Hello everyone, first time posting in this forum.

I was reading a book last evening that talked about keeping track of variance. This is easy to do in internet world, but in the real world, it's a little more difficult.

My initial thoughts on the matter are that the system must first be accessible real time, meaning that you have to be able to do it quickly and at the table in a notebook or calender. Second, you have to track a usable and reproducible metric. In other words, you've have to have seen all the cards.

So, I play 1/2 and 2/5 NLHE. My first thought is that you keep a ticker of every hand that you are 1) all in, 2) have at least a 3-1 advantage, and 3) lose. Additionally, you might keep track of the number of hands you get all in on with the worst of it and win.

Does anyone out there do this, what metrics do you track, and what do you think of my idea for a system?

Thanks.

X

There are poker tracking applications, but there will be left one info when you don't know the opponents hand when you sometimes win the pot. Worry less about variance, and just try to learn winning poker IMO

Variance as in change in your bankroll? Everyone should do this. Track: date, hours played, stakes, in $, out $, expenses and bankroll total. Variance in equity versus result hand to hand is not going to happen unless you write down every hand - I did once play with a guy who did this with pen and paper at The Orleans in Vegas - I was new to the live game and found it really odd/weird. Hand to hand most memorise as many details of the big hands as possible to evaluate later away from the table with Poker Stove etc. As time rolls by you will get a feel for a lot of the more predictable scenarios that happen again and again as you gain experience.

- edit - just seen your post count so perhaps I mised your point as it's unlikely you're inexperienced player but maybe are live? If not ignore

Thanks, guys. I already keep track of basic stats in a calender book. Time in, out, buy in, location & win/lose. I was looking for a way to keep a finer tally of the frequency I am sucked out on when I get my money in with the best of it. This is more of a psychological/mental game exercise than it is about playing winning poker. I've gotten to the point where I am noticing a change in my play based on how bad I run. I lose confidence and play less aggressively than I should. This is a leak I need to fix.

In live play, keeping track of large statistics isn't an option. It might take a year and a half to play 100,000 hands. Often times, the sample size is too small and a single day's win or loss is going to come down to, usually, one or two key hands. Thus, if I want to track variance, I've got to put a finer point on my stats.

Hope that clears things up. Thanks.

I'm not sure that I have anything to offer, but I'd like to hear what other ideas you come up with.

I track my stats on a monthly basis. I log everyday, but check at the end of the month where everything landed. In a month of play, you might only play 10 really big pots. Swinging a couple either way will massively change your total profit for the month.

I like the idea of keeping track of all pots above a set amount of $. Keep track of all the information you have. Run poker stove and see where you should have been on those hands and what your real results were.

Sub - Poker Journal app tracks everything u Ned including standard deviation.

What does part 2 or 3 have anything to do with what you are looking for?

I think OP is confusing variance (standard deviation squared) with All-In EV, which many online players may refer to as "running good" or "running bad" when their actual win rate deviates from their all-in expected win/loss.

Poker Journal and others will do standard deviation and variance for you. Also you can estimate all-in EV by figuring out an accurate (be honest) range for the villain and running an equity calculator to find out what your actual equity was against his range and then calculating all-in EV for your live sessions...so if the pot was 100 and Hero is 60 percent to win, he will pick up 60 theoretical all-in dollars ).

Variance happens, there's no point in tracking it. It's like asking to see the river card when you fold your draw on the turn.

If you use an app on your phone like Poker Journal, after you've logged many hours, you'll see the variance in your charts and in your list of sessions. In my experience, the worst downswings are about 10 buyins, assuming you're playing well and not just donking off chips.

Thanks for the replies everyone. I will soon be upgrading my technology to the 21st century and will have a smart phone. Just behind the curve on that one.

@ livesinabarn - Parts 2 & 3 have to do with how bad I run, which is kind of the point of the exercise. As I said, law of large number statistics isn't going to apply to the magnitude of hands that you can play in a live session, even a long one like 18 to 20 hours. The way I figure, things can go 4 ways, which are all worth tracking over time. I could get it all in and a) have the best of it and win b) have the best of it and lose c) have the worst of it and win or d) have the worst of it and lose. (d is not really variance, it's just a bad misread or bad bluffing, but it does happen occasionally) All four are probably worth tracking.

@ cowboy - Bad runs can go on for a lot longer than 10 buy-ins. I've had bad runs in computer land that lasted 10,000 hands & more. To play that out live might take 3 months and might set you back 20-30 buyins depending. It's sobering. When you are running bad, the worst thing that could happen is playing worse because you get beat down mentally. Without tracking specific bad beats in some way, you might doubt your play or change it in some way that isn't +EV. So, there is some benefit, psychologically speaking, to objective evidence of cold cards. It is reassuring to know you are getting people to put their money in with the worst of it, even if you are continually getting beat.

@ kimoser - very possible I am using the wrong term here. I tend to think of standard deviation as a mathematical expectation and variance as something that actually happens.

I said "downswings," not bad runs. I've had some long stretches of being break-even, and I've suffered through several 10 buyin downswings in the past five years, but nothing worse than that. If you experience downswings of 20-30 buyins, i'd be willing to bet my roll it's your play that's to blame, not variance.

You're probably right. If I was nit, my swings would be smaller.

But we're getting way off topic if this is going to turn into pissing contest about who's a better player. I take it that you do nothing to keep track of how bad you're running beyond what you're little phone app tells you?

You probably want to come up with a better set of terminology to avoid confusion. "Variance" as a term of art in poker is typically understood to also be a mathematical construct having nothing to do with "what actually happens", it is instead a measurement of distribution about a mean. it is nothing more than the square of the standard deviation (or rather, SD is the square root of the variance): http://en.wikipedia.org/wiki/Variance

Certainly possible that OP's play is to blame, but (if player is an overall winner), if you approach this mathematically, you'd be at considerable risk of losing your bet.
Also, in your own case, if you're already capable of experiencing one downswing of 10 buyins roughly every year or so (which isn't unusual, even for excellent players), the likelihood of your experiencing two of them in a row (or separated by a breakeven period), sooner or later, is actually pretty high over several years. I hope it never happens to you, but odds are, it will, sooner or later.
Variance is a bitch....

One of the tracking programs I use, has a customizable field available, or you could use a spreadsheet and customize yourself. I used to track by game/table selection, as in is the game "soft", "moderate", "difficult". But as I've learned to table select better, and move tables if my game is not "soft", this value has become less useful to track for me.

So I've recently thought about tracking suckouts and coolers that affect my session dramatically as well. I also think there's value in seeing if a downswing is play based, or just a run of bad luck.

Today for example, I was playing great, had command of the table, and lost most of my profit from my session in one hand where I flopped a boat. I had a player call me down with good sized bets on a flush draw(ie I allowed him to make mistakes, and he did). The turn and river came QQ, and his Q10 flush draw became a higher boat (flop 667) and I had pocket 7s. There were 3 players in this hand, and the pot was about $400 at a 1/2 nl table. I ended up only $68 after a 4 hour session, when it would have been about $468 if I don't lose that hand.

Unfortunately, I'm currently in a run of this sort of thing happening. I've been losing a lot of monster favorite hands. I believe starting to track these as true suckouts and coolers, will help me stay focused on playing well, and not fix what's not broken.