Posted by Dr Fro 4:26 PM

I read Freakenomics over Christmas, which I thoroughly enjoyed. The authors attack a lot of "conventional wisdom" in their book. I am often skeptical of conventional wisdom, a tendency that earns me the label of "contrarian" from some friends and motivates other friends to tell me that I have my opinions "just for the sake of being different." Of course, I disagree. When you recognize just how

1. hesitant people are to stray from the pack and
2. lazy people are when it comes to substantiating / corroborating experts' claims and
3. the very people that spout opinions out the the public are the same ones that have a self interest in their claim being true,

you should quickly put these things together and realize that much of what is commonly agreed upon is quite likely to be false. The best contemporary example of this was the lead up to the war in Iraq. A high percentage of Americans in 2002 would argue until they were blue in the face that there were, in fact, weapons of mass destruction in Iraq. (I was one of those people). Not even Karl Rove would try to convince you of this now. We were duped because we do not question what is commonly assumed to be true. (this is not a partisan thing, people on both sides of the aisle suffered the same lack of skepticism)

So, anyway, I make it my policy to not believe much of anything I hear. That is not to say that I immediately believe the opposite to be true; I just don't immediately take a side. There are a lot of hot topics out there with experts arguing both sides, both claiming to have the facts on their side: global warming and "peak oil" both come to mind. Anyway, the book has reinvigorated my skepticism.

It also got me thinking on a few other matters. The first one is randomness. I have found that sports media do not understand randomness at all. You should read the entry on the Freakenomics blog on randomness. Sports media types are quick to make bold claims based on results that partially are a function of a team's abilities but also a function of randomness. When Ohio State won the national championship in 2002, they won a lot of their games by < align="justify">Same goes for poker. When we win, it is because we are fantastically skilled, but when we lose, we were "unlucky". If you don't have this hypocrisy embedded in to your poker self-assessment, you are in a very small minority. So, when I lost on Friday night, I attributed it to luck. After all, in over 5 hours of play, I won 2 pots, split 1 and quartered 1. That is some terrible luck. And I am sure that the next time I win a big pot with AA, I will convince myself it was my fantastic skill the deserves all the credit.

Speaking of some crazy luck (or skill), Huge managed to win almost $900 in the $0.50-$1.00 NL game over the same 5 hours. Winning 900 big blinds is f***ing nuts. Sure, the game has a lot of action and a lot of bluffing and a lot of Juan, all of which up the potential for big nights, but 900 BB's is just a lot.

I often wonder what length of time is sufficiently long to be considered "the long run." That is, over what stretch of time should most of the effect of randomness in results be sufficiently minimized such that the good poker player can expect to have profits. This can be calculated with inputs such as the desired confidence interval and the standard deviation. Problem is, what the hell is the standard deviation in poker? Quite a lot, I imagine. After all, I read all the time about professionals (on their blogs) going through terribly long slumps, often lasting several months. I play a lot less often, so I would probably need a couple years worth of play to glean any meaningful analysis of my results. I track my gambling results by the year, and for 2006, I lost about $200 gambling on the year. Interestingly, if we had stopped playing poker 12 hands earlier than we did on Friday night, I would have had a winning year. So, one of two conclusions can be drawn from this:

• the sample size is too small, or
• I really am a darn close to break-even player

The evidence is starting to suggest that the latter is true. Although I won a ridiculous amount gambling from birth through 2003, I have been pretty much break-even from 2004-2006. The data includes all gambling, but it is overwhelmingly related to online poker. All the usual explanations come to mind: the opponents have gotten better, we have all read (and we all apply) the same books, yadda yadda yadda. But I think I have identified the true cause...

I have a formal system for my online play that resembles Junell's system. One component of that system is that I never buy in for >20% of my stack. This has the result of moving up in stakes when doing well and reducing the stakes when doing poorly. If you believe that better players are at the higher tables (which is true enough) and that I can (in the long run) easily beat the lower limit games, it is easy to see how this system of mine almost guarantees that I stay near break-even. I eventually promote myself to table stakes that I am not good enough to beat. This knocks me down to games that I do beat, which then knocks me up again to the tables I can't. It is like an EV equilibrium.

I am beginning to question if this means that my system is fundamentally flawed. Perhaps I should put a cap on how high of stakes I will ever play at. I have evidence to suggest that if I set that cap at the $1-$2 game, I would do quite well. But such a cap is contrary to my nature. I am only happy playing at the low stakes games because it helps to build a bankroll to fund the high stakes games. Take away that carrot, and fighting over nickels and dimes suddenly loses its luster.

