PDA

View Full Version : QTo favorite over 2's


11-07-2005, 10:08 AM
watching wpt with mike and VVP up here in canada yesterday. probably wasn't on in USA and it probably was old.

they had QTo as favorite over 22. i thought pairs were always favorites (even over suited connectors, while this is one-gap offsuit).... the one thing to notice though is that the pair doesn't cut into straight-making ability.

is this correct?? i can not seem to get the same odds from different odds engines...

i know i've seen this before in a book or online (for free). what's the best source for going thru every situation heads-up??? is it harrington??

PaultheS
11-07-2005, 10:54 AM
Pairs are always somewhere around 50/50 with overcards, but they aren't always slight favourites. Many suited and/or connected overcards are actually slightly stronger headsup. For example, from twodimes.net:

cards win %win lose %lose tie %tie EV
9d 8d 894863 52.26 782248 45.68 35193 2.06 0.533
2d 2h 782248 45.68 894863 52.26 35193 2.06 0.467

And, like you said, QTo vs 22:

cards win %win lose %lose tie %tie EV
Td Qh 863394 50.42 828520 48.39 20390 1.19 0.510
2d 2h 828520 48.39 863394 50.42 20390 1.19 0.490

Note that if the QT doesn't cover 22's suits then it's 0.501 vs 0.499.

As for examining heads up situations, I would recommend twodimes (http://twodimes.net/poker/).

11-07-2005, 11:46 AM
thanks paul!!!

i can't seem to get the different evaluation engines to agree... i guess i just think it's interesting as to whether you are technically a favorite or not... but it's arguable that 51/49 and 50/50 are not too different.

wonder why these engines don't agree.

and one engine said QJo was worse than QTo against 2's. i don't see that. obviously the fact that the straight is constrained at the higher end, but i'd rather take the open-ended nature of it.

bobman0330
11-07-2005, 08:26 PM
[ QUOTE ]
thanks paul!!!

i can't seem to get the different evaluation engines to agree... i guess i just think it's interesting as to whether you are technically a favorite or not... but it's arguable that 51/49 and 50/50 are not too different.

wonder why these engines don't agree.

and one engine said QJo was worse than QTo against 2's. i don't see that. obviously the fact that the straight is constrained at the higher end, but i'd rather take the open-ended nature of it.

[/ QUOTE ]

QJ can make a straight with AKTxx, KT9xx, and T98xx. QT can make one with AKJxx, KJ9xx, J98xx, 9876x

11-07-2005, 11:33 PM
Pocket 2's against Q10o is a slight favorite to win. It is around a 55% favorite

11-08-2005, 12:20 AM
[ QUOTE ]
Pocket 2's against Q10o is a slight favorite to win. It is around a 55% favorite

[/ QUOTE ]
no.

uuDevil
11-08-2005, 03:19 AM
[ QUOTE ]

wonder why these engines don't agree.


[/ QUOTE ]

If you are doing a Monte Carlo simulation, the accuracy of the result depends on how many deals are simulated. If too few deals are simulated, the error may be larger than the difference between the equity of the hands.

11-08-2005, 05:32 AM
Hi. If I had to be in a preflop race to the river with one of these hands, I would take the pair of twos with NO hesitation! On coinflip type races I would always prefer the one currently in the lead preflop. I agree that the odds favor the QTo very slightly. But that QTo (no matter how imposing it looks) HAS to improve to win...and the pair can do the job on its own. (This assumes that the board does not produce 4 of a kind.) Have a nice day!

mackthefork
11-08-2005, 10:24 AM
[ QUOTE ]
watching wpt with mike and VVP up here in canada yesterday. probably wasn't on in USA and it probably was old.

they had QTo as favorite over 22. i thought pairs were always favorites (even over suited connectors, while this is one-gap offsuit).... the one thing to notice though is that the pair doesn't cut into straight-making ability.

is this correct?? i can not seem to get the same odds from different odds engines...

i know i've seen this before in a book or online (for free). what's the best source for going thru every situation heads-up??? is it harrington??

[/ QUOTE ]

22's is a small dog to QTo 49.5/50.5 against QTs it is 47/53, JTs is a 54/46 favourite against 22. If a 2 is known to be dead then QTo becomes a 56/44 favourite.

