Would somebody run a sim for me?

I need somebody to run a sim for me.


I want to know what % of the time 22, 55, and 88 will win the pot when they flop a set. I know how often they will flop a set but they won't necessarily win the pot even when they do.


If anybody has any non-sim data which shows how often you will win the pot when you flop a small/medium set, I would love to see that data as well.


I'm trying to determine under what circumstances a player can profitably cold-call a pre-flop raise with a small/medium pair. A comment by Tommy Angelo in Medium Stakes that he would call 3.5 bets pre-flop (from the small blind) with a pair as small as 22 has resparked my interest in doing a mathematical analyis of when these calls can be profitably made. Of course, I will post any worthwhile analysis I make.

think in 2+2 books they say youll win after flopping set ~80% time.


thats general, i dont know how you would tweak for # players.


brad

download "poker" from http://www.geocities.com/TheTropics/Cove/6357/programs.html


that will show you exactly what you are looking for

not sure if this is the type of sim you were looking for, but here is what i did.


1. turbo texas hold'em version 5

2. "tough" lineup

3. 6-12 game (structured like "mirage" games)


10% max $3 rake, .75 cent tip

4. seat 1 gets pocket dueces with a duece on the flop. run auto test for 1 million hands.

5. results: win 9.8% with a net profit of $2.62 a hand.


hope this is of some help, remember that your results will probably be better as you play better than the computer player. though that is somewhat negated by the fact that some of your opponents also play better than your computer opponents.

good luck

I tried this simulator out, think it may be buggy. I got the following results for 88, 55, 22 each flopping a set against 7 random hands all staying through river.


88 won ~70.5%

55 won ~67.5%

22 won ~70.5%


Maybe I did it wrong, anyone want to try and verify these results?

Huh? You flop a set and only win 10% of the hands!?!?

The problem with your sim is it does not limit the wins to just sets. It could include unimproved pair, and straights and flushes.


Stu

yeah 10% does seem low, i'll double check my setup and re-run it.

thanks

the computer player only saw the flop 17.6% of the time ( he didn't know he would flop a set).

when he did play he won 42.4% of the time

i agree with your results, against six players


maybe it has something to do with "all straights contain a 5 or a Ten" - so if you take out three fives there can't be as many straights out against you?

I ran this through Poker Probe.


10 players. Player one holds a pocket pair and the only board card specified gives the player the set. All other board cards and all other players' cards are random.


Across 100,000 deals.


22: 62%

55: 60%

88: 63%


Of course, see Mason's essays for why this type of simulation has problems.


Regards,


Troy

Right, so that means that 55 should do better than 22, not only because it is higher and will beat lower sets, but also because it eliminates more possible straights than 22. This result can't be right.

The 5's doing worse that the 2's, and the 2's the same as the 8's looks wrong, but it may just be sampling error. You didn't say how many hands you ran. It could take millions of hands before your percentages converge on the theoretical numbers.

i tried, for 250,000 hands each against six random hands, to see how many won with all hands going the full way:


22 2xx 70.66%

33 3xx 69.85%

44 4xx 68.70%

55 5xx 67.56%

66 6xx 69.04%

77 7xx 69.96%

88 8xx 70.44%


i can't explain it, but there seems to be a reasonable curve appearing

I ran a sample size of 50,000 for each hand several times. It kept coming up with those results for each hand within 1 percent. I think it is a bug in the program.

great program!!

Just the opposite is true, as there is now a 5 on the board to make a straight.

Yeah, I guess you're right. I can see it now.


So, flopping a set with 22, 33, or 44 is better than flopping a set with 55. Son of a gun...


I guess 55 is still a slightly better hand because of its better chances of holding up unimproved and the slim chance of making a higher flush.

thank you


i wouldn't have got to where i am today without it


when i am playing at partypoker i have time to run a 50000 hand sim before the bet gets to me


and that is if i am utg

I can't understand the 2's doing that much better.

There are more straights possible when a five comes on the flop than with a 2. So the set of 2's will lose to fewer straights. Apparently, this more than makes up for losing to higher sets and full houses, so I believe it.


But, in real play this may not matter much because a lot of those hands that make a straight when a five is on the board would already be in the muck (47, etc).


Funny though, I just won a huge pot last night by flopping the weakest set possible, 555. /images/smile.gif

But wouldn't that mean that a set of 555 wins MORE often then a set of 222?

apparently not - see jimbo's: http://www.twoplustwo.com/cgi-bin/newforums/holdem.pl?read=80118