03-04-2006 , 02:06 AM
This is a how-to post, but it doesn't belong in the software forum; it's to teach people how they can use the Poker Tracker data to find flaws in their game.

Not a week goes by that someone doesn't ask if they are playing the game right. In that post, they include a dozen numbers from Poker Tracker and hope that the old-timers on SSNL can fix all their holes. While it's true that Poker Tracker can help find problems, this is not the way to approach it.

I'm going to try to give you a rough guide for the things you can do to check on your game. These are all just my opinion; they're all subject to interpretation, and other people may disagree with me strongly. The best way to play is usually player-specific, but these strike me as some things you can check on that are frequent flaws in the small-stakes player's game.

1. Do you have sufficient preflop aggression? To answer this question, open up your ring game statistics and go to the "position stats" page. For each position other than the small blind, divide the "PF Raise %" by the "Vol. Put \$ In Pot." If you get a number smaller than 0.5, you're not aggressive enough out of that position. See, aggression is a relative term; it should be a function of your level of looseness. You can be a consistently winning player at SSNL with a VPIP of 12%, and you can be a consistently winning player at SSNL with a VPIP of 30%, but only if you are sufficiently aggressive. My general guideline is that you should raise at least half the hands you play, from every position on the table.

2. Are you positionally aware? Positional awareness means that you understand Ed Miller's comment when he said:
Quote:

Total all the dollars you've ever bet playing poker. The large majority of those dollars should have been bet from late position. Only a small percentage of your total handle should have been bet from up front.

To test this, go to the Position Stats and look down the list of VPIP from Button to UTG. You should see that VPIP steadily dropping the farther you get from the button. I'd love to see my button VPIP at double my UTG VPIP, but if my Button VPIP is at least 50% larger than my UTG VPIP, I'm happy with the situation.

3. How's my stealing? To check on your performance when trying a blind steal, go to the General Info. tab. Where it says "Att. To Steal Blinds" I'd like to see that number at LEAST 20%. (Personally, I like mine to be over 30%, but I'm very aggressive in these situations. If you're trying to steal the blinds less than 20% of the time, you're leaving lots of money on the table.) Now click on "Filters..." and under "Chance to Steal Blinds" click "Chance to Steal &amp; Raised." Select OK and look at the numbers. This shows every time you've tried to steal the blinds, and how the attempt turned out for you. Under "Totals" see the "BB/Hand" statistic. That shows your per-hand winrate on blind steals. If you multiply this number by 100, it should be at least double your "PTBB/100" average winrate. If it's much less than that and you have a decent sample size, you have a hole in your game when it comes to blind stealing. This should be an exceedingly profitable thing to do when you try it; if it's not, you need to work on your strategy.

4. Defending the blinds. Click on "Turn Filter Off," and then click on "Filters..." again. Under "Blind Status" click on "Either Blind." Now under "Vol. Put \$ In Pot" click on "Put Money In." This shows you if you're bleeding money out of the blinds. A "BB/Hand" of about -0.375 would indicate that you were no better off putting money into the pot than if you had folded. If your "BB/Hand" is larger than that, then you typically win back some of your blind money when you put money into the pot from the blinds. That's all you can really hope for. If you click on "Filters..." again and go under "Steal Attempted Against Your Blind" and click on "Steal Attempted." After you click "OK" you'll now see how you did when you chose to defend against a blind steal. Again, the magic number is for your "BB/Hand" to be bigger than -0.375; that means you're making back some of your blinds when you try to defend against a steal. If either of these numbers is lower than -0.375, you'd lose less money by always folding rather than doing what you're doing.

5. Heads-up play. Click on “Turn Filter Off,” then click on “Filters…” again. Under “Hands With Between…Players Seeing The Flop” change the range from “0 to 10 players” to “2 to 2 players.” Hit “OK” and see what comes up. This shows you how you’ve done when you were heads-up preflop, but a flop was dealt. See how you’ve done in these situations. If things look OK, go back to “Filters…” and under “Pre-flop Raise” select “No Raise.” This will show you how you’ve done when you didn’t raise preflop, but the hand was heads-up on the flop (this includes pure limping and when someone ELSE raised preflop, but not when you were the preflop raiser). Is this number positive? If not, it could be an indicator that you have trouble when you are not the aggressor preflop, especially without padding in the pot.

