PT Auto-Rate 6 max formulas

[B00T]

I rarely use the icons, and just go by PT #'s but I figured I might as well update the icons since they are being portrayed anyway...

Was there ever a discussion or formula posted to autorate players based on 6max and not full ring?

I skimmed over the 6max FAQ type post but did not see anything covered there.

Thanks a bunch.

[bonanz]

there's this thread (http://forumserver.twoplustwo.com/showflat.php?Cat=&Board=headsup&Number=1060670&fpa rt=1&PHPSESSID=)

[naphand]

Just wondering?

Are you intending to just copy the formulas? Seems to me if do not understand how to set these up (i.e. how to categorise people) how are you supposed to understand what they mean? The setting up of your own formulas is an exercise in understanding your opponents, and a very useful one. Copying other peoples work teaches you NOTHING.

[cab4656]

[ QUOTE ]
Just wondering?

Are you intending to just copy the formulas? Seems to me if do not understand how to set these up (i.e. how to categorise people) how are you supposed to understand what they mean? The setting up of your own formulas is an exercise in understanding your opponents, and a very useful one. Copying other peoples work teaches you NOTHING.

[/ QUOTE ]

Why ever read a poker book then? It's someone else's work. It teaches you NOTHING.

Everyone knows that someone with a 50 VPIP is loose and that someone with a 10 VPIP is a rock. Manually entering your own autorate rules is a lot of work. I appreciate someone posting them and intend to take advantage of it.

[naphand]

[ QUOTE ]
Why ever read a poker book then? It's someone else's work. It teaches you NOTHING.

[/ QUOTE ]

That has to be the stupidest post I have seen this week. You not only do not understand the benefit of working through what V$IP etc. means but you also make a completely inappropriate comparison. Books teach theory and can give examples. Copying someone else's test answers instead of studying books amounts to nothing. If you cannot see the difference between these things then no wonder you just want to take shortcuts. Sooner or later you will learn this is not possible in poker.

[ QUOTE ]
Everyone knows that someone with a 50 VPIP is loose and that someone with a 10 VPIP is a rock.

[/ QUOTE ]

And you believe this sums up the understanding that is required to appreciate player groups? So you are now saying that a player with V$IP 10% PFR % is the same as a V$IP 10% PFR 10%. Simpleton is a word not big enough to describe this kind of thinking.

[ QUOTE ]
Manually entering your own autorate rules is a lot of work.

[/ QUOTE ]

Is it? Only if you don't understand what you are looking for. If you don't understand what you are looking for, the categories will mean nothing to you. Welcome to the bot-generation.

[B00T]

Then I guess the people on the other thread are cheaters and bot people as well.

I understand what they mean. Its like having an assignment of having to write the phrase "I want to win the WSOP" over and over until it fills 10 pages. I can either continue typing it over and over or just spam CTRL+V until the page fills.

Obviously I will see the formulas that make up each one. I can apply them to what I want, if I couldnt I wouldnt be asking. Being able to take a screenshot and looking cool by having little icons and elephants on my computer means nothing to me.

Like I wrote in my first post, if you read it, I dont even use them, but since I have them popping up they might as well be somewhat accurate and not that of 10 handed games.

[cab4656]

I do not want to simply look at little autorate pictures and make all my player reads from them. I understand exactly what I am looking for. I just do not want to sit there and go click, click, click, click x 10000 to configure all the settings.

I don't know why you decided to make a bunch of silly assumptions about the way I think, but that's your business.

[naphand]

I don't believe I used the word "cheater" at all. Lazy scroungers is the phrase I am most likely to use when pushed for one... /images/graemlins/mad.gif

[ QUOTE ]
Its like having an assignment of having to write the phrase "I want to win the WSOP" over and over until it fills 10 pages.

[/ QUOTE ]

No it is not, you need to balance the formulas to ensure they are both meaningful and leave no gaps. This is an excellent exercise in understanding what the categories mean and how they interact with each other. Copying and pasting is what you do when you are importing the formulas, or physically when you copy them from threads on the forum. By creating them yourself you have to engage brain, which is really what you have to do when playing. By spending time off the tables creating formulas than mean something wrt your own personal understanding of your opponents, you will be able to use them more effectively, done properly you will understand the typical hand types they play and what to look for post-flop to refine what PT is telling you.

Take a short-cut using an incomplete map, and miss the sign that says "Danger Bridge Unsafe"...

[naphand]

[ QUOTE ]
I understand exactly what I am looking for.

[/ QUOTE ]

Yes, a shortcut. There isn't one. Do you even understand what you said in your post? Re-read the garbage you wrote about comparing copying to reading, and ask yourself if it makes sense (Hint: it doesn't, it just looks dumb).

Your comments subsequent to my post demonstrate a remarkably superficial understanding, as they are not consistent with your stated purpose. Let's look at the most obvious:

(1) You don't want use the "little autorate pictures" but you do want to categorise the players.

The icons are just indicators of those categories, "not using the icons" means, quite literally, "not using the categories". So why do you want the autorate rules?

(2) You do not want to sit there and go click, click...

Well this is a description of working on a computer, and playing on-line is pretty much this. If you are too lazy to take the time to configure PT to give you meaningful classifications, then you either (i) do not have an understanding of what those classifications mean and/or (ii) are too lazy to play and learn poker. So what do you mean by "I understand exactly what I am looking for"? You appear not to understand the classifications, nor have the mental resolve to learn.

(3) You accuse me of making "silly assumptions" about the way you think, yet you cannot name those assumptions.

When you use absolute statements such as "Why ever..." and "Everybody knows..." what you are really saying it "Why bother..." and "I don't know...". This is mental reasoning of the school-playground variety.

In two weeks you will be whining on the forum about coping with playing AKo against a LA-A etc. but because you have not understood how to play the players, you won't know the difference between a LA-A and a LP-A, and so on...

The only people I have known who use the word "silly" have been also been incapable of descriptive argument. Do you not see the irony of this?

[Grisgra]

Nap, I think maybe you need more fiber in your diet, or something /images/graemlins/smile.gif. Or did someone run over your dog?

[naphand]

I ate the dog when I ran out of bran...then I took to chewing on trees... /images/graemlins/ooo.gif

[B00T]

Do you just like to ignore what I write which is the MAIN REASONING of my post?

I HAVE ICONS UP NOW OF PLAYERS AUTORATED BASED ON TEN HANDED GAMES...

OK I tried capitals this time, maybe you'll get it. I am not aruging your point about learning them, but if I am going to have them sit there, I'd prefer them to at least be relevant to the game I am playing.

3rd times a charm, maybe you'll get it this time.

[naphand]

So if you have the 10-handed icons, HOW HARD IS IT TO ALTER THEM FOR 6-HANDED? There I put it in capitals for you.

[ QUOTE ]
I'd prefer them to at least be relevant to the game I am playing.[/i]

The same principles apply, copying is not learning, not learning is not understanding, not understanding is losing. Clear yet? If you have a partial understanding of the classifications, or a full one for 10-handed, then your job (of adapting) is even easier.

[B00T]

just for reference...

1st post...[ QUOTE ]
since they are being portrayed anyway...

[/ QUOTE ]

2nd post [ QUOTE ]
Like I wrote in my first post, if you read it, I dont even use them, but since I have them popping up they might as well be somewhat accurate and not that of 10 handed games.

[/ QUOTE ]

Cool, jah??

[B00T]

I cannot believe you are arguing this point. Honestly, what the hell do you care what I do?

[naphand]

[ QUOTE ]
I dont even use them

[/ QUOTE ]

So why do you need auto-rate rules?

[cab4656]

Fine. I want a shortcut. Call it what you want. I'd simply like to better organize my database. I do not want to see a moneybag icon and then automatically play every hand the same way vs. a moneybag icon.

Can you at least assume that I have a little bit of intelligence? Just a little? Because I don't see how you feel the need to flame everything I post on this thread and pick apart my every word.

I appreciate your persistence and willingness to help though.

[naphand]

I don't care what you do, what is important is that people reading this forum understand that importing PT rules is not going to improve their results. Understanding opponents will, and this can be refined somewhat by using PT. Those reading the forum are influenced by what they see, they might actually believe that importing rules into PT is an important step in their quest for poker profits. However, the rules that apply to the auto-rate feature are arbitrary, they are not the rules of poker, they are not the key to correct play. PT is a tool that helps in the development of accurate reads, and the exercise of creating those rules based on what we understand is a far more useful process than applying them in the large majority of cases (I am talking here about game-play, table-selection is another matter entirely).

My initial response to your post was just meant as a reminder of this fact. cab4656 piped up with a mind-numbingly stupid response I felt it was necessary to explain to forum members why such thinking (his, not yours) was deeply flawed.

