Stats Post

[Bona]

No answers but some questions:

According to the graph you played 10k hands in 22 days. Thats a lot. Do have/use any study or review time for hands previously played? I know it varies a lot from person to person but I only spend about 2/3 of my "poker time" playing the rest is review, the forum, the books, in that order.

I have a tendency to play weak tight myself and one of the ways I am fighting it is playing some six max. You have to loosen up or die in six max (or so it seems to me)

Try playing the hands as though you were one seat farther to the left. That is if you are in MP3 play it like you would in the CO. If in the CO play it like you would on the button.

[NIX]

I do believe losing 0.22BB/hand in the BB is pretty rough. You should be a long term loser in that position, but improving to -0.15BB/hand or something wouldn't be unreasonable and would up your overall winrate a little bit. Again, some 6max can help with blind play.

[Octopus]

[ QUOTE ]
I do believe losing 0.22BB/hand in the BB is pretty rough. You should be a long term loser in that position, but improving to -0.15BB/hand or something wouldn't be unreasonable and would up your overall winrate a little bit. Again, some 6max can help with blind play.

[/ QUOTE ]

I would not be too worried about these numbers at this point, particularly on this sample size. A survey in SS a year and a half ago (link) had the forum averages from the blinds at -0.18 and -0.11 at 2/4 and -0.19 and -0.06 at 3/6 (with its 1/3 blind structure).

[Octopus]

You could be at least a couple points looser from every position. For example, look at your UTG stats. I almost never say this, but you could play more hands up front. Your stats show a VPIP of 5.3% from first position. That is something like TT+, AJs+, AQo+ and KQs. I would *raise* with all those and more. I am not much of an early position limper (actually, it has been about 20,000 hands since I last limped first in), but at a passive table, all kinds of stuff becomes limpable, like any Axs, any pair and a bunch of suited connectors, for example.


But most importantly, stop looking at win rates after 10,000 hands. It just doesn't mean anything. The standard deviation of your win rate at this point is something like 1.7BB, meaning your true win rate is probably somewhere between -3.1BB and +3.7BB per 100 hands. (Technical note: That assumes 10,000 hands is enough to make the distribution at least close to normal. I have seen some results which make me doubt that this is the case, though it is probably not too far off.)

[NIX]

[ QUOTE ]
[ QUOTE ]
I do believe losing 0.22BB/hand in the BB is pretty rough. You should be a long term loser in that position, but improving to -0.15BB/hand or something wouldn't be unreasonable and would up your overall winrate a little bit. Again, some 6max can help with blind play.

[/ QUOTE ]

I would not be too worried about these numbers at this point, particularly on this sample size. A survey in SS a year and a half ago (link) had the forum averages from the blinds at -0.18 and -0.11 at 2/4 and -0.19 and -0.06 at 3/6 (with its 1/3 blind structure).

[/ QUOTE ]
Yeah, that's probably a better estimate that what I remember reading. Good link.

[mojoyoyo]

Here are my comments:

You are a winning player over 10K hands. In and of itself, this is a good thing. You are experiencing a downswing in the 100BB range. You don't mention your experience level, so I am going to take a stab that since you play .50/1, 10K is pretty close to the whole thing.

1. Statistically, a 100 BB downswing full ring should be viewed as a *common experience*. I look at my last 25K hands and they include at least 4 distinct 100 BB downswings, a rough 12K sideways swing with at least 5 attempts to pierce the top, and many smaller downswings. I did a post a while back that showed with a year's play and a sample size of 10K players, you would expect a 7% chance of a 200 BB downswing and almost a full percent chance of a 300 BB downswing, this is not to mention the poor s.o.b. who gets the honor of having the 500 or 600 BB downswing. I recommend wholeheartedly Mason Malmuth (2+2's book) called "Gambling Theorey and Other Topics" - in it he shows how damning the standard deviation can be in the short/medium run. In my simulations, the good news is that winning players win money at the end of the year despite significant downswings.

2. As a computer scientist for 20+ years I can say that the proper way to debug a situation is to change at most 1 variable at a time. You are obviously in a situation in which you do not know what to change, and are looking for advice. I would suggest that it is at least feasible that you shouldn't change anything right now. If you change one thing or two things right now, you might get 5K hands down the road and find that the changes are for the worse, that the changes had a positive effect but are incorrect which will hurt you in the long run, or some other insidious combination which throws off your baseline play which you are playing right now. I would start by reading and re-reading SSH. Look for things that you can tweak (I use the word tweak very specifically, I view poker play that is a key that has been cut, and errors in play are a burr on that key - fixing errors is a process of small tweaks, not large adjustments (most times)) that are *clearly* correct. For example, a while ago I noticed that I was a loser with the weaker offsuit hands, namely KT QT JT KJ QJ. I reviewed all the 2+2 material I could find and decided I was misplaying these hands. Now, I rarely play these hands outside of the blinds, and when I do, it is late. Another area you might want to look at is the number of raises you are calling. SSH recommends playing small pairs for 1 or 2 bets in a lot of cases - but there is a big difference between being the 7th person to put 2 bets in a pot and being the first person right after the raiser to put 2 bets in the pot.

