I guess this could be considered another stats analysis post...

The other day a Stars reg who came over to FT asked why Rush seemed to be so aggro lately.  I too have noticed that the full ring games have had a lot more 3, 4 and 5 betting than just a few months ago.  I have traditionally done very well in 3bet pots.  But I suspected that part of my struggles last quarter was that I was not making as much as before in 3 bet pots.  Although I've made a number of changes (which I think are "improvements") in my 3bet game, this has not translated well in terms of more money won.  I think the competition is just getting much more fundamentally solid in 3bet pots, as you would expect over time.

I decided to take a look at my DB to see if I could confirm my suspicions.  The above graph depicts my win rate in 3bet pots for each month of 2010.  I took my EV win rate when I 3bet and combined that (weighted) to my EV loss rate when I got 3bet.  I figured out what 3bet EV win rate was as an overall win rate (over all hands).  I didn't feel comfortable putting my actual win rate numbers on there, but that's not really material to the analysis.  The X axis is zero, fwiw.  Since there are some fish from time to time at the tables, you should consider it a leak if your data points are noticeably below a zero win rate.  As you can see the win rate from 3bet pots has a noticeable downward trend.  AND this is when I think my 3bet game has gotten BETTER.

Even tho I averaged over 100K hands a month, sample size is still going to be an issue.  It's very plausible that the Oct and Dec data points were just on the low side of variance (just as May was on the high side) - most of the other points have been hovering around the same area.  A second plausible explanation of the downward trend is that October is when I stopped playing 1/2 - so it may just be tougher competition.  But I played mostly 2/4 the first part of the year, so I'm not sure that's a good explanation.  A third plausible explanation is that there are fewer "weaker" players around, so there's less "free money" to be had (ie. fish gets it in pre with his AJ vs your AA).  Since I've been switching games and levels, I can't say if I've noticed a change in the number of weaker players.

Anyway, it would be interesting if anyone else has noticed this trend.  But poker is all about making adjustments.  So far in Jan 2011, my data point is higher than any of those on the graph, but I have been running pretty good over only 26K hands.  I made a few more adjustments, so it's plausible the minor tweaks are paying off.  Or maybe the MSFR guys are right and I *am* a luckbox.

Anyway, I want to illustrate the fact that you can use your HEM data for tons of different analyses.  You don't have to be confined to the pre-made tables they have for you.