I am not alone in the opinion that the forum of late has become obsessed with PT stats and auto-rate rules (yet it is clear that many who post these do not understand the basis of mathematical significance). "Reads" for hands now are more likely to consist of 58.6/12.4/1.8 than anything that refers directly to the plays observed. As a direct consequence the standard of discussion has fallen, as people look for forumula responses, the same people I would bet, who would be first in-line to buy the next bot that claims to be able to beat the Party $1/$2 game....

If people just want/get bot-responses then this forum is in terminal decline. I would rather talk about players, their plays and adjustments required based on the cards before me. This leads to a deeper understanding of the game and our opponents, which will lead to other statistics so-drooled over, the 3BB+/100 WR. Bot-poker, standard responses, opponents-as-numbers? No thanks. That is the point. I want this forum to be useful to all (including me) and not decline into a colour-by-numbers copy-and-paste wasteland, hardly worth the bother of perusing the 50-60 daily posts for 1 or 2 weekly gems. There are reasons why players such as Schneids, stripsqueeze, El Diablo, Peter_Rus et. al. are so well respected. How many times have you seen these people post reads based on PT stats? Why do you think that is so?

[naphand]

I don't assume intelligence in another poker player, I base my reads on what I see they are capable of... /images/graemlins/grin.gif

[Schneids]

Wow you more or less did half of a rant I've been very close to making in the general forum (about PT and Playerview). I will just state that for multiple reasons I wish so much that PT and Playerview had never been invented.

[B00T]

I played fine without the Playerview stuff. It just makes general reads easier (especially when multi-tabling). It just waves red flags on spotting the very people who feed us day in and day out.

I would like to see a little fishy icon pop up where it reminds me to pay attention to him and develop an independent read on him so I can make my own player notes on my own observations of him.

Human reads are stronger than statistical data. I use the statistical data as a means of filtering through and prioritizing who I should focus on out of 40 opponents at once.

I agree with you on the comments about PT data not being the key to poker. I will not automatically 3 bet some clown with 60VPIP just because I have A9. I do think just having an extra tool in your arsenal to use at one's personal discretion is beneficial.

If I use the icons to help setup a plan on who to read by just C+P them from a generic text to make and it increases my winrate by .06/100 I'll be happy.

It is by all means not a solution but for what I want them for, a C+P copy is fine.

[naphand]

The rules you are using are based on inaccurate and arbitrary assignations; such as 25%, 1%, 50%. Are these figures significant? How are they derived? How do we know they are not losing you money? We can be pretty sure thay are not close to optimal, and they are severely limited in what they tell us.

This thread

How I Do Against 10 Player Types (http://forumserver.twoplustwo.com/showthreaded.php?Cat=&Number=1718210&page=&view=&s b=5&o=)

has started a more useful discussion, one which involves studying which players are providing $$ and which are costing. This is where PT will really earn money. Note how highly personal these figures may be, and based on what? Classifications that are arbitrary! How reliable are they? But at least this thread is addressing how our personal game is performing and may provide useful clues to game development.

I have posted elsewhere about attempting to more accurately define player groups by datamining and assigning players according to classifications based on how profitable they are generally. This is still ongoing but will result in player groups that are fast-coverging and others that are slow-converging (so will split after n hands. The main purpose of this exercise is to aid table selection, not gameplay.

The classifications you are so keen to employ use arbitrarily assigned values (e.g. why is figure A 25%, perhaps it is in reality 22%, or perhaps it is 22% when combined with figure B and 27% with figure C) and you do not know if they are accurate, you also do not know they are providing you with reliable information (on one table with <100 hands I had V$IP 7%, the another 121 hands with WenttoSD 49% etc.). As such you should consider them to be experimental.

I agree that it is nice to see a player with V$IP 57%, perhaps we had not noticed how loose he was. We also don't know what hands he got or how he plays them. Putting this player into the "fish" category could easily cost you bets. Don't try to make up for weak player-reading skills with PT "reads".

You cannot quantify if your Copy & Paste Rules are helping you until you start the kind of exercise in the thread given above. And that kind of exercise requires rules that are meaningful, or you will find yourself on another dead-end-street.

[ QUOTE ]
I do think just having an extra tool in your arsenal to use at one's personal discretion is beneficial

[/ QUOTE ]

Just how useful is a radar to a mediaeval knight? First you need to understand what you have got, as at the moment you are using your $1million rader as a club. "Yeah it's a great club". Yes, but the radar is used launch guided missiles at your opponents ships....