3. Be careful of internet advice (including this post) as a lot of it is just plain wrong.

4. Be careful of 6max play. I am not saying don't do it, but understand that if you are a 1.5BB/hr 12BB stdev player full ring, your stdev might reasonably go to 25BB/hr 6max. If you have a 300BB roll full ring you might find you need a 600BB roll 6max. If you don't know the difference between ABC poker and Attack and Defend poker, and how to identify which mode the game and/or hand is in, you will most likely not have a pleasant 6max experience. If you don't know how relative card values change when going primarily heads up (hint: pairs and high card strength go way up in value) you will most likely not have a pleasant 6max experience. 6max play can be useful but I would make a wager that most people don't know what they are getting in to. There is a lot of incorrect information on 6max. Ed Miller says you should not play looser or more aggressive 6max, its just that your blinds have more equity - yet the 6max starting hand guidelines *on this very site* recommend raising utg (mp2 6max) with KT!!! KT is a marginal hand full ring and an awful hand 6max (except perhaps very late). You will find misinformation like this all over. In any endeavor 50% of the people will do better than average (by definition), so remember that a lot of people succeed *despite* what they do not *because* of what they do.

I would suggest signing up for the 200 hand exchange and review thing on this board, if someone looks over some of your hands, you might get a perspective on things you hadn't considered.

I would suggest essentially playing on unchanged for now. Look for specific holes that you want to plug and address these systematically. Remember that even though (at this point) the downswing seems to have come out of nowhere and all of the sudden you are down 100BB the turnaround will be gradual and filled with failed attempts. Grinding it back is a slow and sure process. Have confidence in the things you have done to get you to this point - playing outside your roll and/or making large random changes to your game will probably have the results that you might imagine. Be glad that your downswing came after a winning stretch especially if you are just starting, now you have the motivation to learn and re-learn whereas the other way around it would have probably been just frustrating.

Good luck.

[milesdyson]

[ QUOTE ]
4. Be careful of 6max play. I am not saying don't do it, but understand that if you are a 1.5BB/hr 12BB stdev player full ring, your stdev might reasonably go to 25BB/hr 6max.

[/ QUOTE ]
no

[ QUOTE ]
If you have a 300BB roll full ring you might find you need a 600BB roll 6max.

[/ QUOTE ]
not really.

[ QUOTE ]
There is a lot of incorrect information on 6max. Ed Miller says you should not play looser or more aggressive 6max, its just that your blinds have more equity - yet the 6max starting hand guidelines *on this very site* recommend raising utg (mp2 6max) with KT!!! KT is a marginal hand full ring and an awful hand 6max (except perhaps very late).

[/ QUOTE ]
i just opened up the hand chart thread located in the small stakes shorthanded forum, and it recommends raising KTs 3 off the button and folding KTo 3 off the button. it recommends raising KTo 2 off the button (in the hijack). this all seems fine to me.

[mojoyoyo]

I stand corrected on the KT thing.
But in my defense, I did say that things on the internet (including this post) are often wrong. ;>

What do you recommend is a valid stdev for 6max play? Its more variable than longhand, so does saying "you might find you need double" strike you as wrong? If so, are you saying that no more of a roll is required (therefore that it is not more variable)? If you say that it is more variable then a bigger roll is needed, are you saying 2x is not a reasonable suggestion whereas 1.5 or 1.3 or 1.7 would be reasonable? I didn't get a whole lot of info from "no" and "not really".

[Octopus]

[ QUOTE ]
I stand corrected on the KT thing.
But in my defense, I did say that things on the internet (including this post) are often wrong. ;>

What do you recommend is a valid stdev for 6max play? Its more variable than longhand, so does saying "you might find you need double" strike you as wrong? If so, are you saying that no more of a roll is required (therefore that it is not more variable)? If you say that it is more variable then a bigger roll is needed, are you saying 2x is not a reasonable suggestion whereas 1.5 or 1.3 or 1.7 would be reasonable? I didn't get a whole lot of info from "no" and "not really".

[/ QUOTE ]

It depends on your style and, especially, the style of your opponents, but I have found my SD at 6max to be only marginally higher than at full. Most people seem to see around 17%-20% at 6max (and 14%-17% at full).