Using VP$IP to Put Opponents on a Range of Hands
Frequently, players use Heads-Up Displays (HUDs) to display PokerTracker stats on their opponents in real-time. Some HUD programs available include PlayerView, GameTime Plus, and PokerAceHUD. However, many players utilize displaying stats like VP$IP and PFR in addition to auto-rate data to get a "read" on opponents.
Many players, however, don't have the slightest idea of how to interpret this data into anything useful. This article hopes to change this.
My auto-rate rules for Limit Holdem utilize a few different classes of players, including "tight," "semi-loose," "loose," and "extra-loose." But how does this specifically help you assign a player to a range of hands? The classifications alone don't really do much to help, since these classes utilize ranges of VP$IP's to place players into these categories. A tight player, for example, is any player with a VP$IP < 21%. Semi-Loose players are those with VP$IP's between 21-34.99%. This is likely the least useful classification, as so many different players fall into the category. We will examine these ranges of VP$IPs in terms of Sklansky's hand groupings to more accurately estimate the quest for knowledge contained in the second level of poker thought--"What does my opponent have?"
There are 1,326 unique combinations of 2 cards possible to make a 2 card Holdem hand. Actually, there are 2,652, but I am considering Ah-Ac to be the same hand as Ac-Ah, as we all should, as the order the cards are dealt has no bearing on the hand. To continue with the methodology, there are 6 ways to make a pocket pair, 4 ways to make a suited hand, and 12 ways to make an offsuit hand. Now, on to Sklansky's groups:
Group 1 Hands: AA, KK, QQ, JJ, and AKs. There are 28 ways to make these 5 hands, 2.11% of the possible hands.
Group 2 Hands: TT, AQs, AJs, KQs, and AK. There are 30 ways to make these 5 hands, 2.26% of the possible hands.
Group 3 Hands: 99, JTs, QJs, KJs, ATs, and AQ. There are 34 ways to make these 6 hands, 2.56% of the possible hands.
Group 4 Hands: T9s, KQ, 88, QTs, 98s, J9s, AJ, and KTs. There are 50 ways to make these 8 hands, 3.77% of the possible hands.
Group 5 Hands: 77, 87s, Q9s, T8s, KJ, QJ, JT, 76s, 97s, Axs, and 65s. There are 98 ways to make these 18 hands, 7.39% of the possible hands.
Group 6 Hands: 66, AT, 55, 86s, KT, QT, 54s, K9s, J8s, and 75s. There are 68 ways to make these 10 hands, 5.13% of the possible hands.
Group 7 Hands: 44, J9, 64s, T9, 53s, 33, 98, 43s, 22, Kxs, T7s, and Q8s. There are 106 ways to make these 20 hands, 7.99% of the possible hands.
Group 8 Hands: 87, A9, Q9, 76, 42s, 32s, 96s, 85s, J8, J7s, 65, 54, 74s, K9, and T8. There are 132 ways to make these 15 hands, 9.95% of possible hands.
Taking these percentages not alone, but cumulatively, yields a different perspective on the whole VP$IP argument. Remembering that in a full ring game, between 2-5% must be added to the derived percentage to obtain a true idea of VP$IP (to compensate for looser calls from the blinds), and in a 6-max game, between 6-10% needs to be added. We'll take the mid-range for the sake of argument and use +3.5% for full ring and +8% for shorthanded tables.
As you notice, this is by far not exact, in that ranges overlap, and that players do not have the solid cutoffs as displayed in the Sklansky groupings. This should give you a good idea, however, of hands you can expect a player to be proceeding with based on their VP$IP.
The statistic that I've felt has been overlooked by many PokerTracker and HUD users is actually not PFR, but a statistic that is not immediately visible, unless it is calculated. For example, a player with a VP$IP of 16%, and a PFR of 8% has a raising frequency of 50%. To me, the overall PFR number tells you a great deal about how aggressive a player is pre-flop, and about how to proceed with a weaker holding with this player to your left. The raising frequency statistic (which can be calculated quickly while looking at a player's stats) tells more about what hands this player will actually execute a raise with.
The difference: A player to your right has a VP$IP of 27%, and a PFR of 1%. You are dealt 55 in MP. Should you call here? Yes. You are unlikely to be raised by this player, and if you are, you know that this player likely holds a premium hand. This tells you that this hand should likely be even more of a "set it or forget it" proposition. The VP$IP tells you that this player is playing hands down to Group 6, but only raising with his monster hands (the top 1%), Unless you hit a 5 on the flop, you need to be folding. You're likely beaten, especially if an Ace or a King hits the flop.
Now, on the other hand, this player has a VP$IP of 27%, and a PFR of 18%, Same hand, 55 in MP. Calling here and putting the action on him is telling you that since this player raises 2/3 of the hands he plays, you are very possibly going to have to call another bet if you call the first one. However, it is also a signal that you are not necessarily beaten without a set. He may be raising with QJs, or even less.
Raising frequency just puts another idea into your head as far as playing a hand properly goes. Much of this information becomes less useful as players become trickier, but this information, and understanding how to use it, can go a long way into providing "reads" on players whose faces you can never see, and into getting the most from your PokerTracker license.
Many players, however, don't have the slightest idea of how to interpret this data into anything useful. This article hopes to change this.
My auto-rate rules for Limit Holdem utilize a few different classes of players, including "tight," "semi-loose," "loose," and "extra-loose." But how does this specifically help you assign a player to a range of hands? The classifications alone don't really do much to help, since these classes utilize ranges of VP$IP's to place players into these categories. A tight player, for example, is any player with a VP$IP < 21%. Semi-Loose players are those with VP$IP's between 21-34.99%. This is likely the least useful classification, as so many different players fall into the category. We will examine these ranges of VP$IPs in terms of Sklansky's hand groupings to more accurately estimate the quest for knowledge contained in the second level of poker thought--"What does my opponent have?"
There are 1,326 unique combinations of 2 cards possible to make a 2 card Holdem hand. Actually, there are 2,652, but I am considering Ah-Ac to be the same hand as Ac-Ah, as we all should, as the order the cards are dealt has no bearing on the hand. To continue with the methodology, there are 6 ways to make a pocket pair, 4 ways to make a suited hand, and 12 ways to make an offsuit hand. Now, on to Sklansky's groups:
Group 1 Hands: AA, KK, QQ, JJ, and AKs. There are 28 ways to make these 5 hands, 2.11% of the possible hands.
Group 2 Hands: TT, AQs, AJs, KQs, and AK. There are 30 ways to make these 5 hands, 2.26% of the possible hands.
Group 3 Hands: 99, JTs, QJs, KJs, ATs, and AQ. There are 34 ways to make these 6 hands, 2.56% of the possible hands.
Group 4 Hands: T9s, KQ, 88, QTs, 98s, J9s, AJ, and KTs. There are 50 ways to make these 8 hands, 3.77% of the possible hands.
Group 5 Hands: 77, 87s, Q9s, T8s, KJ, QJ, JT, 76s, 97s, Axs, and 65s. There are 98 ways to make these 18 hands, 7.39% of the possible hands.
Group 6 Hands: 66, AT, 55, 86s, KT, QT, 54s, K9s, J8s, and 75s. There are 68 ways to make these 10 hands, 5.13% of the possible hands.
Group 7 Hands: 44, J9, 64s, T9, 53s, 33, 98, 43s, 22, Kxs, T7s, and Q8s. There are 106 ways to make these 20 hands, 7.99% of the possible hands.
Group 8 Hands: 87, A9, Q9, 76, 42s, 32s, 96s, 85s, J8, J7s, 65, 54, 74s, K9, and T8. There are 132 ways to make these 15 hands, 9.95% of possible hands.
Taking these percentages not alone, but cumulatively, yields a different perspective on the whole VP$IP argument. Remembering that in a full ring game, between 2-5% must be added to the derived percentage to obtain a true idea of VP$IP (to compensate for looser calls from the blinds), and in a 6-max game, between 6-10% needs to be added. We'll take the mid-range for the sake of argument and use +3.5% for full ring and +8% for shorthanded tables.
| A Player Plays Hands Up to | Full Ring VP$IP | 6-max VP$IP | Classification |
| Group 1 | 4.11-7.11% | 8.11-12.11% | Very Tight |
| Group 2 | 6.37-9.37% | 10.37%-14.37% | Very Tight |
| Group 3 | 8.93-11.93% | 12.93-16.93% | Tight |
| Group 4 | 12.7-15.7% | 16.7-20.7% | Tight |
| Group 5 | 20.09-23.09% | 24.09-28.09% | Tight/Semi-Loose |
| Group 6 | 25.22-28.22% | 29.22-33.22% | Semi-Loose |
| Group 7 | 33.21-36.21% | 37.21-41.21% | Semi-Loose/Loose |
| Group 8 | 43.16-46.16% | 47.16-51.16% | Loose |
| Any Two Suited or Worse | >47% | >55% | Extra Loose |
As you notice, this is by far not exact, in that ranges overlap, and that players do not have the solid cutoffs as displayed in the Sklansky groupings. This should give you a good idea, however, of hands you can expect a player to be proceeding with based on their VP$IP.
Utilizing Pre-Flop Raise Statistics
The statistic that I've felt has been overlooked by many PokerTracker and HUD users is actually not PFR, but a statistic that is not immediately visible, unless it is calculated. For example, a player with a VP$IP of 16%, and a PFR of 8% has a raising frequency of 50%. To me, the overall PFR number tells you a great deal about how aggressive a player is pre-flop, and about how to proceed with a weaker holding with this player to your left. The raising frequency statistic (which can be calculated quickly while looking at a player's stats) tells more about what hands this player will actually execute a raise with.
The difference: A player to your right has a VP$IP of 27%, and a PFR of 1%. You are dealt 55 in MP. Should you call here? Yes. You are unlikely to be raised by this player, and if you are, you know that this player likely holds a premium hand. This tells you that this hand should likely be even more of a "set it or forget it" proposition. The VP$IP tells you that this player is playing hands down to Group 6, but only raising with his monster hands (the top 1%), Unless you hit a 5 on the flop, you need to be folding. You're likely beaten, especially if an Ace or a King hits the flop.
Now, on the other hand, this player has a VP$IP of 27%, and a PFR of 18%, Same hand, 55 in MP. Calling here and putting the action on him is telling you that since this player raises 2/3 of the hands he plays, you are very possibly going to have to call another bet if you call the first one. However, it is also a signal that you are not necessarily beaten without a set. He may be raising with QJs, or even less.
Raising frequency just puts another idea into your head as far as playing a hand properly goes. Much of this information becomes less useful as players become trickier, but this information, and understanding how to use it, can go a long way into providing "reads" on players whose faces you can never see, and into getting the most from your PokerTracker license.
6 Comments:
Great freakin' post.
Came here through a link from donkeypuncher and i must agree. this is some prime meat right here. thanks for offering.
Mighty good stuff here.
Really good post. I like your blog a lot and I constantly read and attempt to apply some of the concepts you speak of.
Keep up the good work.
Excellent value-add, PokerShark. Well done.
I likes. Just reiterating others comments, but good stuff.
Post a Comment
<< Home