6. Multiway pots. Clear the filter and go back under filters. Change “Hands With Between…Players Seeing The Flop” to “3 to 10 players.” This shows you how you do in multiway pots. If things look good, go back and select “No Raise” under “Pre-flop Raise.” Is it still positive? If so, you’re selecting good times to play/limp multiway pots, and you’re playing them well postflop.

7. Pocket pairs. Under “Filters…” change the “Type of Hole Cards” to “Pairs.” This will show you how you generally play and perform with pocket pairs. Your Total VPIP with these should be EXTREMELY high; unless you play at highly unusual tables, I’d be surprised to see this number below 85%. Pocket pairs make extremely powerful hands that are extremely well-hidden; if you’re not playing them almost all the time, you’re leaving money on the table. Also, your Total PFR% with these hands should be rather high -- at least 1/3 of your VPIP, if not 1/2. Some people have this number higher still, and I don’t have a problem with that, especially at short-handed tables. If you have enough hands, I’d expect every one of these lines to be positive, and reasonably significantly so. If you have any glaringly negative numbers, especially AA-88, it may indicate bad play. Look over individual hands where you lose lots of money and see if you played too timidly early in the hand, or if you went too far unimproved in the face of resistance. Also, look at the hands where you won to see if you played too timidly, or if you routinely forced weaker hands out when you should have been milking them for profits.

8. Suited connectors. Under “Filters…” change “Type of Hole Cards” to “Suited Connectors.” I’m much less likely to play suited connectors than pocket pairs, but some people play them religiously. As a result, I don’t really have a good suggestion as to how high your VPIP or PFR should be. However, your BB/hand should be positive; if it’s not, you’re probably not playing your suited connectors well. Remember: these hands play best in a multiway, unraised pot, or as a steal move. In the “Filters…” change “Vol. Put \$ In Pot” to “Cold-Called.” When you hit OK, you should have almost no entries to view. Of the times you cold-called, you should be able to come up with a specific explanation for why you did so in each and every one of them. Review the hand histories; if you can’t come up with a really good reason why you thought it better to cold-call, rather than raise or fold, you need to rethink your suited connector strategy. Good explanations: the raise was very small, villain is passive post-flop, I had position on villain, villain and I are both extremely deep-stacked, villain is incredibly aggressive preflop, my suited connectors are particularly strong, there are several cold-callers in front of me, etc. I’m not saying you shouldn’t ever cold-call with suited connectors; rather, I’m saying you shouldn’t AUTOMATICALLY do so. Your default play here should be to fold weak suited connectors and reraise strong ones.

9. Unsuited connectors. Clear the filter and then go back into it. Change “Type of Hole Cards” to “Off-Suited Connectors.” Your VPIP for these hands should be noticeably smaller than your VPIP for suited connectors. Check your winrate and make sure it’s positive. Filter for cold-calling and see if you had good reasons for doing so, keeping in mind that the reasons need to be even stronger than for suited connectors.

10. Postflop aggression. Clear the filter. Select the “More Detail…” button above the “Filters…” button. Scroll down. There is a section marked “First Action on Flop After A Pre-flop Raise.” This shows your likelihood of continuation betting. If you add Bet and Raise, the total should be at least 40%. If it’s not, you’re probably giving up too soon on your good hands, and that will cost you money in the long run. Remember: people who cold-call a preflop bet are often in fit-or-fold mode. If you don’t bet, you don’t give them a chance to fold. The pot is already decent-sized, and there’s no reason to give some donk a free look at a turn card that could sink you. If you raised preflop, you need a good reason NOT to raise the flop. Continuation betting should be your default play. Scroll down a bit farther to “Aggression Factor.” Your total aggression factor should be at LEAST 2. No-limit is not a game where you can call frequently and turn a profit. You should always be looking to see if you can raise or fold; only if you have a good reason why you CANNOT raise or fold should you call. As a result, calling should be an infrequent occurrence in your play, which gives you a large aggression factor.

