Open Side Menu Go to the Top
Register
Stats and Graphs Analysis Thread Stats and Graphs Analysis Thread

02-01-2012 , 06:42 PM
Quote:
Originally Posted by mpethybridge
OK, sorry it took me so long to get back to this post. My sleep schedule got all screwed up in the last few days.

So here are a few things I want you to look at:

Button play:

a. As it is in all the other positions, your c-bet stat is too high. I won't say what it "should be" because the optimal number would depend on a lot of things such as your table selection. But where it is is definitely too high. So I want you to go back through your steals and look for c-bets you should not have made--maybe air on a wet board, maybe hands that flopped as middle pair and have good showdown value against the villain's preflop calling range but not against his c-bet calling range--situations like that.

b. You are calling a very wide range on the button. I suspect this is a leak. you should run some filters in your tracking software to see if you can determine what leaks you have here. I suspect that if you run cold call preflop = true from the position page, you will see that you are losing money with at least a part of your range.

c. Calling preflop 3 bets too much. Run this filter and look at your loss rate. The closer it is to the size of your average raise from the button, the bigger your leak. If you are losing more than your average raise size, you have a major leak. For example, if you typicllly raise 3 bb from the button, and your win rate when you call a 3 bet is -240bb/100, you have a significant leak; if it is -310bb/100, you have a major leak.

Blind Play:

a. You are playing the blinds far too passively. Look at your VPIP/PFR stats from the blinds:

Big blind: 15/3
Small blind: 37/14

MrJones, if I sat you at a table full of players playing 15/3 and 37/14, you'd be pretty excited, wouldn't you? You'd think, "holy crap, this table is full of loose passive and tight passive players. I'm going to clean up here!"

Well, with all due respect, the reason you are getting slaughtered in the blinds is because you are playing the two toughest positions at the tale in the fishiest possible way. You need to tighten up and play a TAg style from the blinds. Here are the steps I want you to take to tighten up:

1. In your tracking software, filter for did 3 bet = true and go to the position page. Look at your win rates from the small blind and the big blind. If it is positive, continue, if it is negative, work on plugging that leak.

Assuming your win rate is positive, I want you to go to note the total number of 3 bets for each blind position. Then go to the hole cards filter and remove premium hands: AA, KK, QQ, JJ, AK.

Note the new total of 3 bets. If it is less than one third of the previous total, you should be 3 betting air/light value more often.

2. You should also stop calling such a wide range from both of the blind positions. In general, you should have a very tight range to call an EP raise, and your range should progressively open up the later the position of the initial raiser. But it should never get very wide. Only the players with the biggest edges at their stakes can play weak hands such as KTs or T9s or QTo for less of a loss than folding them.

Steal defense: before you call or 3 bet a steal attempt with the middle or bottom of your newly tightened up range, you should qualitatively describe the player making the steal as:

a weak player
b. about as good as you
c. better than you.

If the player falls into group c, you should only defend with the top of your range--hands that are way ahead of his stealing range, and I do mean waaaaaay ahead.

If the player falls into group b., you should only defend with hands that are significantly ahead of his stealing range--say, in the top 1/3 of his stealing range.

If the player falls into group A, you can go ahead and defend with maybe the top half of his stealing range.

The rationale behind this system of steal defense is very simple--there are four advantages you can have at a poker table:

1. Initiative
2. Position
3. Hand strength
4. Skill

When you defend your blind with a weak hand against a player who is as good or better than you, you play the hand with zero of the four possible advantages. How can you possibly expect to play a hand profitably when your opponent has all 4 possible advantages? You can't. It is impossible. So when we defend from the blinds, what we have to do is consciously manipulate the four possible advantages. The stealer has the preflop initiative and will retain position; there's nothing we can do about that. So what we can do is make sure that we come to the hand with a big hand strength advantage and a big skill edge, to compensate for the fact that the other player has the other two advantages preflop.

This system of blind defense calls for humility. You have to be able to look at a stealer's stats and say, "ok, this guy is too good to make a profit on, I fold." But you have to do it. Getting involved OOP against good players with the initiative is a recipe for blind losses that are, well, about like yours.

Middle and Early Position:

a. Open up from every position--to about 6% UTG, and to about 15% on the hijack. You are way too tight. Open up with raises, not calls or limps.

b. Reduce your c-betting frequency. Just find the worst spots you are currently c-betting--wet boards with air, etc., and stop c-betting them.
Good, detailed insights and much appreciated. Will think on this and make some adjustments. To be continued!
Stats and Graphs Analysis Thread Quote
02-01-2012 , 06:45 PM
Quote:
Originally Posted by LazyAce
mpethy you are awesome for helping out so many people itt for free. I mean, its probably super bad for the quality of the games, but it is a pretty selfless act to just help people the way you do.

