Lies, damn lies, and statistics

[Irieguy]

I'm starting to think that keeping stats may be bad for my game.

I've been playing poker for 15 years, and winning for a few without ever keeping track of anything other than winnings I had to claim on my taxes. Then internet poker comes along and it's like a glorious gift from the heavens. These people are bad, bad, bad bad bad, and I finally see the much over-used poker phrase "giving away their money" applied to a group of about 40,000.

I'm cashing out more than I'm buying in and I'm happy. But all the smart people keep talking about their ITM and ROI and EV and I don't see how it matters much, but I read and think a little more and decide I should keep track and see how I do. So, I play 500 SNGs and track my ROI and my ITM and I prove to myself that I'm winning... and now I know how big my johnson is compared with all the other people who claim to have big johnsons.

But here's the problem: I used to have two kinds of poker emotions-
1. Running good (happy), and
2. Running bad (sad).
Now I have 8 poker emotions:
1. Winning at a rate greater than my long term ROI (happy)
2. Winning at exactly my long term ROI (indifferent)
3. Winning at slightly less than my long term ROI (sad)
4. Realizing that I'm negative for 10 tournaments (pissed)
5. Realizing that another OOTM will drop my ITM% below 42% (depressed)
6. Realizing that I've lost 78% of the time with the best of it over the last 57 showdowns (raging hatred)
7. Realizing that I'm setting a personal OOTM record (homocidal)
8. Completing detailed analysis of past 100 SNGs, deciding to stop playing between 1200 and 200pm, tightening up 5 handed, eliminating the suited-connector limp from my early stage play, becoming less aggressive with weak kings short-handed and more aggressive with big Queens, and finally figuring out how to stop leaking when getting check-raised only to get 7 consecutive 4ths and 3 10ths in the next 10 $55 SNGs (suicidal).

Then I read Aleo's post about his statistically improbable super-turbo-horrendo bad spell and realize that most of his mental anguish is being caused by the fact that he knows precisely how improbably his streak is and how long it will take to repair the ITM and ROI% he worked so hard to build over months and months and a thousand SNGs.

Hmm. Maybe a cashier history is the only spreadsheet I need. Maybe hand histories on the replayer is the only part of a result I should review.

Play the right way,
Irieguy

[codewarrior]

Do you see why?

You make some great points, and it is well written.

I may expound more later.

Welcome to the forum.

[dethgrind]

You are completely right.

I've spent hours and hours wasting my time thinking about risk of ruin and bankroll management, and what a good ROI is, and what my "real" ROI might be given my results. I've wasted hours of my life tinkering with computer simulations. Enough.

There are only two valuable pieces of information I've gotten from all this:
1) I'm playing small stakes. If I go broke, I can easily replace my bankroll.
2) Even if I'm as good as I think I am, I'll probably still have some huge losing streaks.

I guess it was nice to attach some numbers to those two lessons, but that doesn't really matter. Now I really wish I had just spent all that time studying, improving my game, and just playing.

Thank you Aleomagus and Irieguy.

[NotReady]

If you only play for fun, don't keep any records (seriously).

If you're serious about poker, want to move up, play as high as your bankroll will allow, keep all the records you can. If you can't handle what the records say, play for fun.

[pzhon]

If you play for fun, and you don't find statistics fun, don't maintain more than you need for tax purposes.

I enjoy figuring out how much I win. This can help to answer the following questions:

/images/graemlins/diamond.gif Do I make more from poker than my day job?
/images/graemlins/diamond.gif Should I move down in limits?
/images/graemlins/diamond.gif How much do I need to play at a higher limit to determine whether it is worth moving up?
/images/graemlins/diamond.gif Do I win more in SNGs than ring games?
/images/graemlins/diamond.gif How much more do I win more when multi-tabling?
/images/graemlins/diamond.gif How much can I withdraw?
/images/graemlins/diamond.gif Is it worth it to take advantage of a promotion that takes up time or locks up my bankroll?
/images/graemlins/diamond.gif Have I improved in the last few months?

In addition, I think many players don't know how good streaks or bad streaks look from up close. Sometimes I don't notice that I have been running abnormally well.

