Yeah I'm not saying someone will always make a badugi, I'm saying that IF it is known that they do have one (i.e. they are pat, you are assuming they are not snowing, and they are standing on ANY badugi), it is possible to narrow down their range based on the amount that they drew, and therefore have an idea of what % of the time your hand is good should you improve to a badugi yourself. In fact here are some more detailed results, for the top n% badugis in increments of 10%:

top 90%, 80%, ... down to 10%
xxxx drawing 4: KJ9A, K98A, K53A, QT7A, Q752, J976, J52A, T652, 942A
Axxx drawing 3: KT7A, K73A, QT8A, Q74A, J97A, J53A, T72A, 965A, 843A
A2xx drawing 2: K72A, QJ2A, Q52A, J82A, T92A, T42A, 942A, 842A, 652A

But like with everything, it's garbage in -> garbage out. If you think they will not pat a queen badugi on the first round then of course you can't assume that should you make a J it is good >50% of the time.

I see. Good point. The flaw is not so much that opponents in my games are snowing as that they are not necessarily standing on any badugi. You'd have to know individually what opponents would stand on.

As I wrote in my last post, I got badly burned the last time I stood pat on a ten badugi. And a ten badugi is really the same if you make it on your first draw or your second (or third); it's still a ten badugi. It still loses to a nine badugi.

I think, in retrospect, not breaking a ten badugi is probably not a very good play for me in a pot-limit or no-limit game. I can't tell if someone betting has a better badugi or not. If I have taken a draw, someone could be betting a three card badugi into me. If I stand pat, any bet into me is probably a badugi, but maybe at least a seven. (Depends, of course). And what do I do with my ten badugi? Do I fold it if someone is bold enough to bet into me? (rhetorical).

The long and short of it is I don't think I want a ten badugi. I want to either get a better badugi or simply miss and have an obvious folding hand. I don't want a hand where I'll be faced with a difficult, mistake prone, decision.

What about a nine badugi? I don't know. I have to think about that some more. (I'm obviously not an expert at badugi).

And my opponents are regularly breaking badugis and drawing for better ones.
Thanks for the data.
My current thinking is it's poor play to pat a ten badugi, (and I don't know about a nine).

Buzz

pretty much spot on from my experience... mostly about a couple months of pot-limit freerolls @ oddsmaker.
nine badugis are in that gray area, like pat eights in triple draw-the smoother
the better; altho you have a 9-card range vs. a 7-card range, villain has about as many outs w/ a 1-card draw in either case.

Is pot limit actually a fairly common / normal way of playing? I'd imagine that medicore badugis have a big reverse implied odds problem in big bet games, and strong 1 and 2 card draws go up in value. I only play limit, and in the game I play regularly (micro stakes MTT with many clueless opponents), a pat ten is basically the nuts.

WHERE CAN I FIND THESE GAMES?

www.carbonpoker.com

Mr.Advantage-

The P/L Badugi tourney i mentioned is a freeroll, @ 5:15PM est daily on oddsmaker.com
hope that helps
$pike

P.S. RubbishCards, the reason i stay w/ 9's or lower is that, this being a F/R, i have EVEN MORE clueless opponents

New version available here, now with support for multiple draws. I am reasonably confident that it should spit out the right numbers.



And before anyone asks

Badugi sim, 1000000 trials, 3 draws:
Dead cards: 3s

Hand 0: Ah 2c 3d --: 2nd draw: keep K badugi. 3rd draw: keep K badugi
Hand 1: Th Jc Qd Ks: standing pat

Hand 0: Ah 2c 3d -- : EV 50.5% - WIN/LOSE/TIE %: total 50.5/49.5/0.0 - badugi 50.5/0.0/0.0 - 3 cards 0.0/49.5/0.0 - 2 or 1 card 0.0/0.0/0.0
Hand 1: Th Jc Qd Ks : EV 49.5% - WIN/LOSE/TIE %: total 49.5/50.5/0.0 - badugi 49.5/50.5/0.0 - 3 cards 0.0/0.0/0.0 - 2 or 1 card 0.0/0.0/0.0

RubbishCards

I'm getting a Vista "side-by-side configuration is incorrect" error.

Any suggestions?


PS - I f'cking hate Vista.

Hmmm... I am running Vista myself and haven't seen this problem, but googling for it, it seems lots of people are reporting this error in all sorts of programs.

Things I would suggest:
- if you are running the .exe directly from the .zip, try extracting it first
- do a windows update. Some of the cases I googled were to do with an out of date system .dll
- you could try running it as administrator (right click and run as administrator, instead of double clicking). It would be pretty bad security practice to do this on some executable uploaded by a random guy on an internet forum, and would be ridiculous if this were necessary, but it's an option.

Is there any other info from the error message, like a button you can click to get more details?

If anyone else has the problem please let me know, thanks.