WP.
+1

Respect for taking the time to do this for a number of people, hopefully those who have been given help will be more inclined to do the same thing if/when they find themselves in a position to do so.
Stats and Graphs Analysis Thread Quote
02-01-2012 , 07:13 PM
Quote:
Originally Posted by LazyAce
mpethy you are awesome for helping out so many people itt for free. I mean, its probably super bad for the quality of the games, but it is a pretty selfless act to just help people the way you do.

WP.
Thanks. I'm just paying it forward. Back in 2005 and 2006 when I first joined 2+2, some really awesome posters who went on to be big successes helped me with my game. I'm thinking of Beluga Whale, who was playing like NL $100 back in 2005, Fimbulwinter, AJmargarine and Orange. All of these great players helped me and a lot of other people for free, and never asked anything in return. I'm just following their example.
Stats and Graphs Analysis Thread Quote
02-01-2012 , 08:08 PM
Lazyace,

Thanks alot. Recently i reduced my Fcbets significantly 'cuz i feel like im burning money and getting floated. Nowadays average table fcbet% is 45-66%. Last time used to be 100% on most tables. Lol. Else i think my overall cbets% would be even higher. Heh.

Will post trouble 3bet hands in the future, regardless of win/lose.
Stats and Graphs Analysis Thread Quote
02-02-2012 , 10:01 AM
HEM2 how do you copy the text of a hand? tilting me so bad cant work it out lol plzzz help.

i want to copy the text to put in converter for 2+2
Stats and Graphs Analysis Thread Quote
02-03-2012 , 08:45 PM
Schwan, you are nitty, at 11/8 you must be leaving money on the table. Blind stealing under 20% is way too passive, if you are selecting tables with a couple of nits to your left you can get this to 30% pretty easily. Winning at showdown at 54% seems quite high, you may be folding the best hand a smidge too often. FWIW my impression is that you should open up your pre game a bit, especially stealing and being more aggressive on the button.
Stats and Graphs Analysis Thread Quote
02-03-2012 , 09:12 PM
Quote:
Originally Posted by Cangurino
MrBeer, first impression: You ran bad (EV-adjusted you broke even). You 3-bet a lot out of position. 19% resteal looks excessive to me, especially if people don't fold.You probably c-bet too much, in particular in late position. Finally you call a lot of 3-bets.
+1

Definitely c-betting too much, especially in late position.

Your button win rate being negative is a huge issue. You need to filter for every situation--steals, cold calls, calling 3 bets, etc. to find out what other problems you have in addition to the c-betting.

The same thing goes for the blinds, where you are getting absolutely massacred. I suspect that Cangurino nailed it, and that you are getting creamed 3 betting a nitty field way too often and way too light. But look also at your small blind steals and at your cold calling from the blinds. I suspect you will find several minor problems there as well.
Stats and Graphs Analysis Thread Quote
02-03-2012 , 10:55 PM
What is the smallest sample necessary to get a good stats review?
Stats and Graphs Analysis Thread Quote
02-04-2012 , 02:09 AM
Thanks mpethybridge, I appreciate the help.
Stats and Graphs Analysis Thread Quote
02-04-2012 , 06:10 AM
Quote:
Originally Posted by Uthanak
What is the smallest sample necessary to get a good stats review?
I'd say 10k hands is the absolute minimum to get general picture. For a good review may 30k.
Stats and Graphs Analysis Thread Quote
02-05-2012 , 09:47 PM


Can someone post a bigger sample size of their AA and KK winnings. I want a more realistic look. I just got poker tracker this week. So I know my stats can be a bit bloated.
Stats and Graphs Analysis Thread Quote
02-05-2012 , 11:10 PM
Quote:
Originally Posted by Paul D


Can someone post a bigger sample size of their AA and KK winnings. I want a more realistic look. I just got poker tracker this week. So I know my stats can be a bit bloated.
AA 1038 bb/100 (519 BB)
KK 643 bb/100 (321 BB)

this is from a ~400k sample size mostly from 100-600nl
Stats and Graphs Analysis Thread Quote
02-07-2012 , 04:38 AM
Quote:
Originally Posted by Paul D
Can someone post a bigger sample size of their AA and KK winnings. I want a more realistic look. I just got poker tracker this week. So I know my stats can be a bit bloated.
I wish my aces won 100% of the time like yours did! :P
On FR 2NL on Stars, my aces win about 85% of the time for me at a clip of 750bb/100. Kings win around 80% and 650bb/100. That's over 40k hands, some being played with a half stack. With 100bb stacks (or deeper) the bb/100 will typically be higher.
It looks like you've had some rungood over your small sample size, as you'd typically only get each pocket pair 13 times in 2900 hands. Be prepared for the aces to come less frequently in future... and also expect them to get cracked about 20% of the time.
Stats and Graphs Analysis Thread Quote
02-07-2012 , 09:07 AM
Thanks for the replies.
Stats and Graphs Analysis Thread Quote
02-08-2012 , 03:00 AM
Paul you probably do not want to be playing 29/14
Stats and Graphs Analysis Thread Quote
02-08-2012 , 09:48 AM
hi guys nice thread, im mainly a mtt, sng player but want to move to cash in the near future, i started at NL50 FR to learn the game, small sample i know but is there already something remarkable, or some kind of obvious leak in my stats ?