PT is useful as a way to analyse our own game and secondly as a loose method to label other players. Note that we must understand what our own PT figures mean and their significance before we can hope to understand what it means for our opponents. Why do you not just look at the stats for a player and assign a relevant icon (even during game-play)? Assign the icon, make notes. This would provide just as much useful information at-a-glance as trying to set up auto-rate rules based on arbitrary values. Ah yes, with one proviso: you must be capable of meaningful observation at the tables. And that is a skill worth developing, and one which ultimately makes PT redundant...

[ QUOTE ]

Human reads are stronger than statistical data.

[/ QUOTE ]

You win the prize for understatement.

[ QUOTE ]
I use the statistical data as a means of filtering through and prioritizing who I should focus on out of 40 opponents at once.

[/ QUOTE ]

So why do you need auto-rated rules? The answer is; you don't. And they are unreliable, arbitrary and near-useless without accompanying notes. Yet that is how they are used, look at the posts with PT "reads" based on a set of 3 numbers. Very, very sad.

[Nate tha' Great]

You and Schneids are barking up the wrong tree. Pokertracker and Playerview are EXTREMELY useful resouces. I wasn't good about using them until this year and I wouldn't be surprised if they've added something like 0.30 BB/table hour to my win rate. I don't get it. If it's a crutch, it's a damn good crutch.

[cjx]

[ QUOTE ]
Why do you not just look at the stats for a player and assign a relevant icon (even during game-play)? Assign the icon, make notes.

[/ QUOTE ]

Because I'm lazy and find the pictures to be nearly meaningless.

Quote from the future:
[ QUOTE ]
Then why autorate the players if you aren't using the information?

[/ QUOTE ]

Because I like pictures.

[ QUOTE ]
Quote:

Human reads are stronger than statistical data.



You win the prize for understatement.

[/ QUOTE ]

I disagree. As the sample size approaches infinity the mathmetical read will approach 100% accuracy. Granted parsing the information and then breaking down the possible range of hands would be extremely tough on the spot as there are a variety of variables (which can all be definied as poker is a finite game even if some situation occur rarely). I guess I'm arguing for argument's sake though.

cjx just thinking out loud

[Grisgra]

[ QUOTE ]
[ QUOTE ]
Quote:

Human reads are stronger than statistical data.

You win the prize for understatement.

[/ QUOTE ]

I disagree. As the sample size approaches infinity the mathmetical read will approach 100% accuracy. Granted parsing the information and then breaking down the possible range of hands would be extremely tough on the spot as there are a variety of variables (which can all be definied as poker is a finite game even if some situation occur rarely). I guess I'm arguing for argument's sake though.


[/ QUOTE ]

Yeah, you're being a dork. Just because someone has an aggression stat of 0.8 doesn't mean they don't bluff. I've been surprised at how many times I've caught a loose-passive-passive player bluffing, sometimes all the way to the river. PT only goes so far.

That doesn't mean nap isn't being a bit of a doof, of course /images/graemlins/smile.gif.

[J.R.]

not only are you acting like a thick headed jerk but many of your assumptions and generalizations are outlandish and wrong.


I am not sure why you have chosen to annoint yourself god but:

[ QUOTE ]
Bot-poker, standard responses, opponents-as-numbers? No thanks. That is the point. I want this forum to be useful to all (including me) and not decline into a colour-by-numbers copy-and-paste wasteland, hardly worth the bother of perusing the 50-60 daily posts for 1 or 2 weekly gems.

[/ QUOTE ]

Sorry if you refuse to be open minded and explore other ideas, leave if you like. Some people dervive value from that which you do not chose to understand and condescendingly cast aside. The forums are not just for you, don't bother reading pt oreinted posts you don't care for.

Its obvious how you feel, let us run towards certain doom and failure like the leemings you seem to contend we are. And thanks for making it clear its of no value to read any pt oreinted post, I wasted so much time doing that that my win-rate has actually improved.



e.g.,

[ QUOTE ]
what is important is that people reading this forum understand that importing PT rules is not going to improve their results. Understanding opponents will

[/ QUOTE ]

thanks, I am now glad to know these things have been deemed mutually exclusive, I must have been mistaken. How do you know importing pt results is not going to improve anyone's win rate? will auto-rating make one's winrate decline decline? will it have no effect? I am sure you have lots of ancedotal evidence to support your position.