Regards Mack

fnord_too
11-08-2005, 10:46 AM
[ QUOTE ]
Hi. If I had to be in a preflop race to the river with one of these hands, I would take the pair of twos with NO hesitation! On coinflip type races I would always prefer the one currently in the lead preflop. I agree that the odds favor the QTo very slightly. But that QTo (no matter how imposing it looks) HAS to improve to win...and the pair can do the job on its own. (This assumes that the board does not produce 4 of a kind.) Have a nice day!

[/ QUOTE ]

This is a huge error in thinking. Do you also prefer the 22 against 9 /images/graemlins/diamond.gifT /images/graemlins/diamond.gif with a board of 7 /images/graemlins/diamond.gif8 /images/graemlins/diamond.gifK /images/graemlins/spade.gif because the 9T still has to catch?

yellowjack
11-08-2005, 12:03 PM
[ QUOTE ]
Hi. If I had to be in a preflop race to the river with one of these hands, I would take the pair of twos with NO hesitation! On coinflip type races I would always prefer the one currently in the lead preflop. I agree that the odds favor the QTo very slightly. But that QTo (no matter how imposing it looks) HAS to improve to win...and the pair can do the job on its own. (This assumes that the board does not produce 4 of a kind.) Have a nice day!

[/ QUOTE ]

This is an incorrect way of thing. You need to re-evaluate it, as it is 100% certain to be wrong. The other forumer's example with the T/images/graemlins/diamond.gif 9/images/graemlins/diamond.gif illustrates why this is the case. I personally cannot think of another example.

The board could produce 4 of a kind to render the 22 useless, but a much more common situation is the board double pairing (e.g. 55996) where the overcards' kicker comes into play.

11-08-2005, 12:23 PM
[ QUOTE ]
This is a huge error in thinking. Do you also prefer the 22 against 9 T with a board of 7 8 K because the 9T still has to catch?

[/ QUOTE ]

Hi. No. I was referring to a preflop situation. Have a great day! /images/graemlins/smile.gif

LetYouDown
11-08-2005, 12:57 PM
[ QUOTE ]
but it's arguable that 51/49 and 50/50 are not too different.

[/ QUOTE ]
Most casinos would tend to disagree with you /images/graemlins/grin.gif.

Komodo
11-08-2005, 02:26 PM
I think the WPT computer count the folded hands in the calculations.

11-09-2005, 11:18 AM
It doesn't matter preflop or not. He is saying your chances to win with 89s on that flop are greater than 22 even if you still have to improve. Which is the EXACT same thing as Q10 vs 22 PF

fnord_too
11-09-2005, 12:37 PM
[ QUOTE ]
[ QUOTE ]
This is a huge error in thinking. Do you also prefer the 22 against 9 T with a board of 7 8 K because the 9T still has to catch?

[/ QUOTE ]

Hi. No. I was referring to a preflop situation. Have a great day! /images/graemlins/smile.gif

[/ QUOTE ]

Ok, so do you like 22 over 9Ts preflop? The point is the favorite is the favorite, if you would rather be a dog, however slight, just because you have a made hand, your reasoning is flawed. It is a common error in reasoning that a lot of players have. Anywhere, pre flop or post flop, you want the hand with the highest expectation. In all in situations, that is always the hand favored to win, regardless of whether it is made or not. If you are going against a completely random hand, would you rather have 22 or TJs? What about 22 or TJo against a random hand?

Race situations are rarely exactly even (the only time they are is when both hands are isomorphic). There are a lot of times when it is correct to call an all in with an unmade hand where it isn't correct to call with a small pocket pair. These small percentage plays really add up. Consider STT's, where push or fold and call an all in or fold come up all the time. Over 100 STT's, you will probably see 3000 (guess, but probably pretty close) of these decisions. Of these, maybe 500 (another educated guess) will be relatively close. How you handle these 500 decisions will have a huge impact on your ROI. STTer's live and die by how well they make these decisions. (yes, as you move up the field makes these decisions better, so you need to garner advantage in other stages of the game, but mistakes here cost more than all but the most eggrigious mistakes in the other stages.)

11-09-2005, 05:53 PM
[ QUOTE ]
I think the WPT computer count the folded hands in the calculations.

[/ QUOTE ]

thanks, that's a good explanation!!! not something we really have access to, although i suppose number of people folding could give you a slight idea.

d10
11-09-2005, 09:57 PM
[ QUOTE ]
I think the WPT computer count the folded hands in the calculations.

[/ QUOTE ]

This is the correct answer. If the other 2 deuces were already seen folded, and a bunch of cards not queens or tens were folded, QT could have a significant advantage over 22. Typically you see this going the other way, where a pocket pair becomes something like a 60/40 favorite because some of the overcard outs have been folded already.

