Tuesday, August 07, 2007

Streaky Up, Streaky Down

The one thing about poker variance that's sometimes hardest to understand is how streaks within streaks can go in opposite directions at the same time. One would like to think that winning or losing streaks are in some form influenced by how well or not well one is playing. That is, when one is hot and confident and playing well, it seems reasonable to expect that the streak extends across all sites where one is playing, or across all formats (assuming comparative skill) or whatever.

It's really not the case. The vagaries of random distribution are such that one can run hot as hell on one site, could as ice on another, totally independent of whether one's results could be looked at as showing tilt or not. In fact, the variance can get so extreme, it can challenge the concept of what tilt really is.

Case in point, ready to be served.

I mentioned in my previous post how well I'd been running over at the Merge network, and the last week, since that post, has seen more and better. I could barely miss. The nightly $3K Guarantee that I mentioned last week, that I whined about being reformatted? After it's reworking, I cashed in the thing the next four nights in a row, including two final tables, bringing my streak to seven and three (including a win) overall. And then to top off my best-ever week on the site, I did *real* well in the $20K. I was a bad beat away from freakin' awesome, but having delivered my own rancid beat to someone earlier in the tourney, I ain't complainin'. I got by far the best side of the suckout trade.

But random distribution has a flip side, too.

I also played some at Bodog. Mind you, I'm fond of the site. I managed to find time over the weekend to play a couple of dozen tourneys, both MTTs and SNGs, and for the weekend I achieved a goose egg. Absolutely nada. Given that there were at least a half dozen SNGs in the mix, you'd think I'd be able to squeeze out a measly cash somewhere in the entire run. But nope, not a one.

So, running good, running bad. To tilt or not to tilt? That's the curious thing. The truth is that a handful of recent showings that seem unusual have a way of sticking in the mind, whether those showings are good or bad. These 'signature' showings can cloud our judgment as to how well we're really running... or if we're running any different at all.

The truth is that short-term results mean squat in terms of real skill, but it's so easy to forget that. A few dozen events, or even a few hundred, mean very little in the absence of true statistical outliers, something I've mentioned previously. Cash in seven of ten MTTs? Or even seven straight? Good run! But nothing more than that. It all ends soon enough, and then it's time to work for the next cash, however long it takes.

No comments: