Math around Set over set situations.
Join Date: Sep 2006
Posts: 447
Hello,
I've been reviewing my personal set flopping frequency and the frequency with with I am in set over set situations. It caught my attention when I realized I had been in 5 set vs over set situations (where i had the under) and lost all 5, and 0 zero set vs under set situations. Now this has only been the last month or so, roughly 50K hands or so.
I went further back, to the start of the year, ~140K hands, and I have discovered I have been in:
5 set vs under set situations (won 4)
18 set vs over set situations (lost all 18).
My question is can anyone explain to me the math of these situations and if my sample size is significantly off? I would have expected the two situations to be closer to even...then again, sample size becomes a question as well...i may try to go further back...
Ok, i've gone further back, to August 07, so in total this is ~275K hands.
22 Set vs under set (won 21)
36 Set vs over set (won 2)
Though it doesn't seem they are converging, it must simply be an unlucky run as of late. Oh well.
Last edited by uber pea; 04-11-2008 at 07:21 PM.
Reason: went back further in database
Join Date: May 2007
Posts: 3,486
wrong forum again. try probabilities. thats why people go when they need to start a thread called "OMG i got aces 3 times in a row and all three times it got cracked what are the odds?!?!?!??!"
but despite the dumb threads people still give actual calculations.
Join Date: Jan 2008
Posts: 31
Join Date: Jan 2005
Posts: 2,224
Maybe you've been in tons of set over set situations where your opponent owned you and folded.
Join Date: Jun 2007
Posts: 1,558
Maybe you play smaller pairs a lot more often that other players
Join Date: May 2005
Posts: 4,240
In 275,000 hands you will get 16,176 pocket pairs. If you play them all, you will have 1,862 sets. You don't say anything about the size of the tables you're playing, but a reasonable guess is something like 104 of those times, someone else at the table will also have been dealt a pocket pair that catches a set on the flop (I use 104 instead of 100, because on average you will have 8 sets of each rank). For simplicity, assume no one folds after getting a set.
On 52 of the hands, you had the higher set, and the other player folded 30 of them preflop. On another 52, you had the lower set, and you folded 16 of them. That suggests you're only folding 2's and 3's (or maybe playing some of those but sometimes folding other low pairs) preflop; while the table average is playing only top pairs (if they play J's and up, T's about half the time, the math works about right).
This is going to cost you money in set-over-set situations, plus you're going to lose a lot of preflop bets when your low pair doesn't hit on the flop. The question is whether you make it up when you get a low set and no one has a better hand. My guess is no.
Join Date: May 2006
Posts: 2,405
A simple explanation:
When you have a big pair you nearly always raise and reraise, thus you don't see that many flops with them.
With small to medium pairs you are more likely to limp after limpers and call raises. You see the flop more often and therefore you flop more small sets.
Join Date: Dec 2006
Posts: 2,336
The total times we both flop a set and ours is higher than our opponents compensate for the other times it's the other way around. So we only have to bother about the difference in ranks of the pocket pairs that we play.
That means that the times we flop a set with a low pair (that our opponent doesn't play) AND he hits a set as well are pure losses that will have to be made up by the times he doesn't flop a better hand.
Suppose our opponent plays 10% of his hands against a raise. AA-66 will be 41% of his range. He will have a better set about 5.5% of the time. In this scenario my guess is no too. Since I don't think we'll win a large (comparable) pot more than 5.5% of the time when he doesn't have a set.
If he plays 25% of his hands though, the chance of him having a better set is less than 1%. My guess is you can overcome that. So for example coldcalling with pocketpairs after a raiser or limper with 25% VPIP is a lot more profitable since his range is much wider.
Also the numbers above assume that he will play the top range of his pocket pairs the same as the others. And we will probably not when you have 44 and the flop comes 24A you can be reasonably sure he doesn't have top set. Same if the flop is KQ4 and you we're flatcalled preflop. There are a lot of flops you can assume your bottom or middle set is likely to be good. It's dry flops with medium ranked over cards, like 379r that are most likely to be trouble imo.