It must hurt my winrate to sit down and have know an opponent I have never seens before has raised preflop 25% of the time over a 1500 hands sample of his play, or that i can quickly glance at an opponent and know that over a minimum number of hands at a given limit their vpip, pfr and agg all fall within certain ranges. That must be meaningless, and I am sure you have lots of experience to back up that claim. Cause I kinda dissent from that viewpoint.

[ QUOTE ]
I would rather talk about players, their plays and adjustments required based on the cards before me.

[/ QUOTE ]

then do it and stfu about other sh!t. There is certainly some truth to what you are saying, but it gets lost in your unfair and overbroad characterizations of pt, auto-rating, and the intelligence and poker skill of those who use these tools.

I suppose I must be an idiot for auto-rating, its too bad that I try to take advantage of all of the information I can gather prior to making decisions. I guess I wasn't aware that auto-rating precludes me from watching my opponents and oberserving their play as well, I wonder what I have been doing for all this time. I guess I can't learn anything from reviewing soemone else's auto-rate rules, or learn something by employing those rules and then refining them as a I play.

Pt can be a crutch, and there can be negatives associated with using a crutch. But pigheadly refusing to use the tools available to you and criticizing people you don't know for relying on those tools in a fashion you do not know or understand is silly. There may even be some positives associated w/ auto-rating in specfiic and the use of pt stats in general. To each his own, I don't rail on you for not using something I find of value.

I enjoy many of your posts, and certainly hope you continue to grace us with your presence. Stay constructive and get off your high horse.

[Schneids]

Nate notice I never said anything about their usefulness. I would agree that using them correctly adds to your WR.

[naphand]

[ QUOTE ]
but many of your assumptions and generalizations are outlandish and wrong.

[/ QUOTE ]

Name them.

[ QUOTE ]
you refuse to be open minded and explore other ideas

[/ QUOTE ]

Do I? Where have I refused in my post? I seem to remember directing people to a post that explored the use of PT more deeply. Hardly the actions of someone who refuses to use it. In the process of writing your response you also ignored my reference to another post I have made talking about data-mining to better define players with rules that represent actual percentages, not numbers picked form the air because they are nice and cosy (50%, 10% etc.).

[ QUOTE ]
that which you do not chose to understand...

[/ QUOTE ]

Far from me "not understanding", I seek to use PT in more meaningful ways than the utterly incorrect way seen so often on this forum. Example "Opponent was a 56.1/2.6/0.5 player, but only 39 hands..." You are suggesting that this sort of thing is anywhere near correct and reliable? You are in no position to label others condescending and ignorant.

[ QUOTE ]
and condescendingly cast aside.

[/ QUOTE ]

So I "cast aside" PT do I? That is a very ignorant statement. I use PT every time I play. Had this occurred to you? Obviously not...you can now feel a silly as you want.

[ QUOTE ]

thanks, I am now glad to know these things have been deemed mutually exclusive

[/ QUOTE ]

Did I say they were mutually exclusive? I don't remember saying that at all, hang on, I am remembering something, I think it was...

[ QUOTE ]
By spending time off the tables creating formulas than mean something wrt your own personal understanding of your opponents, you will be able to use them more effectively, done properly you will understand the typical hand types they play and what to look for post-flop to refine what PT is telling you.

[/ QUOTE ]

and...

[ QUOTE ]
you need to balance the formulas to ensure they are both meaningful and leave no gaps.

[/ QUOTE ]

and...

[ QUOTE ]
Understanding opponents...can be refined somewhat by using PT.

[/ QUOTE ]

I'm sorry, but did you read this thread? I am saying PT is of no use here am I?

And of course you continue to liberally misunderstand/misquote me...

[ QUOTE ]
And thanks for making it clear its of no value to read any pt oreinted posts

[/ QUOTE ]

No, what I was saying that posts that start "Opponent is 56.1/7.2/1.5" are really not giving a useful read as far as very specific hands go. People make posts about very specific hands and some examples of how opponents play those cards/boards is far, far more relevant than their overall V$IP or Aggression. Do you have a hard job understanding this?

[ QUOTE ]
your unfair and overbroad characterizations of pt

[/ QUOTE ]

The point I am making is, it is PT that makes overbroad characterisations. I'm not sure sure how that missed you.