11. Check-raising. Some people never check-raise; others check-raise infrequently. I personally like to check-raise at least once in awhile; 1% would be fine, 0.5% would be acceptable. The goal of the check-raise is to remind your opponents that just because you checked does NOT mean that you don’t have a hand. However, circumstances need to be very specific for a check-raise to be appropriate. Typically, I check-raise on the flop when OOP against a preflop raiser, or on the turn when OOP against a flop bettor/raiser who was clearly not on a draw (uncoordinated flop). If you are check raising much more than 2% of the time, you’re being entirely too tricky for a SSNL table, and straightforward play would probably be more profitable for you.

All of this is just an introduction to the kinds of self-analysis you can/should do with Poker Tracker statistics. Notice how much more in-depth it is than just glancing at a few VPIP numbers. Typically, the only person who can truly do a “check-up” on your playing style and ability is YOU. As always, if in your searching you find hands that indicate you may have a flaw in your poker reasoning, post them up (one at a time, of course). Tell us the problem you are worried you might have, and why you think this hand might indicate the problem. Then, open the discussion up to see if 2+2ers agree or disagree.
03-04-2006 , 03:02 AM
Cant let this gem die out this fast, very nice post, going over this once i get off my shift!
03-04-2006 , 03:17 AM
This needs to be sticky-ed or FAQed. Great post.
03-04-2006 , 03:18 AM
pokey you are truly a sweetheart
03-04-2006 , 03:46 AM
Great stuff Pokey. I am consistently impressed with the quality of your posts.
03-04-2006 , 06:53 AM
Quote:
This shows every time you've tried to steal the blinds, and how the attempt turned out for you. Under "Totals" see the "BB/Hand" statistic. That shows your per-hand winrate on blind steals. If you multiply this number by 100, it should be at least double your "PTBB/100" average winrate.
If mine is insanely higher (83 bb/100) does that mean I should be stealing way more? I'm only at 10% right now. What range do you recommend stealing with in a typical loose NL 25-100 game?

HU: 0.09bb/hand
HU no raise: 0.00bb/hand
Multiway, no raise: 0.01bb/hand

Looks like a leak!

Thank you very, very much for this post.
03-04-2006 , 09:08 AM
Quote:
This needs to be sticky-ed or FAQed. Great post.
Copied, pasted, stored

Thanks for the help Pokey
03-04-2006 , 10:40 AM
Awesome post Pokey, totally bookmarked!
03-04-2006 , 11:06 AM
When you are talking about stealing blinds - does that mean you're on the button and it's been folded to you or do you mean raising from the button with a few limpers too?

This is one area i've been working on, but I wasn't sure if it was profitable raising with a few limpers too
03-04-2006 , 11:54 AM
Vey nice post Pokey, I actually was wondering how to find out a few things you pointed out. One thing though.

I raise almost nothing pre flop. My pre flop AF is .11
Yet I have a 10.69 PT BB/ 100 winrate over the 54K hands in my poker tracker. Is pre flop raising really as important as people think? Obviously this translates to steals as well.
03-04-2006 , 01:26 PM
I might make another post on this, pokey said this was his opinion and I dont want to put him on the spot and I am curious how many other players dont raise pre flop.
03-04-2006 , 01:44 PM
Pokey,

excellent post. i just use poker tracker to see how much i win (or lose, [censored] january).

Mods: Please sticky this or something.

Rgrds,

NESS
03-04-2006 , 01:46 PM
Quote:
Vey nice post Pokey, I actually was wondering how to find out a few things you pointed out. One thing though.

I raise almost nothing pre flop. My pre flop AF is .11
Yet I have a 10.69 PT BB/ 100 winrate over the 54K hands in my poker tracker. Is pre flop raising really as important as people think? Obviously this translates to steals as well.
Scrappy,

WTF?! I have never heard of someone that pretty much never raises preflop and wins (with a nice winrate I may add).

Maybe this will generate some interesting discussion.

rgrds,

NESS
03-04-2006 , 02:09 PM
Quote:
Pokey,

excellent post. i just use poker tracker to see how much i win (or lose, [censored] january).

Mods: Please sticky this or something.

