Moving up to 10/20 or adding another 5/10 table?

[srw5n]

Now it is you who didn't read the comment. The post said chessmaster is being too result oriented. And my contention is that looking at long run statistics is being result oriented, in a literal sense. That was my point.

My initial comment was actually in agreement with the majority of what you said. I never contended that 5K was an ample sample size, BUT I simply said that if one was winning over 5K hands I would anticipate they were doing "something" right. Similarly when I have had a downswing, I have usually been doing something wrong (in addition to everyone's favorite excuse the dreaded variance). Poker is not like chess or other games devoid of chance. When you lose in chess you must accept that you were beaten. When you lose in poker you can still blame chance.

My point is simply this: many believe that they are playing good poker and are simply on a protracted downswing. I would heartily agree with you that 5K is not an adequate sample size, but I have yet to seen any collective data establishing an adequate sample size. Not having any of my own raw data in front of my I would wager that the something around 15K would suffice, however, poker is vastly different from other arenas and involves a large degree of personal autonomy and growth, so to get a true number I would prefer 50K (again this is conjecture without having my raw data in front of me to make a more educated guess).

Complaining that I do not understand statistics is not useful, I do, I was in antitrust economics for a couple years, but who cares you don't need to be convinced and I don't really care to...it doesn't effect either of us...

My main point is that "variance" is used as the great lie on this forum. Any time someone loses it's "variance". I think blaming variance for losses and gains is counter productive - I for one know that anytime I have dropped 100BB it has been more than variance that lead me in that direction. Variance is a caution to those elated with their short run winnings, BUT if chessmaster wants to play 10/20 let him.

[Fiddler]

This graph posted in the Micro-forum a couple of months ago illustrates how long the "long run" can be for a 2BB/100 winner with SD that is 16BB/100. Kind of depressing.

[srw5n]

Thanks Fiddler for hunting this down. A 16BB SD for a short handed player is probably generous too...My guess is the figure is probably closer to 18, so that can be even more alarming.

I simply feel that knowing variance will play a roll in our results, it would be more valueable to look at "downswings" and see if we are playing well or if we have leaks. Further, it would equally beneficial to look to "upswings" to examine what one is doing correctly.

[naphand]

[ QUOTE ]
Not having any of my own raw data in front of my I would wager that the something around 15K would suffice...

[/ QUOTE ]

You would lose the wager.

There have been extensive posts on this matter, they may be in the archives now, dealing with the math of what consititutes a "reliable" WR figure. 15K is laughable, the very least you need is something around 100K, and possibly twice this much. If you want to read do a search here and in the archives by searching for threads in small stakes forum that have "SD", "Standard Error" "Variance". You could refine the search better by also making sure the posts only include threads with usernames "Kiddo", "Naphand" as the ones I am referring to were posted/discussed with us.

Here are some:

Thread1

Thread2

Thread3

These threads also contain links to other similar threads, so you should have plenty to read.

[ QUOTE ]
Now it is you who didn't read the comment.

[/ QUOTE ]

I read what you said. It still reads the same to me.

[ QUOTE ]
I for one know that anytime I have dropped 100BB it has been more than variance that lead me in that direction

[/ QUOTE ]

This has also been widely discussed, most agree that on a big downswing they play less well but that this is only a contributory factor, and in no way accounts for the large majority of that swing. I prefer to listen to the advice/experience of those who have played 100-200,000 hands at this level and talk of these swings (and still maintain excellent WR). 100BB is also a downswing of tediously small proportions. Again if you were a regular reader you would know that many are not aware of the kinds of swings that can and do happen, I think we are talking 200-300 BB.

[ QUOTE ]
My main point is that "variance" is used as the great lie on this forum

[/ QUOTE ]

By the most experienced players? If you are referring to the less experienced I would say this "great lie" is in fact smaller than the "small sample, high win rate" lie that is commonly known as unrealistic expectation.

