Stat-reads vs. real reads for multitablers

[Harv72b]

Grunching.

At most times, I'm only playing 3-4 tables, so I consider myself to be a borderline multi-tabler; the guys playing 12 tables at once, they're the real multi-tablers. So perhaps because of that, I was struck by one statement made in your first paragraph:

[ QUOTE ]
But the ones that play 3-4 or even more tables at a time are too busy playing cards to be able to notice what goes on at the table during the hands we folded.

[/ QUOTE ]

That's simply not true, or at least it isn't in my case. So long as my mind isn't wandering (I don't dispute the ADD accusation) to where I'm playing minesweeper or surfing pr0n during hands, I do a pretty good job of taking in all the action on all my tables. I think a big part of that is prioritizing.

If one of my tables is filled with guys that I have hundreds of hands on, then I'm not paying much attention to that game. I instead focus on the other table(s), where I have unknown or less-known opponents; when one of them is involved in a hand that goes to showdown, that's where my old friend the last hand window comes into play. Pop that open, see what he mucked, give a quick look at that hand vs. the board & action, and make a mental note. If it's that unusual (particularly if he bluff raised on a late street), I make a physical note on him.

Honestly, most hands that you play don't really require a whole lot of attention, at least not after you've logged your first 100k hands. The proper decision usually pops into mind immediately, and if it doesn't, then I can almost always make the decision within a second or two by glancing at pot size & the stats on my opponent(s). When I'm still not sure what to do, I open my PAHUD popup window on the aggressor in the hand, which has all kinds of little goodies plugged into it: aggression by street, won @ SD, went to SD, and my personal favorite, won SD when raised turn. Of course if I have any notes on him which I haven't looked at yet in the session, then I'll check those out quickly as well. This may take anywhere from 3-5 seconds, depending on how long it takes me to click the right spot on the table.

I've never bothered doing any detailed analysis on my opponents because, well, I really don't think it's necessary. 99.9999% of opponents can be fairly well shoehorned into a pre-existing definition, such as LAG, calling station, TAG, good TAG, etc; if they ever exhibit any behavior that differentiates from that definition, that's why I have my notes. Often times, as you briefly mentioned with that 75o example, those "reads" can be firmed up over just one or two hands.

As for Everest & other sites where the software doesn't work, well, I still don't think there's much of an adjustment to make other than taking a few more notes. And, of course, to always take the size of the pot into consideration when determining whether to call or fold a late street raise--I've dragged more than a few pots over the years after a loose/passive raised a river scare card, simply because the pot was too big to justify folding for 1 more bet (they do bluff every so often, especially if you've got them frustrated after consantly isolating & value betting them to death).

As far as reasoning, it's mostly a case of simple human psychology (which reminds me, Psychology of Poker--good read). At these stakes in particular, I don't think there's much point in trying to delve any further than the basics, which pretty much tell us that a LAG likes to take risks & is therefore more prone to bluffing postflop & making bad (cold)calls preflop, a rock hates risk & is therefore more likely to go passive (especially preflop) with his big hands, and a calling station hates the thought of ever folding a winner. A TAG we know has at least a passing familiarity of the concepts of +EV poker, and we might actually learn something about him & his style of play by analyzing the hands afterwards...except that I still don't think it's worth it, as we're not making our money off of him (unless he's a bad TAG, in which case all we need to know is that he gets married to hands & doesn't understand position/adjust to his opponents).

One thing that you didn't at all mention in your post but which definitely bears bringing up to is to consider your own image as it relates to the table & specific opponents. If you're stuck, and especially if you're stuck & the whole table just saw you make an unusual play, then you can expect to get more action with your hands than you would if you were running the table over. If you've got 5k hands on an opponent, then you can pretty safely assume that he's got 5k hands on you as well & probably has a decent idea of your style of play. If you're a dumbass like me and play under the same screenname you post here as, then you can expect other 2+2'ers to play a little differently against you based on your posts. If you have been value betting the bejeezus out of a guy, then eventually you can expect him to stop calling (even if he's a calling station)...so that's when you start bluffing against him. Things like that should always factor into your close decisions at the table.

I dunno, I feel like I've rambled forever & haven't really addressed the primary topic at all, but maybe there's something worth reading in all this. At any rate, I have to be up for a 13 hour workday in 6 hours, so I should probably wrap this up.

Excellent topic. [img]/images/graemlins/smile.gif[/img]

[neurotiq]

[ QUOTE ]

neurotiq said just like 2 days ago:

[ QUOTE ]

Maybe fold this guy preflop. A 4% PFR is icky for our AQ.