[Dooner]

I totally relate to what you are saying. Then again, in the back of my mind, the old saying "Ignorance is bliss" speaks up. I'm not sure which way to go. Some people may really be helped by stats and others might be hindered by them.

Over the last year, I've started keeping track of my winnings in ring games. I also keep track of how many SnG's I've played in and how I placed (i.e. 1st, 2nd, 3rd, and out of the money). Then I add the total winnings from both together and that is all.

I don't want to know percentages. If I hit a bad streak, I might focus more on trying not to go below a certain percentage than what I need to do to break the streak. I like to use my bad streaks to analyze my game. A tweak here and tug there and usually come back full steam.

But, for some people the stats really help them take a deeper look into their game. So, if the stats aren't helping you, tone down. Don't throw everything out the window, because keeping some stats may help you in the long run. Hope this helps.

/images/graemlins/cool.gif

[t_perkin]

[ QUOTE ]

I like to use my bad streaks to analyze my game. A tweak here and tug there and usually come back full steam.


[/ QUOTE ]

The whole point of a bad streak is that it it not your poor play but the bad run of cards that is causing you to lose. Adjusting your game will not prevent bad streaks.

In fact one reason people look at stats and such like is to help distinguish between bad streaks and bad play.

Tim

[Irieguy]

Yes, Tim, exactly my point. I'm not disputing the fact that a statistical record of your play is the only thing that can "prove" whether or not you are winning. But once you are a winning player, I don't think statistics such as ITM and ROI are going to be helpful (for me, at least) along the conitinued path towards improvement.

It seems like human nature to start to look for leaks when you're running bad... but I think you need to tread cautiously here. You should be looking for leaks all of the time, and I think the best way to find them is to have somebody else watch a HH on the replayer. If you are running badly (ie, getting unlucky), and you respond by changing your game, then your record keeping may be hurting you.

Here's a question about which I'd be interested to hear some comments:

If you met two players and asked how they were doing, and one said:
"Great, I play all of the time and cashed out for 30K this year, here's my cashier history."
And the other said:
"I played 1000 SNGs, had an ITM rate of 40% and an ROI of 24% at the $100 level... but I think I can get that up to 30% next year if I plug a few leaks."

What would you think about each player? Which player do you think would be happier? I don't know, I'm still using my spreadsheet, just wondering how helpful a tool it actually is.

Play the right way,
Irieguy

[ilya]

Great post, Irieguy! And by that I mean "I agree." No, but seriously. I do think people spend way too much time thinking about their ROIs and such. On the other hand...it's good to check up on such things once in a long while. I've set up my poker spreadsheet so that it only lets me see my ROI and ITM once a week....works for me.

[Pubknight]

There is one thing about tracking/statistics/etc that may not be a factor for Irieguy, but is often overlooked.

Almost everyone that plays poker thinks of themselves as a good poker player... but as we all know, that just cannot be true.

It's easy to remember cashing out $1,000 ... but selective memory has us overlook that one may have done 5 different $300 buy ins over the year prior to that cashout.

I guess my point is, while I understand Irieguy's great post about the difference between being happy, and being a stat maniac, without that stat tracking a lot of people will be able to falsely convince themselves they are making money. Or, they aren't making as much moeny as they think they are.

[kyro]

[ QUOTE ]

7. Realizing that I'm setting a personal OOTM record (homocidal)


[/ QUOTE ]

Homocidal, the tendency to kill gays? Yikes!

/images/graemlins/smile.gif I'm kidding, nice post. Enjoyable to read.

[sabre170]

This is one of my all-time favorite posts.
I wish I'd written it.
Well said.

[Irieguy]

"gays" is NOT the prefered nomenclature. Heterosexually impaired.... please.

[Dooner]

[ QUOTE ]
The whole point of a bad streak is that it it not your poor play but the bad run of cards that is causing you to lose. Adjusting your game will not prevent bad streaks.

In fact one reason people look at stats and such like is to help distinguish between bad streaks and bad play.

[/ QUOTE ]

I totally agree with you on most bad streaks being a bad run at cards. However, unless you are a robot, everyone is human and will slip from time to time and play a weaker game than usuall. The problem I have with relying mainly on stats is that you might overlook your weakness, and blame it on the cards. So, when I hit a bad streak, I look at my game first and make sure that I am not slipping in one way or another. After that, if it is not something I can fix, I know it is just a bad luck streak and not a bad play streak.