[ QUOTE ]
I guess I wasn't aware that auto-rating precludes me from watching my opponents and oberserving their play as well

[/ QUOTE ]

Man, now you really are getting desperate. I never said anything of the sort. In fact I said the opposite, re-read the lines above about using PT to refine your reads. Maybe see the optician about some reading glasses while you are at it.

[ QUOTE ]
I guess I can't learn anything from reviewing soemone else's auto-rate rules

[/ QUOTE ]

You know, I never said this either. I am all for people reviewing the auto-rate rules, and understanding them. Did I not make this point when I said...

[ QUOTE ]
The setting up of your own formulas is an exercise in understanding your opponents, and a very useful one.

[/ QUOTE ]

and...

[ QUOTE ]
By spending time off the tables creating formulas than mean something wrt your own personal understanding of your opponents, you will be able to use them more effectively...

[/ QUOTE ]

Perhaps your take on these lines is that I am telling people to stop using PT? but I cannot see how you could see it that way, unless you need new spectacles...or did not read the thread properly.

[ QUOTE ]
pigheadly refusing to use the tools available to you

[/ QUOTE ]

See above, I use PT every day I play. Wrong again.

[ QUOTE ]
criticizing people you don't know

[/ QUOTE ]

I don't know anyone on this forum, does that mean I am not allowed to criticise? Interesting take on the use of Forums in general.

[ QUOTE ]
I don't rail on you for not using something I find of value.

[/ QUOTE ]

I think it is clear by now, that you don't know what I use.

[ QUOTE ]
Stay constructive and get off your high horse.

[/ QUOTE ]

No I like my high horse, particularly if it gets a few people to bother to develop the thinking behind PT and its use. Rather than posting "where are the auto-rate rules" why are there not more threads on what the numbers actually mean? There have been few; so far the discussion produced has pretty much resulted in the exposure of just how limited those figures are when it comes to making reads. I believe the Aggression Factor was exposed as one weak area, in that similar figures could be obtained from completely different kinds of play (you will need to search back a month or two for this).

[ QUOTE ]
I am not sure why you have chosen to annoint yourself god...

[/ QUOTE ]

I have not, I just am not prepared to accept BS answers. I am quite happy for people to question my own argument and logic, it one very quick way to learn. Sadly, most people just react like you, and lash out without even trying to understand the original posts, the intention behind them, and instead make...

[ QUOTE ]
assumptions and generalizations (that) are outlandish and wrong

[/ QUOTE ]

Such as categorically stating I do not use PT... /images/graemlins/laugh.gif

[naphand]

[ QUOTE ]
As the sample size approaches infinity the mathmetical read will approach 100% accuracy.

[/ QUOTE ]

This is not how people are using PT, they are posting the stats from as few as 20 hands against a player, and calling them "reads".

Large sample sizes, yeah great but most of the posts here involve hands against players that are new to the poster.

I am also asking the question "WHAT" approaches 100% accuracy? Do you understand what is becoming more accurate? The definitions become more accurate, which is not much use when they themselves are "broad generalisations".

[Danenania]

Call me lazy or a schmuck or whatever for using every resource available to me (that's within the rules). I'll be the schmuck with a higher win rate (and less bushido pride).

[naphand]

I am not questioning the use of PT per se, just its application to a lot of situations. If someone has 1500 data-mined hands on a player, then that is a lot of information. This is not the same situation as 20 hands on a player.

The auto-rate rules ARE very arbitrary at present. Considering how much datamining has gone one I am surprised that no-one yet has proffered classifications based on actual profitability or 80:20 type rules (i.e. the 20% tightest players etc.) with corresponding BB/100 WR information.

I am also making the point that not understanding the auto-rate rules and what they mean is a significant handicap to maximising the additional profitability possible using something like PT. I use it all the time, for example, to determine the likely hand types of a particularly aggressive player from their PFR, how often a player steals etc. The recent post on "Ten player Types" is a post in the right direction, as it highlights new ways of learning from PT.

[Danenania]

I think there is a TON of useful information that can be gained from 20 hands worth of stats on someone. It isn't the same as having 500 hands obviously, but it's still very very nice compared to no read at all.

[Nate tha' Great]

[ QUOTE ]
Far from me "not understanding", I seek to use PT in more meaningful ways than the utterly incorrect way seen so often on this forum. Example "Opponent was a 56.1/2.6/0.5 player, but only 39 hands..." You are suggesting that this sort of thing is anywhere near correct and reliable?

