[Pokey]

I see people ask this question all the time. The threads get quickly locked (the mod squad is always on duty) and what few responses they get basically say "blah blah blah don't worry about the rules."

That's mostly correct, but I wanted to explain a bit about WHY the rules are unimportant. More significantly, I wanted to give you all a chance to create your own rules that would be useful for the games that YOU play. So, here we go:

<font color="blue">Why are autorate rules useless?</font>

Mostly because people are not binary. It's extraordinarily rare to find someone who is nothing more than a "rock" or a "maniac" or a "calling station" or what have you. Even the mundane, simple-minded, vanilla, generic opponents that uNL'ers face all the time actually have their own quirks. Maybe that rock has a 72o fetish, and mixes up his game by always playing 72o just like AA. (Of course, you'd never notice this because he never gets to showdown with it, either folding himself or folding his opponents.) Maybe that "maniac" who runs 70/55 has, as one of his rare preflop smooth-call hands, AA -- he knows how much folding equity he gets and he wants max value from his monsters. Maybe that calling station is afraid of aces on the flop, maybe that LAG always gives up on the turn, maybe that TAG three-barrels whenever he raises preflop, etc. The point is that a player cannot be adequately summarized by a cute little picture of a fishy or a bomb.

Another equally important point is that we cannot numerically define any of these categories. Quick, what stats represent a TAG? The people who play $2NL should give me a different answer than the people who play $25NL. The people who play Friday nights only should give me a different answer than the people who play weekdays. It's even possible that the people who play Bodog should give me a different answer than the people who play Stars. The thing is, labels like "TAG" and "maniac" are situationally specific. A good translation of "TAG" would be "noticeably tighter than average and noticeably more aggressive than average," where "average" is defined by the typical conditions at the table. This all leads us to the second half of this post.

<font color="blue">How can I create autorate rules that will be useful for me?</font>

The first step is to decide how sensitive you want your rules to be. Is "tight" the top 5% tightest opponents? The top 10%? The top 25%? Since the "line in the sand" is entirely arbitrary, any threshhold will do. Once you've defined your relevant fraction, scour through your own PT database, or open up 100 tables (over the course of a few days). Look over the sorts of opponents that YOU play and find out (roughly) what numbers correspond to those threshholds. Let's say that you define "loose" as the 10% of players you face with the highest VPIPs. Let's say that in looking over your database you find that 52% VPIP is about the cutoff that marks the top 10% loosest of opponents. That becomes your own personal autorate rule for looseness. Do the same for preflop raises and postflop aggression and you've got yourself the beginnings of a set of autorate rules. It's personal, it's specific, it's well-defined, and it's going to actually tell you something that will be useful to YOU.

Examples of autorate rules that might be handy:

LAG: Loose and aggressive preflop, aggressive postflop. Top xx% highest VPIP, top xx% highest PFR, top xx% highest average aggression.

TAG: Tight and aggressive preflop, aggressive postflop. Bottom xx% lowest VPIP, top xx% highest PFR, top xx% highest average aggression.

Rock: Tight and passive preflop, passive postflop. Bottom xx% lowest VPIP, bottom xx% lowest PFR, bottom xx% lowest average aggression.

Station: Loose and passive preflop, passive postflop. Top xx% highest VPIP, bottom xx% lowest PFR, blottom xx% lowest average aggression.

Ways to make this even more helpful for you:

Define other categories of specific interest to you. Perhaps you're always looking for people who change drastically from preflop to postflop; you might want to make a category for LAPs who are loose and aggressive preflop but become passive postflop. You could throw in a low c-bet percentage or simply go with high VPIP, high PFR, low average aggression. If you worry about blind thieves (high steal %) or frequent c-bettors (high c-bet) or double-barrelers (high c-bet turn) or showdown monkeys (high WSD) or weak-tighties (low WSD) or tricky-trappy opponents (high check-raise numbers) or what have you, feel free to make a category that specifically addresses that issue.

Another VERY handy way to get extra information would be to pick the category or categories that you need to pay the most attention to and create TWO rankings within it. Say you're always on the lookout for calling stations. You could define an Ordinary Calling Station as the top 35% in VPIP, bottom 35% in PFR, and bottom 35% in average aggression. Then you add in another category called a Super Calling Station, defined as top 10% in VPIP, bottom 10% in PFR, and bottom 10% in average aggression. (Make sure you order them correctly so that PT recognizes both categories.) At that point you might choose to play at tables that have at least three Ordinary Calling Stations or at least one Super Calling Station.

There are no "right" autorate rules; there are only rules that will benefit you. Personally, I don't use autorate rules at all -- I simply have a bunch of data reported in my HUD and I glance it over looking for opponents that look exploitable. If you have a hard time processing lots of data, or if you want to trim it down to a more manageable amount of information, feel free to create autorate rules that will tell you at a glance information that YOU consider important. Mark opponents you need to recognize easily, either to chase or to avoid. Ultimately, you're the only one who can say what type of opponent you need to pay attention to, so ultimately you'll have to make your own rules for them to tell you the information you need. So quit asking what good rules are and go make some for yourself!