But this is all besides the point. I personally do not like keeping all kinds of stats, but for some people it might really help their game. And for those, I say "Good for you". But, the original post was about not liking all the stats, and I agreed with him. Either way, the overall game of poker gets better as the players do. So do what you must to make your game top notch.

/images/graemlins/cool.gif

[t_perkin]

I posted because the post I replied to suggested that a "bad run" could be fixed by adjusting ones game.
This suggests that by playing better you can avoid all bad runs. This is not the case.

Most posters on this forum, when they say they are having a "bad run" they mean a bad run of cards, not just playing badly.

[ QUOTE ]

So, when I hit a bad streak, I look at my game first and make sure that I am not slipping in one way or another


[/ QUOTE ]

You should be analysing your game all the time, not just when you are running badly.

[ QUOTE ]

The problem I have with relying mainly on stats is that you might overlook your weakness, and blame it on the cards.


[/ QUOTE ]

Stats don't lie. If you misinterpret them then that is a problem.
Nobody pretends that (ROI, RoR and ITM%) stats tell you *why* you are losing/winning, only whether you are or not.
They also tell you how likely it is for something to occur. If you play 250+ SNGs a week then you are likely to see long losing streaks. Understanding the statistics gives you the confidence NOT to tinker with your game when there is nothing wrong with it.

It is just an important part of understanding the game of poker.

Tim

[Irieguy]

"understanding statistics gives you the confidence to NOT tinker with your game when there is nothing wrong."

Tim, I think that's the most profound comment in this thread so far. The hard part is managing the information while you are accumulating a large enough sample to gain your confidence.

I think it's clear that you need several hundred SNGs before you "know," statistically, how well you are doing. But I think you can figure out whether or not you are winning much sooner than that. Do you usually have the best hand when significant portions of your stack go into the pot? If the answer is yes, then you are winning. You are winning even more if the few times your money goes in with the lesser hand you were there first and there was a significant chance that your opponnent would fold a better hand. (By the way, I'm pretty sure you can beat SNGs without ever bluffing. You can just beat them more by bluffing appropriately.)

Perhaps a log of significant hands would be a more helpful tool than a spreadsheet.

Play the right way,
Irieguy

[Cry Me A River]

I can't help but think this is no different from any athelete who takes his game seriously - If you're going to be a pro or world class athelete in this day and age you pretty much have to be watching film a substantial amount or time, you pretty much have to be on a very specialised diet, you pretty much have to be scrutinising the technical aspects of your game at the most minute level, you pretty much have to be prepared to have a coach tell you how you need to change your form. I don't expect any of these are particularly 'fun' but they're essential to being competitive.

Which of course is completely different from a guy who just plays pick-up games in the park. There's a world of difference between "casual" and "serious" in anything. There's nothing wrong with wanting to take it a lot less seriously - You wouldn't be the first "athelete" to retire at the peak of his career.


I'm sure no track star likes to be told they're running "wrong" or that they've been nowehere near their personal best in the last 4 meets. But how else are you going to improve if you don't learn and build on what you've done before? I think the mistake here is not the use of statistics, it's their misuse. You shouldn't be using the stats to beat yourself up over a bad run* you should be using them to keep solidifying and improving your game.

Sure, any team that goes on a 10 game loosing skid is going to get down on itself, but any coach in the world is going to tell you the only way to get out of it is to emphasise and build on the positive; Keep it simple, don't try to do too much and make sure you get the fundamentals right then the rest will come naturally.



*You can't tell me when you weren't tracking your stats that you never got upset when you were on a bad run...

[Doubling12]

You should read John Feeney's Inside the Poker Mind. Since reading that book, I keep a detailed record on everything *except* P+L. The game is way more fun now, and (I think!) I play better for it. It is ridiculous to, for example, wimp out on a clear-as-day resteal opportunity because if you bubble again, your ITM% will drop to a new low. I review hand histories and keep a lot of notes, that's it...

[Dan Mezick]

CryMe:

There is a new service at www.PokerCoaching.com (http://www.PokerCoaching.com) that is related to what you are saying here, check it out.