Rgrds,

NESS
ditto,,,,
03-04-2006 , 02:15 PM
what do you mean you dont raise preflop?????
and do u plya short ring or full???
03-04-2006 , 02:51 PM
Great post, I went through it step by step as I read it.

A few comments on pf aggression. My pf aggression is less than 1, but my postflop aggression is greater than 6. This is because I tend to limp into pots with somewhat mediocre hands with the intention of picking up the pot if nobody else hit. I think pf raise vs. VP\$IP is somewhat of a personal issue.

The stuff on blind stealing is great, I need to pull back (i'm at 45%... ) You also exposed a huge leak of me overvaluing TT's pair value in your section on pocket pairs. Thanks for that.

aaand... I am winning a lot of money from the blinds? how weird is that..

thanks pokey
03-04-2006 , 02:56 PM
Thanks for the positive feedback, everybody. Answering a few questions:

goofyballer said:
Quote:

If mine &lt;winrate when attempting a steal&gt; is insanely higher (83 bb/100) does that mean I should be stealing way more? I'm only at 10% right now. What range do you recommend stealing with in a typical loose NL 25-100 game?

As I said, I'd consider anything less than 20% to be leaving money on the table. There was a great thread on the Poker Tracker forums (see it here) where a guy claims that your winrate on blind steal attempts does not diminish markedly as you increase your steal attempts from 10% up through 40%. That information was specific to a limit game, I believe, but it wouldn't surprise me at all if something similar applied to NL. If you get a chance to steal a blind once every three orbits (about my average over the last 16k hands or so) and you attempt a steal 10% more often, that means you're going to try to steal one more time every 300 hands. If your winrate on a blind steal averages 75 PTBB/100, that change should add 0.25 PTBB/100 to your TOTAL WINRATE. Sure, it's not earth-shattering, but a gain is a gain.

Nuprin said:
Quote:
When you are talking about stealing blinds - does that mean you're on the button and it's been folded to you or do you mean raising from the button with a few limpers too?

Technically, a blind steal is when you are either in the Button or CO (one off the button), everybody in front of you folds, and you raise. Personally, I recommend attempting "quasi-steals" when one or two players have limped in front of you. If you've got at least a decent hand, your quasi-steals are going to be even more profitable: when they pick up the dead money preflop, it's more than twice as large as the usual blinds stolen, and when you get one caller you're still equally likely to win with a continuation bet or further aggression, but now the pot is triple what it usually would be. I'd be very hesitant to try this with true junk (K8o, T9o, that kind of thing), but if you'd normally limp, it's at least worth considering a raise instead.

scrapperdog said:
Quote:

I raise almost nothing pre flop. My pre flop AF is .11
Yet I have a 10.69 PT BB/ 100 winrate over the 54K hands in my poker tracker. Is pre flop raising really as important as people think? Obviously this translates to steals as well.

There are a variety of reasons why raising preflop too infrequently will hurt you:
1. When you've got the best starting hand, you're most likely to have the best hand at showdown. The more money you get into the pot, the better.
2. Every raise and bet you make has a chance of picking up the pot immediately; why pass up a chance to take down 2 PTBB with a hand like 98s? If you have a large database, check out your PTBB/hand for various preflop holdings and notice how few of them win you more money per hand than the 0.75 PTBBs that are in the blinds alone. If you've got one limper in front, the pot now has 1.25 PTBBs in it. Picking these pots up is a serious bonus.
3. Raising preflop sets you up to win the flop VERY often with a continuation bet. OK, so you got called; that's OK! You've now got one opponent who limp/called into a pot, and the board fell with garbage. Your opponent has checked to you. When you bet the pot here, you're going to take down the pot at least 2/3 of the time, showing a VERY tidy profit. Even if you get called, your hand can improve on the turn, or you can take a free river card. The odds of you making money on this hand remain extremely high.
4. Raising preflop makes hand-reading MUCH easier. So you limped preflop with A3s and five people see a flop that comes 743r. Is your hand good? You can't even GUESS at what BB has, since it could be literally any two cards. Anybody who limped with garbage could have hit this rather nicely. However, if you raised big preflop and narrowed the field to heads-up, your odds of having the best hand on this flop are MUCH higher; it's hard for a stray 7 or 4 to have survived a preflop raise. You can narrow your opponent's holding down dramatically: he's usually holding either overcards (that missed) or possibly a pocket pair (that might be getting scared). If he check/calls your bet on the flop, you can start to narrow his holdings, assuming he's more likely to have a small pocket pair, either 22, 55, 66, or maybe 88 or 99. If a big card comes, you can bet again and easily fold out a stronger hand. Sure, it can backfire on you, but there's nothing more frustrating than having your JJ die on a flop of 733 when the big blind had 83o.
5. Raising more hands preflop disguises your big hands more effectively. If you raise 10% of the hands you play, then your opponents can narrow down your hand holdings extremely effectively when you DO raise. However, if you raise 50% of the hands you play, and you change it up based on table conditions, your hands will be MUCH stealthier. He raised from the button; is that 65s or AA? These are questions you want your opponents to have to ask. Not only does this boost the winrate for AA by getting these hands called more often and building them bigger pots, but it also boosts the winrate for 65s by concealing it much more effectively. Any gain you get from making your opponents' hands more easy to read is mirrored by the gain you get from making YOUR hands HARDER to read.