[/ QUOTE ]

Hell yes, that's reliable. It's *extremely* unlikely that the player is anything other than very loose and very passive. Complicated stats like win rates do not converge quickly but simple ones like VPIP do, especially in a 6-handed game.

[B00T]

wow, for the first time ever a "big name poster" is actually somewhat taking my side...It must because its Valentine's Day. My gf is gonna get it extra good tonite. I will be sure to yell out WHATS YOUR VPIP MOUSE, while her legs are up in the air to continue quizzing myself and keep myself on my toes.

All kidding aside, all points are made to and for it. I got my chart, I am happy.

/images/graemlins/cool.gif

[helpmeout]

You should spend more time playing/studying the game instead of wasting time arguing with people.

People who want things done for them eventually fail, but really how is it your problem?

[Scotch78]

God damn you Naphand. You're such an insensitive ##### sometimes. I can't believe you'd throw a party like this without inviting me. I mean, honestly, is there any flaming left for me in this thread?

I guess I'll just have to post a serious response then. Damn, what's the world coming to?

I don't know how to use Excel, so right now I can only do this for VP$IP and PFR, but I think WSD and AF (excluding pre-flop) would be useful as well. Figure out what VP$IP and PFR percentages cover the highest and lowest 5% and 20% of players. I designate players in the highest 5% VP$IP and lowest 5% PFR as calling stations. LPs are in the corresponding 20% ranges (note: an LP could be in the 5% category for one stat). Maniacs are highest 5% for both numbers, LAGS are in the highest 20% for both. TAGs are in the corresponding 5% categories.

I only use VP$IP, PFR, WSD, and AF-Total in Playerview, and the first two are color coded to the percentage ranges (lowest 5%=red, highest 5%=green, etc.). Since this information is much more specific than the icons, I do not use the latter anymore. I have AF-F, -T and -R under the hotspots for quick access, and once I figure out Excel all of this stuff will be analyzed and color coded as well.

Currently, the only notes I export are hands raised, including frequency.

Unfortunately, I haven't been using this method long enough to say whether the five and twenty percent designations are optimal/useful.

Now, how do I use this info? Naturally, I adjust for looseness, aggression, etc., but I also use the stats as a scouting report. When I see someone in a 5% category for anything, I watch that part of his play extra carefully. Normal play is normal play, but someone on the fringes is more likely to do something unusual that I can pick up on and exploit.

For example, I recently posted a thread where I raised a calling station on the turn with just top pair and felt like I should've raised the river (it paired the turn card) as well when he bet into me again. Normally I would fold the turn instantly, but because PT+PV had brought my attention to how ridiculously passive he was, I had noticed him checking and calling with some strong hands (two pair, trips, etc.). There weren't any straights or flushes possible, so I figured he either turned a set or was taking a shot at me, with the latter being much more likely.

PT couldn't tell me that he had fishy betting standards rather than tight standards, but without PT and PV I would've been watching the whole table equally rather than focusing on this player's (non-)betting patterns, and I probably would've folded the best hand.

So far, this system is working very well for me, but it does leave the icons unused. A while back kiddo posted something about using the icons for non-statistical designations, i.e. habitual bluffers. I'm in the process of moving up a level, but once I'm settled I will probably try to develop a similar method. If anyone has ideas, or if you know how to do what I want in Excel, please chime in.

Okay, I feel dirty after typing something with substance, so that's all. If the original poster hasn't been scared away yet, I hope this helps.

Scott

[Scotch78]

[ QUOTE ]
My gf is gonna get it extra good tonite. I will be sure to yell out WHATS YOUR VPIP MOUSE, while her legs are up in the air

[/ QUOTE ]

POTD

Scott

[naphand]

It is not reliable. As Scotch78 points out, early data from PT can be used to flag only. You do not put someone on a hard read based on 20-30 hands, but that is what people seem to think they can do. And I have not even mentioned how utterly wrong it is to quote V$IP to 1 d.p. when you have a a sample size too small to quote reliably to +/-5%.

People are quoting figures that are meaningless. PT is a program, it churns out figures to 2 d.p. because that is how it is configured, this does not mean these figures are reliable. They are not, and certainly not to the degree of accuracy claimed.

