Open Side Menu Go to the Top
Register
**** Official Stats and Graphs Analysis thread **** **** Official Stats and Graphs Analysis thread ****

04-02-2010 , 12:31 PM
Quote:
Originally Posted by Gallo
Clearly I need some help only have 20k hands hoping I can attribute the negative win rate to a smallish sample size. Although I'm beginning to feel I need to make some real adjustments if I even want to be a marginal winner. All advice welcome.


By null at 2010-03-31


Gallo:

For the time being, you need to work on your blind play and your button play.

You are playing the blinds like a loose passive fish. Look at your vpip/pfr from those positions; you know better than to play like that.

This leak is easy to make some significant progress toward improving--you need to click "fold," a lot more from the blinds.

You are completing far too wide a range in the small blind. Take the best hands you are completing, and turn them into raises. Take the worst hands you are completing, and turn them into folds. FOR THE TIME BEING leave yourself a completion range of small pocket pairs and suited connectors that you can play reasonably confidently when you flop a pair (so maybe K9s, maybe KTs, maybe QTs, etc. Again, for the time being, leave yourself a completion range of just about a dozen hands.

Fold to raises a lot more out of both blinds. Look in your holecards report for hands you cold called from the blinds with and that you were unprofitable with. For types of hands, such as suited aces, suited connectors, broadway, etc, group them into categories to increase your sample size, and then start folding categories that are unprofitable.

You will probably see some good hands that you are losing with when you cold call--AQ, AK, KQ, JJ, TT, 99, for example. Negative win rates with these hands represent a post flop leak. These hands are too good to fold, so you should spend study time away from the table figuring out how to play these hands for a profit from the blinds (more aggressively than you do now, and winning with unimproved hands with a higher frequency). Grind some poker stove to see how these hands match up against various stealing ranges, and develop lines away from the table to use on various types of flops.

As for your button play, your problems are likely very similar to the player in my previous post, so read and run the filters I described there.
04-02-2010 , 12:52 PM
mpethy i got a quick question about the DID COLD CALL filter.

if someone raises preflop and you flat on the button say, is this considered a 'cold-call' by HEM (the bb being the first bet and villains pf raise being the second).
and what about if someone raises pf in ep, villain in mp 3bets and you flat the 3bet on the button. is this a cold-call too?

just wondering if both these scenarios are included if you set DID COLD CALL= TRUE
04-02-2010 , 02:12 PM
Thanks mpethy I knew my biggest leak was out of the blinds but you have given me some great advice on how I can hopefully plug it. Also I ran your filter for late position unopened pfr and seeing a flop and as you suspected I am negative for the button but not the cutoff. I will review many of the hands that fit that filter but what do you think is the most likely post flop flaw? Am I giving up to easily when my steal meets resistance or the opposite? Or should I just be stealing less? Thanks again.
04-02-2010 , 05:43 PM
Quote:
Originally Posted by Gallo
Thanks mpethy I knew my biggest leak was out of the blinds but you have given me some great advice on how I can hopefully plug it. Also I ran your filter for late position unopened pfr and seeing a flop and as you suspected I am negative for the button but not the cutoff. I will review many of the hands that fit that filter but what do you think is the most likely post flop flaw? Am I giving up to easily when my steal meets resistance or the opposite? Or should I just be stealing less? Thanks again.
The reason you are positive from the cutoff but negative from the button is very probably your tighter cut off opening range.

Phrased differently: when you play a weaker range of hands, your winrate goes negative.

This suggests that you are losing with the bottom of your stealing range. The easy (nitty/weak/tight) fix is to stop stealing with the bottom of your range. If you do this, make sure you understand that it is a TEMPORARY FIX aimed at getting you back into profit while you figure out how to play the bottom of your range.

I see people doing a bunch of things post flop with their steals that make me scratch my head. For example, I see people stealing, say, 72s and then checking a flop that looks like A94r. I mean, if you aren't going to c-bet that flop, what flop are you waiting for? 772?

Another thing I see is people going for too many streets of value with marginal hands. They wind up isolating the caller to a range that they lose to at showdown. They also wind up getting bet off their equity. So, for example, if you have T8s and the BB calls your steal and the flop comes down A84r, are you bet/folding? If he calls, are you betting the turn unimproved?

Another thing I see is people with a button W$SD < 50%. Clearly they are getting to showdown too light. But are they getting there by checking or by calling? These are two different leaks, and you need to do a detailed review of your button steals and button stats to figure it out.

I'm not sure I can guess which of these leaks you may have; people who are negative on the button when they see a flop usually have all of them, at least to an extent.
04-02-2010 , 06:07 PM
Quote:
Originally Posted by riksanchez
mpethy i got a quick question about the DID COLD CALL filter.

if someone raises preflop and you flat on the button say, is this considered a 'cold-call' by HEM (the bb being the first bet and villains pf raise being the second).
and what about if someone raises pf in ep, villain in mp 3bets and you flat the 3bet on the button. is this a cold-call too?

just wondering if both these scenarios are included if you set DID COLD CALL= TRUE
Yes to the first.

The answer to the second seems to be no. I don't know this for a fact, but when I filter for being in this scenario, it returns a few results where there was a raise and a 3 bet in front of me and I flatted the 3 bet. then I add "did cold call = true," it returns 0 hands. So apparently "did cold call = true," only counts it as a cold call when your first voluntary action is to call the first raise.
04-02-2010 , 06:07 PM
Quote:
Originally Posted by mpethybridge
That looks fantastic, actually. Curious about your sample size, because EP looks unsustainably high. I mean, seriously, your EP WR is roughly 3 times what I am used to seeing from solid winning players.

Your blinds as a stand alone analysis (nod to doublefly) are right at the arbitrary cut off I tell people is "fine--not leaking at all."
~500,000 hands
04-02-2010 , 10:33 PM
nice one. thanks for the reply
04-06-2010 , 05:14 PM
Can someone hit me up and tell me how i post graphs/stats from PT3 in here? And Im new to keeping track of my rates and sttats. Any recommendations?Ty

Last edited by Marty_McFly; 04-06-2010 at 05:29 PM.
04-07-2010 , 11:30 AM
Quote:
Originally Posted by Marty_McFly
Can someone hit me up and tell me how i post graphs/stats from PT3 in here? And Im new to keeping track of my rates and sttats. Any recommendations?Ty
The basic thing you are accomplishing here is to upload an image onto a photo hosting service and then you copy the URL of your photo on the remote server into your 2+2 post. Using [img] tags before and after the URL tells the software to display the actual image in your post, rather than the link to the image.

The way I do it is to use mspaint.

1. Get HEM to the screen you want to post (say, the position page)

2. take a screenshot by hitting the Prt Scr key on your keyboard.

3. Open mspaint.

4. Ctrl V to paste the image into mspaint.

5. Use the select a field function in mspaint (the square dotted box) to select the field from the screenshot you want to use as your final image

6. Ctrl C to cut out that image

7. open a new mspaint file.

8. Ctrl V to paste the selected field into the new file

9. Save the file.

10. I use photobucket, so I go to www.photobucket.com (you'll have to create an account--it's free). Go to your "album" on photobucket. Click upload an image. Select your image from its location on your hard drive. double click it and it will upload to your album on photobucket.

11. click on the [img] tag box beneath your image in your album. this may work the same as pressing Ctrl C to copy, or you may have to highlight it and click Ctrl C. This copies the [img] tag location of your image in your photobucket album.

12. Click Ctrl V to paste the [img] tag url for your image into your 2+2 post.

13. Submit your post.

It sounds like a lot of steps when broken down in this way, but it is a very simple process that takes about two minutes once you know what you are doing.
04-09-2010 , 08:06 AM
Just checking how my stats are, making sng->cash transiction

Thanks

04-10-2010 , 03:06 AM
Whats a good total winrate from the blinds? What about a great winrate? What should I be aiming for...-40 total? -50?



Also malice, your total winrate is 15.5bb/100? Seriously? How the hell is your EP/MP so high?
04-10-2010 , 04:44 AM
Quote:
Originally Posted by CxF
Whats a good total winrate from the blinds? What about a great winrate? What should I be aiming for...-40 total? -50?