[/ QUOTE ]

This was on a 46 hand stat... [img]/images/graemlins/smirk.gif[/img]

[/ QUOTE ]

Let's not play this game, Bellatrix. I could point out that just the other day, you rejected a 5,000-hand stats post because you said that it wasn't a large enough sample size to recognize any meaningful trends.

I think that Smurph64 said a lot of what I was getting at, only he said it better than me.

As he says, "There should be convergence at some point but I have found stats misleading so many times I just use them as broad point guidelines." He also gives the example of how people have misread him based on stats, thinking that he was a maniac when, in reality, he was just running really hot.

This is the point that I was attempting to make. Over 50 hands, you could have bad luck or good luck, and stats could be incredibly misleading with that in mind.

You yourself acknowledge that stats "can vary wildly," so I'm not even sure how we disagree. You also acknowledge at the end of your post that reads are more valuable than stats.

I couldn't agree with you more.

EDIT TO ADD: Stats also fail to recognize the phenomenon of tilt. A TAG can easily tilt into a LAGgy range at times. If you start recording stats during the 50 hands at which the TAG is tilting, then attempt to use those stats in a later session--well, it probably won't work out too well. [img]/images/graemlins/smile.gif[/img] In a similar vein, if all your stats were on the TAG when he or she wasn't tilting, those stats will mean nothing once the TAG goes on tilt. This goes back to Bellatrix's point about reads being more important than stats; if you're paying attention and get a read that the TAG is currently playing differently than his/her stats would suggest, you can adjust your play accordingly. You can't do that with stats alone, though.

[TimovieMan]

[ QUOTE ]
I dunno, I feel like I've rambled forever & haven't really addressed the primary topic at all, but maybe there's something worth reading in all this.

[/ QUOTE ]
There definitely is! Great post!


In reply to previous posts:

I didn't actually mean that 50 is a magic number at which stats become accurate, that was a simple case of “not my best choice of words”, that's all. In fact, I'm really convinced stat reads are not accurate at all. Someone with a PFR of less than 1% could still only PFR 72o and limp with AA and KK, and unless you see him showdown or check for his mucked hands, there's just no way of knowing.
Real reads however ARE (in most cases) accurate. That’s not something I’m going to discuss, it’s a given fact that stat reads are no good against real reads.


The whole idea of this topic is when you don’t pay all that much attention to the tables you play, when does it start paying off to replay the hands you have on your opponents as opposed to just following the stats PAHud or PT give you?
When I’m multitabling (mostly 4 tables), there are plenty of times when I’ve folded all my hands and then I’m actually following one of the tables and can get a real read for a change. But often my attention gets called elsewhere (new cards dealt at another table – checking the lobby to find better tables or to see if my tables are still good – checking PT for a bit more intel on one or two players – etc.) and because of this I’m really low on real reads on my opponents.

That’s where I think the replaying of opponents’ hands comes in: in doing that I get a better grasp on the way they play and can notice some betting patterns that can get taken advantage of…
But I don’t want to do this for every person I’ve played, that’ll take me WAY too much time, so I restrict myself to the ones that I’m almost certainly going to meet in the future, the ones who I’ve got lots of hands on.

But is that a good way to go about it? Is replaying opponents’ hands a good way to get reads when you don’t get those from the table itself? Or is there another way (in-depth analysis of some PT stats could hold some stuff that is golden – Harv mentioned a great “% won SD when raised turn”)?


Thanks for the replies at least, some good reading in it! ;-)

[TimovieMan]

[ QUOTE ]
EDIT TO ADD: Stats also fail to recognize the phenomenon of tilt. A TAG can easily tilt into a LAGgy range at times. If you start recording stats during the 50 hands at which the TAG is tilting, then attempt to use those stats in a later session--well, it probably won't work out too well. In a similar vein, if all your stats were on the TAG when he or she wasn't tilting, those stats will mean nothing once the TAG goes on tilt. This goes back to Bellatrix's point about reads being more important than stats; if you're paying attention and get a read that the TAG is currently playing differently than his/her stats would suggest, you can adjust your play accordingly. You can't do that with stats alone, though.

[/ QUOTE ]
Tilting opponents are one of the things you're likely going to miss when multitabling, that's one of the many reasons why multitabling is -EV/table when compared to singletabling. But isn't this also something that can get noticed when replaying hands?

As a sidenote, even when multitabling you can still notice some players suddenly cursing in the chat about bad plays others make, and that's when you should start to look in detail at what they're doing.
I've noticed on several occasions that LAGfish who experience two or three bad hands in a row (badly beaten by a nit with 83o rivering a miracle boat and the likes) and that start insulting other fish are either already tilting or about to tilt, so their next hands should receive some special notice.
Stats do tend to get screwy when a maniac suddenly turns into a rock... [img]/images/graemlins/smile.gif[/img]