Open Side Menu Go to the Top
Register
pushbot/callbot for PLO pushbot/callbot for PLO

04-08-2014 , 12:48 AM
Im wondering if theres a pushbot/callbot for PLO like the one JITxpert made for NLHE but for PLO
pushbot/callbot for PLO Quote
04-08-2014 , 04:20 AM
I'm not sure, but just to add one thing: in PLO the structure forces you to play postflop (ie max raise is 3.5 open, and min bi is 40BB in most games), so I'm not sure it would be possible.
pushbot/callbot for PLO Quote
04-08-2014 , 05:54 AM
Even for no-limit O/8 hypers where a HU Nash pushbot would make sense, it's very hard to make it because there are too many pocket card combinations. I believe no PC can yet handle the volume of computations needed to create a Nash chart.
pushbot/callbot for PLO Quote
04-08-2014 , 07:03 AM
If the stacks get short enough you could probably write a decent program that buckets a ton of hands of similar equity together in order to make possible for a computer to handle.

On the other hand even LHE bots bucket situations together because that game already has too many hand combinations lol.. PLO seems... Unreasonable, for now.
pushbot/callbot for PLO Quote
04-08-2014 , 10:30 AM
propokertools does it pretty quickly...

I doubt it would be hard to get one working but getting one profitable is another story.
pushbot/callbot for PLO Quote
04-08-2014 , 10:38 AM
MSS for PLO.... if they sold tilt-prof elixir to withstand the variance I'd try it
100bb+ is high enough variance, and with 40bb my keyboard and mouse would be stuck trough my screen after 10 hands
pushbot/callbot for PLO Quote
04-08-2014 , 01:07 PM
umh 100bb+ has more variance than 40bb
pushbot/callbot for PLO Quote
04-08-2014 , 05:03 PM
Quote:
Originally Posted by Loctus
If the stacks get short enough you could probably write a decent program that buckets a ton of hands of similar equity together in order to make possible for a computer to handle.
Quote:
Originally Posted by Mt.FishNoob
propokertools does it pretty quickly...

I doubt it would be hard to get one working but getting one profitable is another story.
There's no such notion as absolute 'equity'. There's only equity vs a range. And it's not clear how ranges should be composed. They of course can't be just PPT's x% because there are domination nuances (most famously, good A*** and KK** flip poorer vs AA** than some rags). Say, in NLHE their composition is far from that of PPT's x%.
pushbot/callbot for PLO Quote
04-08-2014 , 09:29 PM
Quote:
Originally Posted by coon74
There's no such notion as absolute 'equity'. There's only equity vs a range. And it's not clear how ranges should be composed. They of course can't be just PPT's x% because there are domination nuances (most famously, good A*** and KK** flip poorer vs AA** than some rags). Say, in NLHE their composition is far from that of PPT's x%.
We are only ever using PPT like ranges in our heads, you can go into further layers of domination, from a preflop vpip you can change your postflop expectation because dominating equity occurs a higher % of the time. It is quite simple to rank a hands form out of 10, costing things like gaps and low rank of highest card and relate to to preflop ranges and score its 'dominated' factor just by ranking its form, I guess a computer can do this easily.... there are many combinations sure but its easy to wipe alot of them out due to inadequate form considerations, so you would get a bot having an easy time in raise/fold preflop... postflop then the bot could react depending on how much the avg flopequtiy has swayed from average preflop equity.... You can extend your hand form into hand form + known deck form (u,e flop) and again rank it using its score to determine offensive or defensive measure.

If the idea is to get perfection, then its a matter of how shallow the stacks are, 40bb should be easy to get reasonable close, 20bb plo 6max is pretty much solvable using ppt average equity, but they would haave to be balnced against historical stats, even then limited by a lack of timing tells and instinctual perceptions of tempo change/tilt... combiniatorics is the only problem at like 7 bb
pushbot/callbot for PLO Quote
04-09-2014 , 02:42 AM
Quote:
Originally Posted by cbt
umh 100bb+ has more variance than 40bb
Sort of depends how you look at it I think.

Someone playing $100plo with $100 might choose to ss 200plo with 80 for example, which would obviously have more variance.
pushbot/callbot for PLO Quote
04-09-2014 , 05:55 AM
with 40bb it's hard to protect vs draws for example (speaking from low stakes PLO).
since our FE reduces with every player in the pot, more hnads go to SD.
more SD increases variance.
pushbot/callbot for PLO Quote
04-09-2014 , 07:06 AM
in bb/100 terms, 100bb stacks have significantly more variance than 40bb stacks.
pushbot/callbot for PLO Quote
04-09-2014 , 11:43 AM
strange people say that, I used to MSS in NL10/nl20 but never tried it in PLO, hence where my conclusions are coming from.
pushbot/callbot for PLO Quote
04-09-2014 , 11:59 AM
Link to the NL pushbot referenced in the OP? Just interested from a theoretical perspective.
pushbot/callbot for PLO Quote
06-06-2014 , 10:39 PM
wow completely forgot i started this thread
ok i was super vague but this is mostly for plo mtts where stacks get super shallow and you can get a very high % of your stack in for a single raise preflop, or in restealing situations.
obviously this is when youre far from the money thus no icm considerations.
http://www.mediafire.com/download/l1...ot+Rev+4.2.xls
thats the pushbot im refering to
and this is the link of thread
http://forumserver.twoplustwo.com/23...nk-you-824728/
pushbot/callbot for PLO Quote
06-06-2014 , 10:41 PM
i guess the same could apply for plo cash games where you buy short and dont rebuy and/or 3bet/4bet pots when also buying short.
pushbot/callbot for PLO Quote
06-06-2014 , 10:42 PM
apologies but completely forgot about this thread lol
pushbot/callbot for PLO Quote

      
m