Wednesday, June 25, 2008

Confirmation Bias #2

A great example of the psychology of (mis)understanding what you see at the poker tables came from Yorkshire Pud (see comment in the post below). Here not only was the play not understood by opponents and the result 'confirmed' to them that it was bad... there was another aspect - the agreement of peers to further re-enforce an opinion.

Kind of beats the example I was going to give to start part#2 - but hey!

This came from a WSOP freeroll thingy on Stars last weekend.... 4000 odd players and 3 spots paid with a seat. My bustout hand was actually an error - I check-raised all in from the SB with 9=7 sooted on a 6-8-3 board - stacks were perfect for this move in the mid-game... only problem was I assumed I had fold equity against exactly the hand that called me (j-8) and was insta-called. No good blaming the call in my world, I should have seen it coming and made my decision accordingly.

Oops - that was an aside! Meant to continue with my confirmation bias / blame luck theme, this time for the very inexperienced freerollers.

What was happening again and again in the freeroll was that players were betting like this:

- Suited cards, highish cards, low pairs = Limp
- Decent but not great hands 10's, J's, A-Js = Raise a lot (4+ BBs)
- Premiums = Min-raise (or even worse min-reraise)

This worked because there were very few observant players around (and those that were observant tried to get people with j-8 to fold too often!). The thing is it was actually the majority of the table playing this way... if they mini-raised I folded without a 'big pot' hand (pair / connector etc to hit a flop hard).

Now, lets look at the perception of the players making these hands they feel invincible pre-flop, get scared of losing their audience. The small raise gets money into the pot without making anyone fold for sure - but as experienced players will realise, playing a raised multi-way pot out of position is not ideal for QQ / JJ / AQ etc (or even the better hands than these).

So, the bets would be small on the flop too, giving any player who cared to take it a good price to draw. Inevitably the QQ lost (well, a decent proportion of the time) and was 'shown' by the player in the end... along with a comment along the lines of 'nice catch' or 'joker-stars' etc

Now - another long background piece. Here then is the thinking point: We tend to look for evidence in the outside world that confirms our beliefs or behaviours, and downplay evidence that contradicts or threatens these beliefs.

And the rub: This holds back the poker player, in a number of situations, from being able to objectively improve their play... since many bad plays, both subtle and (like the one above) screamingly obvious are re-enforced by winning those occasional big pots - or by seeing that we had the best hand and so were 'correct but unlucky' (once again).

Well, a lot of words here - will continue this theme with some thoughts about how exactly we can break through our in-built bias and really start to work on our games!

Gl at the tables, Mark


Anonymous said...

I like this series Mark, I completely agree with you that we need to look at things differentyl to improve.

Your example with QQ is the perfect example as 95% of players will not even bother to try to find error in a hand where there QQ gets beat by 87s. As you said they type in "unlucky" "
joker/riverstars" whcich bugs the hell out of me.

I mean they act like they have no control over what just happened yet they min raise or flat call a big PP into a 4 or 5 way pot. and then wonder why they get drawn out.

Interested i nhearing your thoughts on the recent trend of small pp calling on the bubble in the low/medium Sng's and how this affects the proper play. As they could have a small pp a reasonable % of thime and unless you have a bigger one your pushing into someone who is ahead of you. Just an observation of what seems to be the defense to the "proper" bubble play as of late is to call with these holdings that are likely a few % ahead.

Mark said...

Hey Scott,

Sure - calling with a small pair at the bubble is often an error too... all depends on stack sizes / ranges etc.

Guess my point is that when a player who habitually 'calls light' sees that you were pushing 'junk' then the 'bad call' is re-enforced as the 'right play'.

The obvious examples I listed in my posts are to illustrate the point more than anything... where I will try to get to is how we can avoid falling into more subtle versions of this trap once more experienced.

After all - one can not correct a leak if you actually perceive it as a good play!

Cheers, Mark