Stats And Leaks: Study Guide

[jmbreslin 0]

I'm approaching my first 10k hands at 10NL Rush (full ring) and wanted to spend some time really studying my stats to identify some leaks. What I'd like to do is dig down beyond the obvious so I can identify aspects of my game I need to work on. I'd like your input on developing such a study guide, which might be very helpful for others who want to examine their play in depth.To start the ball rolling, here are a few ideas that spring to mind (we can fill in the details as we go):1) Positional awareness - VPIP should be low in EP/SB and increase toward Button.2) Aggressiveness - Small gap between VPIP/PFR3) Continuation bets - Not too often, not too infrequent but just right4) W$SD - Should be >50% - if it's not, look at WTSD to see if you're going to showdown too often5) W$WSF - Should be, what, >35%? Too low indicates you're being pushed off hands6) Flop AF and Turn AF - Too much of a gap indicates giving up on the turn too often7) Review big pots lost - Are you making bad calls in big pots? Using semi-bluffs recklessly? Getting it in badly too often?Keep it going...

[Ninja Ace 1]

one more stat you should check is your RPTAP%that's rush poker to actual poker percentage

[Ninja Ace 1]

Ok... you're sort of asking for something legit and I can appreciate it... but honestly I think you're looking at the wrong stats... those stats are really only meaningful once you hit 100k hands.The VPIP and PFR thing, in general, is correct with around a 2% gap overall... but they're much more meaningful when you look at the positional chart... UTG (FR anyhow) should be something like 5/5 while the button should be around 25/20I'm kind of mad there's no 3bet stat by position bc that would be meaningful... but you've also got to look at your 3bet % and determine if you're re-stealing frequently enough. I tend to 3-bet CO and Button raises with any hand that's equivalent or greater than their open range.Also you should check your aggression by street. In the early streets it should be incredibly high, and drop down as the streets progress. I think mine is PF 3.5 F3 T1.5 R 1.2 or something to that effect. So I guess I kind of disagree with what you say about the gap bc you should start going into check-calling mode on the turn with one pair hands very often... let the other TAGs try to float you, and don't worry about letting them draw for free bc a lot of them won't be able to resist the passive flop draw turned into a float line. Also, it makes it really easy to fold a one pair hand to a river barrel bc it's rare that you find players capable of bluffing both the turn and the river in these situations.Also, a majority of players (with any talent whatsoever) don't have major leaks in big pots... a lot of their leaks come from blowing small pots.

[trystero 0]

BEGIN DERAIL:Well I'm here to totally derail this thread. Here's the best tip I can give you, and it's got nothing to do with PT: don't play Rush. You want to improve at poker. You've started a blog that's going to chronicle your ascension through the riggedness that is Full Tilt. The only way that you're going to improve is to start playing exploitative poker and that's feasible in nonRush games. I mean, seriously. I can't say it enough. Like here, you want to talk generally about what numbers one should try to attain for different poker statistics. That you're only operating on half of the information (nothing on your opponents) means that you're not able to take full advantage of what a poker database offers you. What's the point in knowing your 3-bet % if you don't know your opponent's fold to 3-bet% and PFR? What's the point of knowing your button steal % if you don't know your opponent's VPIP, fold to steal %, and 3-bet %? What's the point in knowing your c/bet % if you don't know your opponent's fold to c/bet % and fold to turn bet %? What's it mean to say "not too much, just enough"? 60% 70%? You should be firing against the right kinds of opponents on the right kinds of boards. Focusing on the statistic alone, like focusing on the redline, is tackling the problem backwards. What's the point of knowing ANY of this if you're not able to know ANYTHING about your opponents because you're sitting down with new ones again and again?I'm a bit of a hypocrite because I'm running w/out a HUD right now. But I'm confident in my game and don't believe I have any major leaks against this player pool, and I also don't play Rush so I can develop reads. Unless you're playing the same regs over and over again, a HUD imo will be distracting, because you won't be able to amass a significant enough sample size for it to be of any use.DERAIL OVER:Now for actual statistical analysis, your 3-betting at 10nl in general should be mostly for value. Like 5%? That includes 3-betting lightly late position openers.Fold to 3-bet % with 100BBs should be in general fairly high, especially out of position, unless you have reason to believe that villain's 3-betting you lightly.

