[PokerAce]

I have a feeling this formula is pretty accurate, regardless of how loose/tight a player is.

Loose players who are passive will have a low aggression frequency. For example, I'm looking at a player right now. The stats on this player are 33/11/.3. The Aggr Freq for this player is 14%, which is low as you would expect. If this player bets or raises, odds are, he has a pretty good hand.

Loose players who are aggressive will have a higher Aggr Freq. A 28/17/2.5 player has an Aggr Freq of 56%. This player can bet/raise with any kind of a hand.

Tight players who are passive will have a low Aggr Freq, and tight players who are aggressive will have a high Aggr Freq.

I'll do some players and their Aggr Freq:

VPIP/PFR/PT AGR = Aggr Freq

13/7/3.75 = 50%
22/12/1.2 = 42%
21/18/2 = 56%
20/15/5 = 74%
14/14/1 = 33%
31/4/.9 = 34%
22/16/2.6 = 55%
54/16/1.3 = 42%
13/8/5 = 65%
43/4/.8 = 34%

[ QUOTE ]

Perhaps if you were able to remove or at least account for frequency of call actions this would likely distill this metric into a much better measure of aggression itself (without the interaction of looseness-tightness to worry about)

[/ QUOTE ]

The more a player calls, the lower their aggression factor is. So a player, no matter how loose/tight they are, if they don't bet or raise much, their aggr freq will be low. If a player with a low Aggr Freq raises, you know he's got a good hand.

If would be very difficult to determine the strength of the hand the player requires before raising.