Example. You have 60 hands with a player, PT says V$IP 6.7%. What does this mean? Well I can tell you that is exactly what I had on one of the tables I played this week, yet my V$IP is more like 23% 6-handed, and this was a 5-handed table where it is more likely to be 25-28%. Just how reliable is that? All that can be gleaned from such small samples is a that the player is probably tight, but given my V$IP is usually +20% from this figure, perhaps people might begin to understand how imprecise samples of this size are.

Of course, people throw extremes back at you, "yeah well if he plays 70% and raises 0% that's a pretty good read". Well, how good is that read? Poker is often about pushing small edges, and if your "read" is +/-20% your edge may be swamped in the statistical variation. While it is understandable that people sing praises of PT, it is quite frankly stunning how they also do not bother to understand the numbers they are looking at. Let's think about it:

(1) Who defined what %ages make a LAG?
(2) Why are the numbers you use(50%, 10%) the defining characteristic? Why are they descriptive, moreso than 46% and 12.5%?
(3) Are these numbers based on what has been discussed on the Forum? or the overall profitability of your sample?
(4) How do the numbers vary wrt to AF and PFR? For example, How proitable is a player V$IP 30% PFR 6% compared to V$IP 29% PFR 20%? or V$IP 12% PFR 6%?

Labelling players over typical sample sizes for a session with no data-mined opponents, can only ever be very approximate. I often hear about numbers "converging", a popular phrase, but where is the math the describe this convergence? Where is the evidence?

Considering how powerful a tool PT can be, why are you guys only using 5% of it's capacity? Are you serious about your poker? Look at the number of posts concerning plays like "Overcards on a raggy flop", this kind of thing crops up again and again. It's one hand. Look at he posts using PT "reads". How many posts deal with refining the statistical data and what it means?

I intend to start another thread for those willing to actually dicsuss this and contribute their own research. I never intend to release the rules I derive from this except to those people, for the reasons given and also I fail to see why a bunch of scroungers, who will never contribute meaningfully to the discussion, should benefit. The rules will also be highly personal, albeit based on math and data-mined samples, and I have already said why copying them is meaningless.

There is one statistic I have mentioned previously (I think) that has hardly been discussed, and it points out where the $$ are. Here it is...

For $5/$10, the average loss by all players is 2.44 BB/100. This is the rake taken by Party. The average number of hands played in the sample I have is 91. Filter for at least 50 hands played (3506) and the loss falls to 0.85 BB/100. What does this mean for us? Does it mean that most of the profit made on Party is from players who play less than 50 hands? Think about it. What implications does this have? How many of the regular players are making $$ (i.e. the ones with more hands tend to be winners) and how much adjustment must we make for this fact? Well, 46% of the 50+ sample were winners, 40% of the full sample were. That is a 6% swing over a very large sample. Does this mean that the very people who provide most of the profits on Party are not playing enough hands for PT to define them with any accuracy? By the time players reach 100 hands they are almost break-even. BUT OF COURSE only the entire sample is break-even, breaking it down further reveals differences, but hw are those differences defined mathematically is a far stickier problem than most here seem to understand.

The significance of these results needs to be determined, as do the implications for using PT to flag or define players. Yet precious little intelligent discussion appears on this forum regarding this. Why not?

[naphand]

What a bizarre post. You appear to argue the case for not using the forum at all.

[ QUOTE ]
how is it your problem?

[/ QUOTE ]

It is an issue (for myself) if I choose to make it one. Why is your thinking so superficial? Can you not even begin to see where a discussion such as this may lead?

After all, how is it my problem if a new poster has trouble playing overcards? How is it my problem is Nemesis is on a big downswing? How is it my problem that no-one is prepared to develop an understading of PT and what it means for table-selection and opponent reading? Ah well, now you're talking...

[naphand]

I am sure Nate is really delighted that his posts give you a hard-on... /images/graemlins/laugh.gif

[Mr. Graff]

[ QUOTE ]

For $5/$10, the average loss by all players is 2.44 BB/100.

[/ QUOTE ]
Actually this number (the rake) has dropped to 2.03 BB/100 in my database since January. Anyone else notice this?

[naphand]

Yes I did notice it, and I figured someone would pick up on this (I just took it straight from my DB) and it occurred after an update. I wonder if they have changed the rake? That is a huge drop in rake, 25%! Unfortunately most of the data-mined hands I have were from before the New Year so I guess I will have to get busy again. This should not affect the player classifications, as they will be just be skewed upwards. I think it is reasonably safe to assume 2.03 = 2.00 or is there an explanation for the 0.03?