Two Common PokerTracker Misunderstandings

[Catt]

It should be obvious to most here that having good reads on your opponents is very, very helpful in determining your course of action in the hands you play. It may be less obvious that some of the most potentially profitable reads are those on specific post-flop plays of your opponents -- but alas, bowing to reality, I understand that oftentimes we don't have the reads and notes we would like, for one reason or another.

Usually we rely on PT data to help inform our choices. Relying solely on PT isn't great, but it is usually better than not using any "read-like" info at all. I say "usually" better because such reliance is helpful if the data is properly understood, but can be harmful if misunderstood. And so, I thought I'd point out two common misunderstandings I see in the Micro Limits forum pretty frequently (and see in SS often enough and Mid-High occasionally).

Post-Flop Agression Factor. The implications of the PT post-flop agression factor is frequently misunderstood. The problem stems from the fact that the AgFactor cannot be read in a vacuum -- it is highly influenced by VP$IP and # of flops seen. In our ML FAQ we restate the consensus view that an AgFactor of 2.0 or higher is cconsidered "aggressive;" 1.0 - 2.0 is "neutral;" and less than 1.0 is "passive." These ranges are by and large considered standard. However, such ranges are based upon an assumed VP$IP of 15 - 20 or so. When we encounter a VP$IP of 65% the same scale does not apply. Why? The AgFactor is calculated as (bet % + raise %) / call %. A 1.5 AgFactor is considered neutral around here, but that's for someone with a reasonable VPIP -- our tendency to bet and raise more than call (and thus get a 1.5 AgFactor) is heavily influenced by the fact that (1) we generally only see post-flop play with strong pre-flop hands, and (2) we generally fold when faced with a poor flop and action from others. To maintin a 1.5 AF (again, more betting and raising than calling) when our VP$IP is 65%, we would need to either: (1) be a very disciplined folder when the flop misses us, so that even though we're seeing ~65% of the flops we are getting away from them immediately when we miss; or (2) bet and raise a whole crapload of terrible and marginal hands. Most 60% VP$IP players are not disciplined folders post-flop, so you should generally assume that a 60/7/1.5 player is betting and raising a whole crapload of terrible or marginal holdings until proven otherwise.

Pre-Flop Raise %. A less common misunderstanding is the relation between VP$IP and PFR%. The fact is that there is almost no connection (the only connection is that VP$IP will always be equal to or greater than PFR%). Simply put, PFR% is the percentage of hands dealt that a given player raises. If his PFR% is 10, he is raising 10% of the hands dealt to him, regardless whether his VP$IP is 80% or 20%.

I'm sure there are additional misunderstandings that are repeated out in the forums, but these are the two I see most often.

[/soapbox]

[jaxUp]

Good post. I was aware of the first problem, but I didn't know that people actually thought PFR was relative to VPIP. Anyways, interesting stuff. Problem 1 was prticularly well explained.

[mosuavea]

Good post.

[Greg J]

Good thoughts. Thanks for sharing them, esp regarding ag factor. I had thought on this same point but never put it in words.

Thanks for a quality post Catt.

[Stuey]

[ QUOTE ]
Pre-Flop Raise %. A less common misunderstanding is the relation between VP$IP and PFR%. The fact is that there is almost no connection (the only connection is that VP$IP will always be equal to or greater than PFR%). Simply put, PFR% is the percentage of hands dealt that a given player raises. If his PFR% is 10, he is raising 10% of the hands dealt to him, regardless whether his VP$IP is 80% or 20%.


[/ QUOTE ]

I think I misuse this stat. I always thought if a player had a high VP$IP% and a high PFR% when he limps there would be a good chance he has very poor cards. If a player has a low VP$IP% and a low PFR% when he limps I feel he has a average hand or good hand. Is this a mistake?

[KaiShin]

Great post. I learned a lot from it.

[Catt]

[ QUOTE ]
I think I misuse this stat. I always thought if a player had a high VP$IP% and a high PFR% when he limps there would be a good chance he has very poor cards. If a player has a low VP$IP% and a low PFR% when he limps I feel he has a average hand or good hand. Is this a mistake?

[/ QUOTE ]

No, I think you're roughly on the right track -- but really there's no need to look at the numbers in tandem. A high VPIP and a high PFR might generally mean that a limp is a weak hand -- the reasoning is that if, say his PFR is 25, then any limp is either deception or a hand that falls in the bottom 3/4's of all possible hands. But, a VPIP of 20 and a PFR of 18 means a limp is likely to be on the weaker side of what a tightish player would play, regardless of the tighter player's PFR. Conversely, a low VPIP would generally mean that a limp is a pretty decent hand, regardless of PFR -- for example, when a 10/6/2 player limps, you should still expect that his limping hand is pretty strong -- he's only playing 10% of all hands dealt. A limp from such a player UTG (if he is position aware) indicates a pretty strong hand. Just remember that VPIP represents the subset of hands played from among all possible hands; and PFR represents a subset of hands raised from among all possible hands played. A 20/15 player and a 20/3 player are probably playing similar hand sets -- the former is just much more aggressive pre-flop.

[Aaron W.]

I enjoy essay posts more than I enjoy most hand posts. This is especially true when the content is solid like in this one. Good work.

[Aaron W.]

*Bump* in response to...

... Villain is loose and extremely passive, 60/0/1.25 over about 100 hands

... Villain is loose (53), marginally aggressive post flop (1.5)

[mojobluesman]

Very useful. Thanks.