Brought to you by Americas Cardroom

On Patterns

The thing is, patterns don't exist. Things happen, entirely at random, and we instinctively attempt to rationalise, formulate, and categorise them under the totally arbitrary and misleading title 'patterns'. Do any of the following sound familiar to you?

'I was running like dogshit so I quit.'

'I lost my first five coinflips and I thought 'uh oh, it's gonna be one of those sessions!'.'

'These big combo-draws haven't been hitting for me recently, so I decided to just fold instead of jamming.'

In poker, the equity of decision-making should exist entirely in a vacuum; or rather, the swings of fortune/misfortune have no bearing on the mathematical correctness of each decision. Just because you are, in fact, running like dogshit, doesn't mean that you should eschew the mathematically best decision (ie. shoving) in favour of the 'safer' option of folding. When that happens, you are letting variance win. You are allowing the totally random will of the Poker Gods to influence you towards making a sub-optimal decision. So much for poker being a skill game!

Just think of how much mental energy and focus you could free if you stop allowing these arbitrary 'patterns' to permeate your decision-making. They don't exist, and yet poker players expend so much focus trying to rationalise them. Let's take a look at my results over the last ten days:

Day one: + $807
Day two: + $347

This is going so well! I've cracked poker! I'm a genius! The poker Gods love me!

Day three: - $1571
Day four: - $528
Day five: - $189

Jesus, I suck. The poker Gods are punishing me. I run so bad. I hate this f game!

Day six: + $1179
Day seven: + $667
Day eight: + $263
Day nine: day off
Day ten: + $138

Cracked it again! Poker is my BITCH!

These ten days yielded an overall win of + $1113. I like to earn money, so this is good news. However, notice how completely arbitrary my means of demarcation are! Had I elected to use a five-day sample rather than ten, I would have been stuck $1134 - rather less impressive. And had I gone for a twenty-day sample, guess what? Back to being a genius again: a win of $1962!

Now, it looks as if my winning and losing days come in streaks. I win for a few days, lose for a few days, win for a few days. There must be something behind that, right?

Well, for me the answer is no. I have a deep-enough understanding of variance by now to recognise that this is just chance, just the swongs of online poker. For the more recreational player who checks his results after every session, is more apt to tilt, and generally allows his confidence to be affected by the short-term swings inherent in the game, then yes, I would agree that he is marginally more likely to book a few winning days in a row while his confidence is high.

However, in my case (and hopefully yours too!), variance plays a miniscule part in my poker life. Some days I run good, some days I run bad. Overall, I run exactly even. If I allow a completely alien factor such as 'how I'm running' to inform my decision-making, then I'm losing the poker battle. And making correct decisions is the ONLY consideration at the poker table. It is all that matters.

Comments (3) Trackbacks (0)
  1. Patterns _do_ exist. It’s a matter of discerning them correctly. Having a large enough sample size. Understanding the deviation over a specific sample size.

    I have a clear pattern of having a stupid day about once every two months. I have small wins for almost two months and a big loss after about that period of time. I’ve been tracking my daily online results for a long time and it’s a clear-as-a-bell pattern. I’ve come to understand that it’s likely because I’m playing so small that I’m frustrated with winning a little bit all the time and one day just decide to move up for the hell of it. To win more. At which time I lose more. Keeping detailed records and finding that pattern has helped me to understand when that’s happening and moderate those losses.

    It’s very similar to people who play loose when they’re up in a session and give it all back. That is a fairly common behavioural pattern in poker as well.

    • Hi Eric,

      Your point is absolutely right, although I think you are misinterpreting the concept of patterns that is being discussed here. The phenomenon you describe is indeed a pattern, but it has nothing to do with variance. You readily admit that your losses come from taking shots at bigger games, which suggests that the pattern in hand is that you win in lower-stakes games and lose at higher-stakes ones.

      This blog refers to those situations in which Hero arbitrarily decides that he is running good, running bad, etc. A lot of poker players’ focus gets drawn towards these pretty meaningless concepts, as most players will skew the evidence to give an unrealistic account of reality. This is why I included my results from a five, ten, and twenty day sample; if I look at the last twenty days earnings, then I have been motoring along nicely. If, however, I try to draw conclusions from that nasty five-day streak in the middle then it might appear as though I ran bad (indeed I did, though the streak was broken pretty quickly and variance evened itself out). Imagine, then, how unreasonable it would have been for me to have complained to my friends about how bad I’d been running due to that anomalous five-day sample, when it is pretty clear that when the sample gets widened the numbers settle down a bit and my true winrate starts to reveal itself. I had not suddenly forgotten how to win at poker, I just caught the nasty side of variance for a short spell and played my way through it. A lot (most?) poker players, when faced with these situations, get sucked into thinking that they had been doing something wrong, and erroneously adjust their game in a misguided attempt to outsmart what is a matter of variance.

      We look for patterns and yet poker is a game with a large random luck element. Control issues indicate a fundamental lack of understanding about variance. The cards do not care that you have lost your last five flips; they run out their random way regardless. I have blogged about this concept before ( and I’m sure I will do so again.

Leave a comment

You must be logged in to post a comment.

Trackbacks are disabled.