Again, I'm really glad people are finding this post useful; I've been thinking about writing this post for some time now, and after another flurry of "do these stats look right?" posts, I figured now was as good a time as any.
03-04-2006 , 04:45 PM
This post is really great, and I've learned alot both about my game and poker tracker from it.

But I'm not sure your advice on "Post Flop Aggression" numbers is accurate, atleast for six max. If I add both my raise + bet percentage up I get 39.67 (36.72 and 2.95 respectivley). This doesn't quite meet your 40% mark that you advocate, but I almost always cbet, as I think my 11% check and 5% fold numbers indicate.

I think it must really depend on what your no flop/no action % is, mine is 42. The higher it is the more you are going to have to cbet to get up to your 40% bet+raise number. I think you may need to take that into consideration, atleast for six max players.

Great post. Thanks again.
03-04-2006 , 04:55 PM
Quote:
Your default play here should be to fold weak suited connectors and reraise strong ones.

weak suited connectors being 23,34, &amp; 45? and what are considered strong suited connectors?
03-04-2006 , 05:11 PM
Wow, this is an absolutely phenomenal post. I learned flaws in my game, such as that I am a loser with both suited connectors and unsuited connectors and that I need to be slightly more agressive postflop.

Pokey - do these numbers and suggesstions apply for full ring, 6-max, or doesn't it matter?
03-04-2006 , 05:34 PM
Quote:
I might make another post on this, pokey said this was his opinion and I dont want to put him on the spot and I am curious how many other players dont raise pre flop.
...you're joking right?
03-04-2006 , 09:41 PM
great post. I am somewhat new to NL cash games and this should help me a ton once I get some hands in the ole' DB

Shortcutted this post immediately BTW, I'm sure it'll end up in the FAQ
03-04-2006 , 10:09 PM
Great post...i am pretty new and i never really learnt how to use poker tracker properly...they were all just a bunch of numbers to me...now i think i can start to make a bit of sense of it all....
03-04-2006 , 10:54 PM
Someone that used to post in here and I wish I could remember who, said they rarely raised preflop, to include big pocket pairs. He had a very healthy winrate and was a well looked upon poster. His defense of this "style" was that he felt that the preflop equity he gave up was more than made up for by the postflop equity that he gained. I'll see if I cant dig up the posts, but with the search function on this system, dont hold your breath
03-04-2006 , 11:31 PM
Thanks for a great post. Definitely warrants reviewing some of my own stats.
Now on the PF aggression topic:
I think the game level ultimately plays a big part in your preflop aggression. If you are playing micro NL with a table of maniacs or call stations (standard micro) the equity of preflop raising can go down a bit. You can argue that it is still better to be aggro preflop in either circumstance but at micro/low limits this often leads to tears. Knowing that the other players thrive on raisewars (in maniac situation) or can't get rid of any 2 (station situation) do you really want to push your connecters hard preflop? I don't think this applies as you move up, even in higher SSNL, but if you know you are going to get called down when you hit it can be profitable to play less aggro. Again, I am talking specifically about uber SS before people tear into my line.

m