*EDIT* okay you could try downloading this crap from Microsoft and I can't believe I have to make people do that for this thing to be usable

hi rubbish. i want to thank you so much for making a utility like this. only having the equity calc of a single card draw isnt nearly as useful though as being able to calc 3 draws. it would be amazing if you could code certain customizable settings like;

-if i make a badugi then opponent keeps 3 card <9 and folds >T, villain folds street x y% of time if doesnt improve to 3 cards.
-if villain makes a badugi off 2 cards i keep 3 card <T and fold >J
-if villain makes a badugi off 1 card i keep 3 card <9 and fold >T
-if villain draws 1 i draw 2 to 2 cards <6

stuff like that. i could come up with more settings that would be extremely pertinent to a strategic player if youd like.

thanks again!

Did you get the new version (from about 3 posts up)? It now works for multiple draws.

Anything that introduces betting/folding action will be a bit beyond the scope of what I can realistically make it do. It might be possible to come up with something for changing drawing standards based on opponents' drawing actions, specifically: giving the option to break when an opponent pats after a draw. But this introduces the concept of position, which complicates things a lot at my end. Still, it's something I can look at in future.

Right now I am adding support for 7 card stud hi with hand ranges and it's progressing quite well, so that should be up within the next few days. After that I'll be doing the same for stud/8.

Hi,

This is nice and I appreciate the work you have put into it.

I am finding that it often, but not always, ignores the Draw to: cards
I enter. If I run your example it looks OK but if I try and change the
Draw to cards it ignores it, for example set 2nd to Q for the 8JQ hand.

I'm using XP.

Tx.

That had me stumped for a minute In fact this makes perfect sense. Changing the K to a Q means this player will not keep a K badugi going to the 2nd draw, but it is actually impossible for this hand to make a K because it needs the Ks, which is in the 3rd player's hand.

Is there another specific case where you think it might not be working properly?

*edit* ok, I can see in the output text that it is still treating it as drawing to a K instead of Q, I'll see what's going on here

OK there's a stupid bug which I've now fixed. For 3 card hands, it was always reading the draw-to value as K. Oops I mostly work on a stripped down version that doesn't have any fancy Windows front end so I didn't notice this as it is working fine under the hood. I should have a new version posted within the next day.

NEW VERSION HERE - now with stud hi support (and fixed the above bug).

Cheers, looks good.

any chance triple draw ?

I've been considering this now that I've actually played a few hands of TD (I'm learning these games in the wrong order IMO). Does anything like this already exist for TD? I know twodimes does a single draw for 2-7, but is there anything that does multiple draws? If not, I'll probably have a go at it.

Not as far as I know for TD, and it would be very nice.

Cheers.

Some hypothetical TD n00b questions:

So let's say I get the simulator working like badugi, where you tell it what to stand pat with, and it sticks to that without any regard for the opponents' actions. And for example let's say you tell it to stand on a pat ten or better, and it is dealt an 8 high straight on the first round.

Presumably it would be normal to have it discard the 7, so you have 8 outs to an 8-6 rather than 4 outs to a 7-6 if you discarded the 8? What about a 9 high straight? Now you could lose the 9 and have 8 outs to an 8-7 or lose the 8 and have 12 outs to a 9-7. So I guess you would want a way to make the user specify this in addition to saying it will stand pat on a ten or whatever?

Are there cases other than possible straights, where it makes sense to discard something other than your highest card(s) when you aren't paired (just based on your own cards / ignoring opponent soul-reads for now)?

interesting I haven't thought this out but you could have an toggle option where you choose "3 card 7 draw OR 4 card 8 draw" for cases like 87654. just let the user decide which to prefer? there aren't that many cases so you could just list them and have user decide which to prefer.

also I would say have some options for the key card 2 and the special trouble card 6. like for 2 have an option to make a 2 in hand mandatory. like for 6 have an option to discard six if drawing x. like if you draw more than 2 then never keep a 6.

Version 0.4 available, now with stud/8

I would strongly recommend that you adopt the model that ProPokerTools uses for razz in two respects.

1. Allow users to not specify the suits. For example if the user imputs (234), this means any incomplete of the ranks 234. The parentheses mean unsuited.

2. Allow users to input a range of incompletes using the "-" notation. For example (-5-4-3) would mean any 3 card badugi of ranks 543 or lower.

MM- yeah I'll definitely consider that stuff if/when I get to doing range support for badugi. Range-vs-range is pretty tricky to get right and I'm not quite sure how to do it yet.

BTW for anyone actually using this thing, there is a bug I am aware of that is affecting some 2 card hands. When the hand contains a double suited pair, e.g. 4s 5s 5h 6h - it will keep the lowest card of each suit between draws (4s 5h in this case), or use those as the cards to play at showdown if there are no more draws. But this will give incorrect results when the pair is the lowest rank present, e.g. As Ah 4s 5h will play the two aces instead of Ah 4s. I'll have this fixed in the next version.