thoughts/opinions? Maybe this isn't as important a concept as I think?

I think that image is much more important in live games - not to say that image isn't something you should watch when in an online game, but others often won't, and when that's the case, you'll be wasting your time/money.

Live, though, it's much easier to keep track of how people are playing because you have more than a screen name (and maybe avatar) to use as a mental image of somebody. So, in those situations, adjusting the way people perceive you early on could be much more beneficial.

Tag early/ Lag later doesn't only work as a combo because of metagame considerations, it works because according to the fundamental Theorem of Poker, it has the greater chance of making us act in a +EV way (we reduce our mistakes while our opponent makes more of them).

Early on, it makes no sense to raise marginal holdings (A10o) OOP because:
a)We are likely to be beaten or not far ahead (and are thus making a mistake preflop)
b)The risks greatly outweighs the rewards: the antes are small and we have nothing to gain by showing strength when we don't have it.

In the later stages (four-handed in a turbo) on the button, even a hand such as K8s is good for a raise if we have a suitable stack since:
a) We are likely to be ahead or at least even to the blinds range and opponents ar elikely to fold (thus committing a mistake if they were favorite)
b) The risk of getting raised is not that great unless facing tricky players and the reward is great (antes are high in relation to stack size)


We are putting ourself much more in harm's way by playing LAG early on than later because of antes/cards considerations. This is not to say that no on ecan profit from playing LAG all the way, or a mix, (because clearly, TAG in the later stages is a mistake), but unless you are facing tricky players and have to use deception or strategies that aren't optimal, TAG/LAG like you are doing is probably best.

Also, relying too much on table image makes two assumptions, wich can be wrong:
a) Players are paying attention to what you're doing
b) Players aren't paying enough attention to see that you have changed gears.

Assuming decent to average observant players, table image is a good tool, against complete fish or great players, not so much, since they will not see what you're doing or they will so clearly see what you're doing that you will become easy to trap.

I don't think the first part is necessary. I very much agree that it's good to raise a lot when the table becomes short-handed and the blinds become big.
But you don't need to play like a nit early on for this to work.

The reason is twofold: For one thing, most people tighten up around the bubble. Even if you played semi-loose early on, the other players will mostly fold to your raises.
And second, your table image is worth much less than you may think. At lower buy-ins people won't even notice, and at higher buy-ins people are smart enough to know that a decent player will change his style as the blinds go up.

While this is true, it is argued that it is easier to identify patterns in people's betting, irregular behavior, and other clues that allow one to label a player as a certain "type" (image) much faster than in live action, merely because of the absence of other distractions and the rapid rate at which hands are played.

An example here is:
The player two to the right of you raises from CO, you re-raise from SB, he folds. After you do this twice, he should already have a label on you as a player who will re-steal from the blinds, and will likely adjust his game accordingly. Of course this may not be the case at $0.1 SNGs, but it is generally true.

Exactly. In Killer Poker Online 2 John Vorhaus notes that the "context density" is much higher online than live (chapter four). Not only is there less conflicting data/information to filter (other than player betting pattern), but you see (similar) situations far more often and in a shorter time frame as well.

If you wanted to figure out if the bb was an easy target for your button steals in a live game you might have to wait hours before you obtained enough (even if only two or three) samples to make a decision; whereas online, you can get this read much much faster.

It's the same for your opponents: if you are coming over the top of someones raises every time they enter the pot online, or if you are stealing someone's blinds every time it's folded to you on the button online, you can be sure that it is getting noticed. While 'folding' might not be as noticeable as doing one of the more aggressive/active actions, it still gets noticed (even if only on a semi-conscious level). In a live game, however, so much time might pass between these actions that unobservant opponents may not necessarily see the pattern.

Because of the deep context density online, image matters a lot more than live. Even your most unobservant opponents will be forced to notice how tight/loose you have been playing-- there's just no possible way for them not to. They may ignore that you haven't played a hand in forever and still call with KTo (at certain $ levels), but that doesn't mean that they didn't notice.

I know this is old, but I play almost strictly STTs and have been thinking about this a lot.

While it would seem important to maintain that tight image in the early stages to exploit in the later stages, I've decided to loosen up slightly, limping with small pairs and suited connectors a lot more often.

The thing is that you've got players who are either donks, or decent multitabling players like me. The donks are rarely observant enough to notice you folding ~%5 more hands, and doing so is pointless. While the 'good' players do notice this, they're also more keen in general, and know that you'll be raising with less than premium hands in the later stages anyways, so again, doing so is pointless.

I still play TAG in the early stages (for strategic rather than metagame reasons), but I've loosened up some.

That elusive 'sweet spot', the holy grail for all poker players is a moving target!

I play the same way Chuck. I have noticed that those players who do notice are the ones who tend to be there when the table gets short handed, those who don't generally are gone by then. Of course this is a generality, but it seems to hold water.

Where you have loosened up, I have been working on the AG in TAG. I took the DonkeyCom test last night when I was feeling especially passive. Flunked.

I think I can read a lot of the online players, but usually I have problems pulling the trigger, especially when I have been tight through the beginning stage. Once the table is half its original size I tend to burst on the scene.

It has been sort of successful.