[jmbreslin 0]

I appreciate what you're saying, trystero, and the main reason I'm focused on Rush right now is to clear my new Tilt bonus. That said, I honestly don't know if I'm going to be able to return to regular cash NLHE because I just don't have the time to put in good volume. I do worry that playing Rush may have the same impact on my desire to play regular cash that trying turbo SnGs had on my desire to play regular speed SnGs (killed it). Regardless, though, whether you play Rush or regular cash you can still use your database to examine your own play and look for possible leaks. The only difference between Rush and regular cash is that in the former you're basing your decisions largely on how a typical 10NL Rush player would play the hand, while in the latter you have more specific info about each player at your disposal. Bottom line is that I can still use my stats to identify general leaks in my game that are also applicable to regular cash games.

[KingJames 11]

BEGIN DERAIL:

:D

[Ninja Ace 1]

Unless you're playing the same regs over and over again, a HUD imo will be distracting, because you won't be able to amass a significant enough sample size for it to be of any use.

This is the stupidest statement about HUDs ever. If you can even halfway read HUD stats you can easily EASILY figure out who the fish are within 2 orbits. Sure... you might not be able to distinguish how TAG/LAG someone is... but you can generally figure out who the passive monkeys are just by looking at numbers from less than 20 hands. I mean... people act like you need to know how someone plays to the percentage point when the reality is that the adjustments you make are such a generalized inexact science that the sample size you have of another player becomes irrelevant. More so, large sample sizes tend to get skewed because people's games change over time

[trystero 0]

That's the stupidest interpretation of what I said ever. You don't NEED a HUD to identify fish. Basic observation skills can do that. HUDs pay off when you have a large sample against regs. When you have several thousand hands on someone, you have a large enough sample of patterns and tendencies that you can begin to exploit. You can then start taking detailed notes, like what you tried to exploit on what date, and if/how the reg responded. Otherwise you just don't have enough data.Of course if you're a mass multitabler then you won't be able to rely on your own observations because of information overload. A HUD is always useful for that scenario.

More so, large sample sizes tend to get skewed because people's games change over time

Weird comment, when you yourself log like 10k hands per day.

[rrumsey 0]

i find i have the best results with about a 65ish AF, and about a 20 vp 15 to about 17 preflop raise in most cash games. In tournys i tend to have around a 50ish or lower AF, depends much more on the table then what i want. Good post man!

[David_Nicoson 1]

I appreciate what you're saying, trystero, and the main reason I'm focused on Rush right now is to clear my new Tilt bonus.

This is the tail wagging the dog, imho. The right game selection is worth a lot more than the typical bonus.

[jmbreslin 0]

Yeah but Rush is absolutely ideal for clearing bonus, especially during happy hour. 2-tabling 10NL Rush I can clear bonus at literally 3x the pace of what I would clear playing even 3 tables of regular 10NL. I fully agree Rush isn't a great way to develop an overall cash game skill set, but you can still work on lots of the basics.

[BaseJester 1]

You're picking the bank to rob based on the interest rate.

[droberts 3]

one more stat you should check is your RPTAP%that's rush poker to actual poker percentage

does this stand for wrap and tap? lulz

[babylondonks 5]

8) hard and fast rules about poker are, in general, stupidHonestly you shouldn't be focusing on changing your stats. If you improve your game, the statistical changes will follow.

[babylondonks 5]

does this stand for wrap and tap? lulz

I know acronyms are difficult things, but he does say what it stands for on the very next line

[Ninja Ace 1]

That's the stupidest interpretation of what I said ever. You don't NEED a HUD to identify fish. Basic observation skills can do that. HUDs pay off when you have a large sample against regs. When you have several thousand hands on someone, you have a large enough sample of patterns and tendencies that you can begin to exploit. You can then start taking detailed notes, like what you tried to exploit on what date, and if/how the reg responded. Otherwise you just don't have enough data.Of course if you're a mass multitabler then you won't be able to rely on your own observations because of information overload. A HUD is always useful for that scenario.Weird comment, when you yourself log like 10k hands per day.

No... you can exploit regs easily after about 100 hands I promise. Also your "basic observation skills" are great when you're at the max 4 tabling... And yes I mass multi and it helps point stuff out for me... but you'd be amazed at what I can pick up and adjust to during a session even with so many tables up. I don't get overloaded but I do wish that sometimes there was a quick button to timebank every table I was at so I could actually *use* my timebank

Weird comment, when you yourself log like 10k hands per day.