11-10-2005, 04:30 AM
i see this all the time in the wsop espn coverage, where a hand like 77 can be anywhere from 57-60% over AKs preflop, depending on aces/kings/suits in the folds.

fnord_too
11-10-2005, 12:27 PM
[ QUOTE ]
thanks paul!!!

i can't seem to get the different evaluation engines to agree... i guess i just think it's interesting as to whether you are technically a favorite or not... but it's arguable that 51/49 and 50/50 are not too different.

wonder why these engines don't agree.

and one engine said QJo was worse than QTo against 2's. i don't see that. obviously the fact that the straight is constrained at the higher end, but i'd rather take the open-ended nature of it.

[/ QUOTE ]

pokerstove (www.pokerstove.com) can do the calculations iteratively. QTo is a slight favorite over 22, QJo a slight favorite but a hair less of a favorite. (Iteritively means it goes over all possible combinations, so the answer is exact. Many engines use a monte carlo method to get their results, so they will have a margin of error.)

51/49 is very different than 50/50. It is a bigger edge than the house has for the pass and don't pass lines in craps. In some situations you would make the same decision regardless of whether you are 49%, 50%, or 51% to win a hand, but slight edges matter a lot, and the cummulative effect of always being on the good or bad side of slight edges is huge, but variance really obscures the value of edges in the short run.

11-10-2005, 02:17 PM
51/49 is very different than 50/50. It is a bigger edge than the house has for the pass and don't pass lines in craps. In some situations you would make the same decision regardless of whether you are 49%, 50%, or 51% to win a hand, but slight edges matter a lot, and the cummulative effect of always being on the good or bad side of slight edges is huge, but variance really obscures the value of edges in the short run.

[/ QUOTE ]

thanks for the pokerstove.com... i realize that casinos thrive on 51/49. i was thinking more of showdown where you don't even know what the true % is. i think we tend to think around 50/50. i don't tend to sweat 50/50 vs. 51/49 (or 52/48) when i don't even know my opponents exact hand. worst thing about casino games is that you know what the odd are.

wouldn't 50/50 always be o.k. for a showdown from an EV point of view as there will always be folded $$$ in the pot and your opponent is putting in $$$$$.

fnord_too
11-11-2005, 03:54 PM
[ QUOTE ]
51/49 is very different than 50/50. It is a bigger edge than the house has for the pass and don't pass lines in craps. In some situations you would make the same decision regardless of whether you are 49%, 50%, or 51% to win a hand, but slight edges matter a lot, and the cummulative effect of always being on the good or bad side of slight edges is huge, but variance really obscures the value of edges in the short run.

[/ QUOTE ]

thanks for the pokerstove.com... i realize that casinos thrive on 51/49. i was thinking more of showdown where you don't even know what the true % is. i think we tend to think around 50/50. i don't tend to sweat 50/50 vs. 51/49 (or 52/48) when i don't even know my opponents exact hand. worst thing about casino games is that you know what the odd are.

wouldn't 50/50 always be o.k. for a showdown from an EV point of view as there will always be folded $$$ in the pot and your opponent is putting in $$$$$.

[/ QUOTE ]

There are actually spots where you should fold a favorite. Here is an extreme case: Last 3 of a tourney, you have 49.9K, villain has 50K, 3'd guy has .1K. 3'd guy folds, villain pushes and shows you AK, you have 66 (say blinds are 500/1K), you should fold your favorite here (assuming normal tourney payout structure), because you lose cashing EV by calling.

Another interesting thing you can do is to determine the face up value of a hand, that is, if you flipped your hand face up and everyone played optimally if you pushed, there is an effective stack size where you should push vice fold.

In practice, you are right, your estimation error is large compared to the equity percentages most of the time, but not always. You can also do things like figure in folding equity. e.g. you can determine how often you need to win if called to make a push +EV (from a chip perspective, it gets complicated when you are looking at a cashing perspective in tournaments). These exercizes are very beneficial and go a long way to quantifying the value of aggression.

I think I have rambled way off topic here, I'm not quite sure where I was going with all of this.

jonnyd
11-11-2005, 04:19 PM
not just 4 of a kind but say board comes 8844x

you are screwed

11-11-2005, 05:24 PM
[ QUOTE ]
I think I have rambled way off topic here, I'm not quite sure where I was going with all of this.

[/ QUOTE ]

yeah, i know the feeling!!!! thanks for your help!!!