Wednesday, October 6, 2010

A Cold Hard Truth

I decided to take a look at my HEM stats for the first time in a while. Not the stats that define how I play, but the stats that define my results. What I found wasn't terribly surprising in retrospect, but really has made me think about what I need to do in order to improve and keep playing this game for a living. The following three graphs pretty much sum it up, which are for all the hands in my database at 2/4 and 3/6 (unilaterally the numbers are better at 2/4, because the games are softer and the rake structure is actually better I think....you tend to get more than 50% more rakeback money for playing 3/6, which means they are raking more than 50% more on average, which means it's a worse structure). First, all hands that I played 7-ways or more:











Without a horrendous downswing early on (which was at 3/6 and was the reason I switched to trying to play 6-max games several months ago...it looks way worse when this thing is plotted in dollars, not big bets), this graph would be stupifyingly good. With it, it's still a very respectable win rate of more than 1 bet per 100, with rakeback (the blue line) adding in some extra win rate and padding out the break even stretches. It's only 17K hands, but it's pretty good looking so far. At 2/4 only it looks even better. Next, we'll have a look at my play 5-6 handed:



Meh....this is not great, obviously, but it is still indicative of actually winning. Actually it started off pretty well, but the 400 bets I've lost over the last 35K hands or so is extremely discouraging. Once again, it looks much, much better for just the 2/4 hands, with me actually losing (before rakeback) at all 3/6 hands dealt 5-6 ways to date. One might be tempted to say this sample size is adequate, but really it's not. It's enough to know that I'm not randomly clicking buttons, but it's not really enough to say with much confidence anything but "you probably are a small winner". Ah yes then, here we go my friends, the the cold hard truth. Here is my graph for hands dealt 4 ways or fewer:



Great googely moogely. Some of that is hopefully run bad; it's almost impossible for me to believe that I could be that inadequate short handed. Losing 5 big bets per 100 in the long run would be legendary. But even though I only have 7K hands here, there can really be no debate that I am a loser when the hand is dealt 4 ways or fewer. So what should I do? I've known that I'm not a great short handed player for some time now, but I thought in the back of my mind that I was getting better, that I was improving, and that it really wasn't that big of a short coming. Well these graphs pretty much nix that idea. I guess it's basically back to the drawing board.....

5 comments:

Captain R said...

Check out DD/MikeL's "Fives a Crowd" series. Pretty good. Also, watching somme HU videos will get you in the proper mindset for super-shorthanded play.

The blindman said...

Is the pattern similar 2, 3 and 4 handed? Because I find the play to be pretty substantially different..

jesse8888 said...

Yeah I'm going through the classic HU/SH videos now also. Mano a Mule, Thug Life, Pistolas de Justicia.

It occurs to me that StDev for 4 handed play is truly massive (something like 20 bets per 100), so my sample size really is pitiful and hopefully things aren't as bad as they look. Hopefully.

that_pope said...

What types of games are you sitting in when you play 4 handed? If there isn't a known poor player in it, just get up. Could game selection be part of the problem? Or improperly adjusting to the fish who aren't adjusting to the short handed nature and still playing the same that they always do.

I should do a similar anaylsis for myself...I have been a big loser at HU tables, but I think I have been a big winner at HU at 6max tables. The main reason for that is I don't have a chance to go on losing tilt at a 6max table because they usually start to fill after 10-20 hands, where I have had some epic 200+ hand sessions where I just get crushed cause I am on tilt at HU tables.

bellatrix78 said...

Wow, great analysis (and very truthful).
I suspect my graphs look exactly opposite ;-) Although, I do find 3-handed games are the most difficult for me. The whole extra SB is kinda hard to grasp. Plus HU, I know the levelling wars, 3-handed you first need to get your read if people adjust, 4-handed it doesn't matter that much anymore.

If you're taking online as a good way to learn (like aren't actually worried that much about results), you should practice a lot, just game creating, game creating, game creating.

I also find that people at Stars game select much more than at FTP, are you sure you are choosing the best seat possible always? Having a good game is great, but having the best seat in a meh game is greater.

I remember that one session we played against the huge maniac fish a few months back. The game was great, but you had the worst seat (to my left :P )... j/k, I don't remember if you adjusted, but that guy was capping any Ace and it's those small sample size sessions that suddenly turn the whole thing around.

Also, the StDev of 20 for 4-handed seems ok, but in today's game, I would contend that it's at least 25 for 3-handed and more like 21-22 for 4-handed (so you're mostly playing against nits haha).

If anything, the hilarious chat you get from hotel99, when he puts you on his "do not play" list is enough to game create a lot.