Also malice, your total winrate is 15.5bb/100? Seriously? How the hell is your EP/MP so high?


It's not that good hah, mpethy this breakdown is ok?
04-10-2010 , 05:26 AM
lol whoops for some reason i was thinking 3 early pos and 3 mid pos. haha
04-14-2010 , 09:31 AM


I am looking to tighten my steal range from CO and BTN, which hands should I begin with eliminating?

Any thoughts/opinions please?
04-14-2010 , 01:19 PM
I'm looking for some thoughts on defending against 3bets. I play at 1/2 (mostly full ring, some 6max).

I know there are opinions on 4bet bluffing, calling with suited connectors and semibluffing some flops in position, etc... I ran some reports in PT and got the following:

150,000 hand sample

global filters: *these filter applies to all
NOT(AA,AKs,AKo,KK,QQ,JJ)

Call 3bet: *I generally never do this OOP
1) suited connectors and suited one gappers
BB/Hand: -1.42 Total hands: 108

2) Pairs 22-TT
BB/Hand: -1.82 Total hands: 286
I've started just folding these or 4bet bluffing

4 bet:
BB/Hand: 0.94 Total Hands: 125

Fold to 3bet
BB/Hand: -0.67 Total Hands: 5,295


So...
4 bet bluffing looks like the best strategy, but you can't just do it willy nilly against regs or they will start 5bet owning you.

This leaves folding or calling.. Folding sucks but calling sucks more for me. Does anyone show profit by calling with suited connector type hands, or at least less of a loss than folding? And if so do you have any thoughts on flop play besides hit/fold?
04-14-2010 , 02:13 PM
Quote:
Originally Posted by azan


I am looking to tighten my steal range from CO and BTN, which hands should I begin with eliminating?

Any thoughts/opinions please?
Quote:
Originally Posted by azan
Basically right now I open so much that it's a leak where I could be saving money. Having to fold to 3bets, having to give up post flop a ton, even though I am still doing my best to 4bet some and double barrel good boards/opponents.

I guess I am not sure which play better, hands like 75o, 96o or hands like Q6o, J4o. Vs a blinds calling range.

Running some quick calcs in poker stove, vs a rough calling range of: 22+,A2s+,A2o+ and broadways, all the following have 35%+ equity, so look like a solid range imo:

Thoughts on this as a base range to work off?
04-16-2010 , 12:14 AM
deleted (wrong forum)

Last edited by SleightOfHand; 04-16-2010 at 12:21 AM.
04-17-2010 , 01:34 AM
ok i have been trying to fix my redline disease for a while now, I have to admit my graph looks a lot like gallo's posted above. having read this thread i have to admit i am still leaking from the blinds despite everything i try. I am losing 19 PTBB/100 = 38bb/100 from the big blind and worse 32bb/100 from the small blind.





so i followed your advice and, starting with the big blind only, i went through every group of hands i am calling in the bb with separately against raises from steal posns and non steal posns.

i wont post the whole chart here but;
v steals, I am positive or at least better than the -1bb i would achieve by folding in ALL categories (even junk) except TT JJ and QQ. This has been alluded to before and obv i need to look at how i am playing those post flop to at least realise my equity in these. But apart from playing those hands better i dont see how cutting out any categories is improving things.

v non steal posn raises, there are a couple of categories that are a problem that I am frankly surprised i have called with at all as i should know better, so i can make an improvemnt there which is good


here is a summary against both types of raises;



any comments on the breakdown v each type? I will get in first and say i am surprised at the low 3bet% v non steal posn raises so clearly i have to up the 3bet and lower the calls here. how about v steal raises?

So how to reduce my lossrate from 38 to 28 or better still 22?
your advice is alwys 'call less' but looking at that chart i have to say it is the figures in the "I folded" row that are more the problem that the "I called"

over 100 hands at Fr we post 11 big blinds or 11bb. assuming we fold 9 of these times (vip=18%) that means in the two times we play we have to win 6.8 bb to make up for the 9 we lose when we fold (to achieve your target of a net -22bb/100), which is 3.4bb /hand on average every time we play.