There is another solution, of course. I could improve. If I played better at the higher tables, I could keep my current system and make a ton of money. Ah, but how? Hell if I know, but I do think that in 2007, I will be more vigilant in implementing a fundamental belief of mine related to poker: Playing by the book is not the way to make money. This often shocks poker enthusiasts, who love "the book", but consider an analogy. Most people intuitively understand that in a large March Madness pool, if you pick all favorites to win, you have no chance of winning the pool unless all the favorites actually do win (which is highly unlikely). I won't get into all the math, but you are going to have to trust me here (if you need reason to believe me, consider that I have run a pool for 20 years that has grown to 400+ entrants and I have done an extrordinary amount of research on the data I have). One of the most interesting results of the research I have done is that consistently, the people that have a set of 63 picks with the least in common with the other participants' picks consistently outperform others. In fact, in all but one year, all of the money winners were in the top quartile on this statistic. Of course, picking all favorites means you have the most in common with other participants' picks. Anyway, if you want circumstantial evidence, I also use a dummy entry every year that picks only favorites, and it has never cracked the top 10% of the pool.

Oh geez, where was I? Ah yes, I think that playing by the book sets you up to play equally as well as the other players at the table (who can also read books). But game theory and my poker experience both say that there may well be a strategy that foils the the by-the-book strategy (assuming that all other players will continue to play by the book when they realize you have strayed from the pack). In game theory, most complicated decision-making scenarios have an optimal strategy, but every strategy (including the optimal strategy) has a strategy that dominates it. Thus, there is no transitive property in poker. The optimal strategy is only considered "optimal" because of an assumption that your opponents could be playing any strategy. But if you know that they are in fact playing optimal strategy, this assumption no longer holds.

Take an example: Mark Junell. I don't know if he wins in the long run, but I have seen him do quite well. He is just as likely to check raise you pre-flop with 8-3 offsuit as he is to do it with AA. He is very difficult to figure out, and I think that does well for him. People do little to adjust their strategy against him, and he is all the richer for it.

So, I will continue to play "by the book" at the lower limits (since it seems to work, presumably because some of the players actually can't read books) but trick it up at the higher tables. What do I have to lose? I was losing at $2-$4 and above anyway, so I can "afford" to experiment with different approaches.

* * * * * *

I had a blast on Friday, and I quickly pinpointed why. My old Houston poker friends are capable of doing something that no other home game can do: talk while keeping the action going. Every other home game I have played in comes to an abrupt halt if someone asks something as benign as "who won the ball game tonight?" I am naturally impatient and more so at poker, so slow action just bugs me. I love that old saying about whose turn it is: it is the guy who is asking "whose turn is it?".

With 12 hands to go, I had $306.50 in front of me, representing profit of $6.50. I then ended up all-in and quartered, losing $125. I lost another hand for about $100. On the last hand of the evening, I re-raised Juan to go all-in with 22. You may think that was stupid, but then I would reply that you must not know Juan Miranda. We ended up with 4 all-ins: 22, TT, AA and KJo. Juan had that AA, but the jack-king-off (hee hee) won the pot. As a result, JBaird won $400 and broke even on the night. Talk about randomness.

* * * * * *

If Minnesota and Missouri won their bowl games, I would be in first place in two different pools (they each have scoring systems that use a "confidence rating") As you know, they both lost in improbable and dramatic fashion on Friday. As a result, I am near last place in both of those pools. The college football gods seemed to continue their assault on my happiness during the first quarter of the UT-Iowa game. Then a penalty took a touchdown away from Iowa and led to a UT touchdown. I am certain we would have lost that game absent that penalty. So, in a sense, the game was decided because one guy on Iowa accidently lined up in the wrong place. Had he not done that, we would have lost and Mack Brown would be ridiculed for being an idiot and people would be concerned that Colt McCoy never fully recovered from his injury. But instead, all is peachy keen in Austin. Again, sportswriters attribute too much to teams and not enough to randomness of events.

* * * * * *

I rarely make New Year's resolutions. I find them to be a little silly. Why, if there is something wrong with my life, do I need to fix it on January 1? If I recognize that smoking is harmful on July 23, why not quit smoking then? But I do have a resolution this year. On December 31, I stepped on the scale and for the first time since 1998 was > 10 pounds (11 pounds to be exact) over my ideal weight. I could have started my diet then, but I waited until today. I have no idea what diet I am going to do. For now, I am just going to "eat less, eat better" and only trick it up if that does not work. I had a handle on my weight for 8 years, but fatherhood has led to more time at home, which means less exercising and more snacking.

* * * * * *

This is a pretty good read on cheating in poker. Some of it is a bit technical, but the gist is that that you are probably a lot safer than the conspiracy theorist believe, but far from entirely safe.

3 Comment(s):

Posted by  Anonymous, at 10:29 AM, January 02, 2007  

Sometimes the gameplay in Dallas gets slowed down when people "fold in their head" but just stare at the two cards in front of them.

__________________________________________________

Posted by  Dr Fro, at 3:16 PM, January 02, 2007  

Touche

__________________________________________________

Posted by  Junelli, at 9:55 AM, January 03, 2007  

Enjoyed your post.

__________________________________________________

Post a Comment