PDA

View Full Version : On the subject of "reads"


Big Limpin'
05-07-2005, 05:00 AM
"I put him on a premium hand because he has been playing tight"

"Villian seemed semi-maniacal, he had been speeding around for levels 1/2, so my AJ looked strong against his probable holdings"

etcetera, etcetera. I read this kind of stuff all the time in ppls posts. And let me say upfront, im brutal at reading a person for playing style. I dont even try. But as i become more and more at ease with the cadence of 4-tabling, i find i can observe the goings on in hands im not involved in pretty well. That is to say, i'm now finding myself recalling previous hands when im trying to put a villian on a hand(s).

But heres the thing...we play on sites with such a deep player pool, where you dont see the same player over and over again (high limit players pls ignore, this doesnt apply to you). Essentially, you are basing your reads on what a player has done in the last, oh, 20 or 30 hands. What is this really, in the grand sceme of things?

If theres one thing we all should have picked up by now, its that poker is a game of swings. You run hot, you run cold. You go 9 in a row OTM. You pimp a 65% ROI for a day. Seriously, what is 20 hands?

I guess what im saying/asking here is how much weight can you put on a guys previous actions in a SnG? I mean, for sure, if a guy is seeing 50% of the flops...sure you can deduce he isnt the most discerning player, but generally, i'm thinking that your "reads" are as much due to the cards he's been dealt as how he plays.

I've played games where i havent seen a flop until level 4. I've seen 8 flops and been raising like a mofo. So have y'all. The observant villian will certainly have catagorized me differently in these two scenarios, but im the same guy. I'm gonna play my level 4 hand for what it is, irregardless of my previous actions.

How many times have you called a "maniac's" push with Aj to been shown KK? Even if you have catagorized them accurately, they are still dealt the same cards as you, and wake up with a hand as often as a rock.

***
Well, i just proofread (semi) what i wrote, and i guess i really haven't much else to say, although it seems incomplete. I suppose i'll just leave it at that, perhaps responses will spur my juices. I feel like i want to say more, but dont know what it is.

I'll leave it open-ended...hoping y'all can run with it.

What i want to know is:
/images/graemlins/diamond.gifHow much weight can/should i put on previous actions?
/images/graemlins/diamond.gifIs this a part of my game that i can work on (like i say, i suck at reads), or, as the tone of this post suggests, am i correct in brushing off "reads" as vauge, as likely to induce mistakes in my play as to gain an edge?

Ok, this kid needs sleep, g'night

shejk
05-07-2005, 05:29 AM
I observe mostly people doing weird things. Like folding when they had odds for calling with any two, or limping utg lvl1 with A2. How I use it then, well not very much. It's reassuring to see that the guy who just pushed into your aa has been really live, and I might push even more late if someone is folding more than they should. That said it is a minor consideration.

So in conclusion: Taking notes and trying to see who sucks the most at the table is good for maximising your earnings when playing postflop. At the later stages I don't use it that much.

Just my thoughts on the subject... hope I spur something.

Big Limpin'
05-07-2005, 05:34 AM
thanks. but,[ QUOTE ]
It's reassuring to see that the guy who just pushed into your aa has been really live

[/ QUOTE ]
How T.F. does this matter? My AA stacks up better against KK than TJ.
But htats small potatoes, i get yer drift, generally, thanks for the input.

pergesu
05-07-2005, 05:36 AM
I don't think you can have 100% faith, but you need to take reads into account. Yeah, 30 hands is nothing to make solid conclusions from, but it can still give you an idea.

To put it better, you can't form any conclusions about an opponent's long term play, but it can give you an idea of how he's currently playing. Imagine if you played with a rock that only ever played AA, folded everything else. You play thousands and thousands of hands with him, and that's all he plays. One night he gets drunk and starts raising every hand. Now you have to adapt your play to this short term read - it won't apply to him all the time, maybe even never again, but it should still help you out in this particular situation.

Pepsquad
05-07-2005, 05:40 AM
This is why it's referred to as a game of "incomplete information". Does that mean the precious little information you've been given should be ignored?

Yes. Sometimes the maniac with 43% VPIP turns over KK.

But 43% VPIP is 43% VPIP is 43% VPIP.

We can only make decisions based on what we know. Failing to do that, I'm able to convince myself that THIS is the time he has KK or THIS is the time the TAG is running a stone-cold bluff despite what I know.

Pep.

shejk
05-07-2005, 05:57 AM
Sorry, I was referring to postflop play. Just called a 800 push on the turn from some maniac when I was holding aa.

I came to think of the weather when thinking about this.

If you had a weather service that announced every day, that the weather would be the same as yesterday, it'd be more correct than the crap the say on the news.

Phoenix1010
05-07-2005, 06:07 AM
It is possible to gain (a very small amount of) useful information from very small sample sizes. Every showdown, every raise and checkfold, every big blind fold, every individual action on every hand provides something. The information from each hand individually is miniscule to the point of being useless. But after five or six hands you can start to take note of trends, especially if they're doing the same things over and over. After 20 hands, the trends become more defined, and more useful. If someone has raised all 20 hands, you know something; if he has folded all 20 hands, you know something; if he has done some mixture of the two, you still know something. You can start to get a feel for someone's tendencies very early, and the reliability of this information grows exponentially as the sample size goes up.

With that said, yes, most online reads at the lower levels are very unreliable, and should almost never be the main determining factor of any decision. We have math and common sense for that. However, sometimes reads can be an important part of refining our math. When you have to put someone on a range of hands, or have to estimate your fold equity, or the chance that your opponent is bluffing, you have two options: you can make your decision based on what the average player would do, or you can base your decision on what THIS player would do. If you're paying close attention, you can sometimes get a pretty good idea of how this player differs from the norm, or how he adheres to it. For close decisions, these observations can be the difference between +EV and -EV.

To sum up: Don't bet the farm on reads, but don't ignore them either. Use them where they are appropriate, and always with a grain of salt.

-Phoenix

Cry Me A River
05-07-2005, 08:32 AM
I think that in SNG's the most important read, by far is the bubble read. Particularly, "too tight" or "too loose", this is really the case with the big stacks and little stacks - Big stacks on the bubble are often either "Fold into the money" or "bully" and you need to adjust your approach accordingly. Similarily, the short stack is either "waiting patiently to blind out or get a premium hand" or "push, push, push". While most mid stacks will be playing too tight, you need to know which are capable of a steal and which will only make a move with a real hand.

You really need to assess your stealing opportunities. It's one thing to say "push any 2" when you're UTG with 5xBB and the BB is going up next hand but you need to know when you can steal with any2 when you're an average stack with 12xBB.