How the hell do people do that? 3 betting a single raiser only nets that if it is successful 100% of the time, which they never are, so it cant be achieved by 3 betting. how about by calling and playing post flop? how many hands can you achieve 3.4bb/hand with full stop, let alone out of position?

playing 3 hands out of 11 reduces the win rate required from those 3 hands to a more manageable 1.9bb/hand but obv adding more weak hands to the calling range cant be right, so to achive this we would have to jump the 3 bet up to about 17% which is kinda over the top.

maybe squeezing more is the only thing i can think of. will have to look at that.

anyway thanks for taking the time to answer all these posts and any comments or help will be appreciated.

I will do the small blind analysis next...

Last edited by village kid; 04-17-2010 at 01:43 AM. Reason: typo
04-18-2010 , 10:40 AM
Hi these are stats of a live player transitioning to on-line. Do these stats suggest an epic rungood...or genuine skill?

04-18-2010 , 03:36 PM
Sample size is way too small. Make it at least 75k hands and come back.
04-20-2010 , 03:13 AM
Ok so I am playing NL50 but I am posting here because I would like to know what I need to change in my game in order to adapt to NL100. I did some "good" changes to my game this month so I am only posting this month stats. Overall my winrate on NL50(in something like 200k hands) is ~3bb/100



Even though while playing I think I defend my blinds very good results tell different story :/
04-20-2010 , 03:44 AM
17.53 vpip / 7.4 pfr in early position.

Are you limping a lot or cold calling preflop raise ?


I personally never open limp and even limp behind in EP since you're likely to be iso and will not have position which sucks.

This might be OK at NL25 and maybe NL50 but it's going to be exploited a lot at NL100.
04-20-2010 , 03:50 AM
17.53 is bb/100 not vpip

God, having 17vpip and 7PFR early would be horrible :P
04-20-2010 , 04:02 AM
lol my bad..
04-21-2010 , 03:14 PM
Quote:
Originally Posted by Daniel1941
Ok so I am playing NL50 but I am posting here because I would like to know what I need to change in my game in order to adapt to NL100. I did some "good" changes to my game this month so I am only posting this month stats. Overall my winrate on NL50(in something like 200k hands) is ~3bb/100



Even though while playing I think I defend my blinds very good results tell different story :/
Daniel:

Your preflop game looks ok except for a few things:

1. You need to loosen up on the button. A LOT. You should be stealing a lot more. You should probably also be isolating a lot more than you seem to be. This doesn't mean that you should just start stealing more regardless of table conditions. It means that you should look carefully at the two players to your immediate left, and what their 3 bet and fold blind to steal stats are. If they are players that fold to a lot of steals, then you should be stealing against them with almost any two cards. I'd be ok if you didn't steal from them with the worst 25 or 30% of hands, but, really, there is no reason not to at least try it once or twice and see how it goes against them.

If the players to your immediate left play back at steals to the extent that you think you should not be stealing against them with any two cards, then you need to ask yourself why you are at that table. Maybe there is a good reason, such as that you have position on a fish. If this is the case, then stay at the table and be careful with your steals, but keep stealing with a fairly wide range.

If there is no one at the table you are targeting, and the two players to your left play back at steals a lot, then you should strongly consider leaving the table.

2. You are getting creamed in the blinds. You need to filter for calling a raise and look at the categories of hands you are calling a raise with when you are in the blinds. Most players lose money (compared to folding) with most of their calling range. You should look at the hands you are losing with when you call a steal, and you should stop cold calling with those types of hands.

Another common leak that players have in the blinds is that they are losing money, or not making enough, with their non-premium 3 betting range. So filter for did 3 bet = true and holecards, deselect AA, KK, QQ and AK, and look at your win rates from the blinds. If those win rates are low, such as around 100bb/100 hands or so, you are ether 3 betting the wrong people or you are misplaying a significant number of those hands post flop. You'll need to do detailed hand reviews to determine the extent to which these leaks are affecting your blind loss rates. Obviously, they are not mutually exclusive.

Another leak you may have is not stabbing at pots where you get a free look at the flop from the BB (limped pots where you check your option preflop). A small negative win rate in this spot is ok, but if it is much higher than -.08bb/100, you are leaking. The best players have a small positive win rate in this spot.

      
m