would be very cool if one of u stats pros would give me some advice about my game,

im also not sure if i hav provided enough of important stats for you guys to judge but im happy to add some more stats if thats helpfull,

and btw thats 24 tabling on stars

steal is 39

thanks




Stats and Graphs Analysis Thread Quote
02-08-2012 , 11:36 AM
Your stats look very solid WhiteWhidow
Stats and Graphs Analysis Thread Quote
02-08-2012 , 06:39 PM
This is the last 25,000 hands I've played at $4NL. Feel like I've been playing much better over the past 10k hands or so but still haven't broken even. Any comments on my stats would be much appreciated!

Stats and Graphs Analysis Thread Quote
02-08-2012 , 08:15 PM
Quote:
Originally Posted by Paul D
Can someone post a bigger sample size of their AA and KK winnings.
Stats and Graphs Analysis Thread Quote
02-08-2012 , 09:09 PM
Quote:
Originally Posted by mReid
This is the last 25,000 hands I've played at $4NL. Feel like I've been playing much better over the past 10k hands or so but still haven't broken even. Any comments on my stats would be much appreciated!

Looks like you should probably play tighter in early position, and you're limping a 4% range UTG too which you should probably not do. I know your results sport a decent winrate in ep but it seems very likely that its variance.

Open up more in the CO. BTN stats look solid but overall you would probably benefit from getting your ATS up to around 40% of the time.

You're cbetting a lot of flops but not following up on the turn as much. You should look at some hands where you might be better off check/folding flops instead of cbetting. I think if you cut it back to under 70%, your turn cbet % would naturally increase because you'd have a stronger range getting to turns and your stats become a little more balanced like 65 flop cbet and 60 turn cbet. It makes you harder to play against.

Your 3bet % seems kinda high for 4nl. I dunno how that stake plays but guessing by your redline it might not be working out so well for you. I'm sure you can definitely 3bet light a decent amount but just make sure you're not overdoing it and be careful of turning good flatting hands into bluffs by 3betting them.

Your fold to 3bet seems kinda low. Solid players at your level often fold to 3bets more than 80% of the time. The reason that this is okay is because there aren't many players at 4NL that are going to exploit you for folding so much but you may be exploiting yourself by calling too often against ranges that are much stronger than you think.

Suggestions only though and not gospel. I'm interested to see if Mpethy or others agree with some of the advice I've given.

Good luck though.
Stats and Graphs Analysis Thread Quote
02-09-2012 , 06:48 PM
Just a quick endorsement for mpethybridge


I just had a one hour db session, and it was pretty awesome!

We covered a number of topics/leaks and the session ran well over time. I now have some concepts to think about and put into action. A lot of info for my tiny brain to cope with.

I only play at 5nl and he doesn't usually go that low, but if anybody is thinking of using his services, I cannot recommend highly enough.

Thanks m.
Stats and Graphs Analysis Thread Quote
02-09-2012 , 07:24 PM
Image disappeared for some reason, re-posting.

Stats and Graphs Analysis Thread Quote
02-12-2012 , 12:37 PM
what is a good fold to 3bet % to have at 6max 50NL overall ?
Stats and Graphs Analysis Thread Quote
02-13-2012 , 08:46 AM
Quote:
Originally Posted by WhiteWhidow
... but is there already something remarkable, or some kind of obvious leak in my stats ?
Yes, there is something. The superwide gap between Flop- and Turn-Cbet-Frequency might indicate that your whole postflop-play needs some work.
Stats and Graphs Analysis Thread Quote
02-14-2012 , 04:59 AM
Hey, I've got the same question about winrates with top value hands.
Mine look something like this:
AA 1246 hands.... winrate 751bbs/100
KK 1251 hands....winrate 400bbs/100
QQ 1198 hands....winrate 115bbs/100
AK(s and o) 2752 hands... winrate 32bbs/100

I also have a negative winrate with JJ and +42bbs/100 with TT... which makes no sense )
If this looks wrong to you, please let me know
Samplesize is 210k hands
Ty.
Stats and Graphs Analysis Thread Quote

      
m