My game adjusts daily to things I've learned... also if you have mega large samples on somebody like me who is massing... your table selection isn't working too well... that's why I don't have craptons of hands on the "regs"

[Ninja Ace 1]

8) hard and fast rules about poker are, in general, stupid

Are you saying Branigans law doesn't apply?

Honestly you shouldn't be focusing on changing your stats. If you improve your game, the statistical changes will follow.

Obv you don't focus on "changing" your stats... but stats do point out where the leaks may be... which is the whole point

[fighter 4]

8) hard and fast rules about poker are, in general, stupidHonestly you shouldn't be focusing on changing your stats. If you improve your game, the statistical changes will follow.

catch 22.Looking at stats gives you an idea of where to look to improve your game. However I guarantee that no one here has a clue what to look for beyond the positional stats and VPIP/PFR.

[Ninja Ace 1]

Looking at stats gives you an idea of where to look to improve your game. However I guarantee that no one here has a clue what to look for beyond the positional stats and VPIP/PFR.

Because all your other decisions should change based on opponents so most of those stats are useless unless broken down by villain type

[SCS 0]

I might have a 3bet% of 10% against villain A and the same 3bet% against villain B. Villain A calls a lot of 3bets and villain B folds a lot to 3bets. Is the range of hands I 3bet the same for each villain?

[Ninja Ace 1]

I might have a 3bet% of 10% against villain A and the same 3bet% against villain B. Villain A calls a lot of 3bets and villain B folds a lot to 3bets. Is the range of hands I 3bet the same for each villain?

No... the one who calls 3bets a lot I would think you'd want to include more low pairs in your 3bet range bc if he is calling too many then you need to have decent equity against his rangeThe one who folds to a lot of 3bets you'd want a lot of suited Ax's and Kx's for card removal... and the ones who fold to a lot of 3bets will also be 4betting you a lot so it is easy to muck these hands confidentlyBut you can't do it just based on how often he folds to 3bets... bc it is also dependent on fold to c-bets per street... for example... if someone almost never folds to flop bets but folds to a ton of turn bets suited connectors go up in value bc you will be seeing more cards to make your big hands as you're setting up the barrels.

[shynepo3 0]

No... the one who calls 3bets a lot I would think you'd want to include more low pairs in your 3bet range bc if he is calling too many then you need to have decent equity against his rangeThe one who folds to a lot of 3bets you'd want a lot of suited Ax's and Kx's for card removal... and the ones who fold to a lot of 3bets will also be 4betting you a lot so it is easy to muck these hands confidentlyBut you can't do it just based on how often he folds to 3bets... bc it is also dependent on fold to c-bets per street... for example... if someone almost never folds to flop bets but folds to a ton of turn bets suited connectors go up in value bc you will be seeing more cards to make your big hands as you're setting up the barrels.

this is good stuff...i wish i played online and used a hud, etc..would change my game drastically...but then, i think my live game would suffer once i got use to having all those stats and extras...

[SCS 0]

No... the one who calls 3bets a lot I would think you'd want to include more low pairs in your 3bet range bc if he is calling too many then you need to have decent equity against his range

Against someone who calls a lot of 3bets you don't want to be 3betting small pocket pairs.Hands that can make good top pairs are much better.

[trystero 0]

No... you can exploit regs easily after about 100 hands I promise.

Some things, maybe, just because so many regs play the steal/fold type game when they're in position. But 100 hands simply cannot develop a meaningful sample on almost anything.

My game adjusts daily to things I've learned...

You aren't changing your game all that much daily. Are you telling me that your game has changed markedly since last week? No it hasn't.

also if you have mega large samples on somebody like me who is massing... your table selection isn't working too well... that's why I don't have craptons of hands on the "regs"

Not all that relevant, because this discussion isn't about game selection, but suppose we're both multitabling. I'm 12-tabling and you're at just 4 of my tables. We play 60 hands/hour and I log a 2 hour session. That means we've shared almost 500 hands already, and it's only our first encounter.

[jmbreslin 0]

Against someone who calls a lot of 3bets you don't want to be 3betting small pocket pairs.Hands that can make good top pairs are much better.

Against this type of villain, shouldn't we only be 3-betting for value?