I do, however, accept your premise that people easily accept that their downswing is not their responsibility. There have been articles on the psychology of this that make very interesting reading. Guy McSucker is one of the few players big enough to refuse to accept this position and go back to rebuild his game, which kind of implies that many players will just blame "variance" or "bad luck". But there are two sides to this coin; there are an abundance of posters who refuse to listen to the advice on the Forum simply because they are short-term winners. People post 7-infinity BB/100 win rates over small sample sizes as if they were meaningful (they are, but only to one person), we have had plenty of posts that talk as if the collected wisdom here only applies "in certain circumstances" and that weak-passive play is a profitable form of poker. It is interesting, though not surprising, that losers can be easily convinced their game is weak but not that bad, as natural variance may account for a lot of their losses; a similar number of players will not accept their game is weak as they are (short-term) winners and this is unlikely to be a consequence of natural variance. Now THAT is funny.

[ QUOTE ]
if chessmaster wants to play 10/20 let him.

[/ QUOTE ]

No-one is trying to stop him, the purpose of this forum is to educate, not just how to play but what to expect. If someones expectations are unrealistic then it is correct to caution, not just for the sake of the poster, but for the many others who read this forum.

When you responded to the "results oriented" line, you were also mis-representing what he said. When people talk of results-oriented thinking they are pretty much referring to the results in front of them, not the last 100,000 hands. So someone who has a bad session is being results oriented by assuming they cannot play, or a bad play on a given hand turns out good does not make it the right play long-term. "Result-oriented" herre is referring to short-term results and is completely valid. Your interpretation of it is not invalid as it could be interpreted that way, it is just incorrect in the context of what the poster was saying and therefore of no use to the discussion.

I am glad to have someone experienced in statistics willing to post, though I think what we need is some statistical/mathematical rigour not "conjecture" and estimates. When you can provide this kind of analysis to support your assertions, then your posts will be devoured with enthusiasm and treated with respect. Please, no more conjecture, especially from those who should know better.

[chessmaster]

[ QUOTE ]
How would losing $2k in 3 days feel to you?


[/ QUOTE ]
I had a strange day yesterday. After 2 hours i was about even with swings from +20BB to -20BB and sat down at a quite loose table where i quickly won 20BB. But then i faced a guy who hit about every draw he had against me (and sometimes it were only longshots), I lost with aces, kings, two pair or even top set against some crap hand rivering a gutshot or some silly backdoorflush. I think i had about seven hands against him were i was 80-90% favourite at the turn but lost anyway and there was little to compensate for that. Somehow this affected my game and i lost a further 15 BB with playing too loose, so i lost about 1.2k in one and a half hours and was at -40BB for the day. After i realized not playing too well i took a break and sat down at a new table afterwards to make a new start (probably i should have done this much earlier). Here everything went fine and i won it all back in two and a half hours to finish the day at +6BB.
But i didnt lose my confidence yet because i know this guy wont continue hitting everything and in the long run he will be punished for his bad plays. I just have to work on not being affected by a bad run like this.

[srw5n]

good post. I continue to disagree with you on some points, but you were well thoughtout.

I continue to hold that discussions of variance regarding poker are useful only if done in context. Variance, to use an analogy, should be a warning light. If you're going too fast and killing a game for 8BB/100 you should look at your game and see if you're simply on an upswing. The same warning light should go off when you drop 100BB or the like. BUT more important is an examination of the raw data. How did you play particular hands? Where did you miss bets? Where could you have saved bets? How are you selecting your tables? etc.

However, what continues to trouble me is the way variance is used by many on this forum as an excuse, and anyone who is winning is met with the conclusive assumption that they are merely experiancing an upswing, because no one can win more than 2.3456 BB/100 or what ever the going rate is... The reality may be that player X is on an upswing, but without having the raw data infront of any of us an equally legitimate conclusion could be that player X is very good and has superiour table selection... the truth is we don't know. However, individuals post conclusively that player X cannot possibly be this good, yada yada yada and he is merely on an upswing.

Reading your posts I see you're not one of the donks who do this and I appreciate that. I have been frustrated in recent months by the proliferation of pokertracker and the improper use of the statistics generated from this excellent piece of software.

Anyway. Good post. I'll add you to the list of individuals who I read. Good luck on the tables.