Hello everybody!
My name is Andrei and I am a Romanian programmer converted to fulltime poker player. Been playing on Stars for the past 3-4 years as Zamoxes mostly HU and mostly 50s last year - however the good results there are due to the fact that I fish select. Moved 2 months ago to fulltilt and started learning superturbos. Mersenneary is to blame for that because of his excellent videos - probably wouldn't have made the switch without those - and for that Mers you have my eternal gratitude !
At the moment I am using my own custom HUD that I programmed specifically for HU - not that HoldemManager/PT3 aren't good but I want more control on the information I have displayed, in order to take advantage of opponent's tendencies. I do include things like action preflop past hands for opponent (so I can see if he is in aggro or passive-ish mood), SB open/limp %, BB call/3bet/raise limp %, postflop aggro, raise %, ck raise %, fold %, also open sizes for the past 10-15 hands (so I can keep track if opp has a 2x or 3x range). Not perfect but works for me so far and I want to add some more information to it in the future (like known opponent hands that he limped/raised 2x/3x / openshoved with).
True to my habits I am playing fishes at 20s superturbos - because of the switch to fulltilt I need to increase my bankroll - but in the future I want to start playing at higher limits and I imagine that the fish population is thinner up there (well ok mostly everybody is probably looking fishy to Mers but that's beside the point now!) so I want to improve my play vs better players. Hence my enrollment in this program.
Lots of information here, need some time to go through the threads and absorb it - it might be a bit before I post any meaningful question/hand in this thread.
Cheers,
Andrei
Good to have you here! And good call on reading through all the information, everybody should be doing that.
Hello again!
I will update this thread with a conversation I had via e-mail with Mers, we agreed it's better to continue this on the forum. All started with a little simulation I made:
-------------
Hi Mers,
I did a little experiment and attached results, hope you will find this useful!
What's it all about: suppose we have 20/15/12 BBs effective stacks and
hero is BB. Villain min opens. Should the hero call,shove or fold ? I
tried to give an answer to that given opponent open range - each file
shows results for villain open 20%/40%/60%/80% for the stack sizes
20/15/12 BBs in this order.
Hero shove and fold EVs are relatively easy to compute (probably I
should look in the future on villain call ranges if hero shoves, for
now I used 22+,A2+,K8+,QJ,Q8s-QTs,J8-JTs,T9s for 15/20 BBs and
22+,A2+,K7+,QT+,JT,Q8s-Q9s,J8s-J9s,T9s,98s for 12 BBs which might be a
little too wide for 20 and too tight for 12 - but for now I think they
might be ok for our purposes!)
Hero fold ev is again easy it's -1 always
How can I compute the ev when hero just calls ? In this case I did an
approximation, for a known hero hand I dealt 5000 random hands for
villain in his opening range and on flop I assumed hero checks,
villain bets 1/2 pot 100% of times and hero check raises if he has 3rd
pair or better or a decent draw or some combo like overcards and
flush/open ended - and villain will call the ck shove only if he has a
decent hand like top pair or better - or 2nd pair + a decent draw - or
a stronger combo.
Obviously this is not what really goes on in reality since villain
might not cbet 100% of the boards or bet half pot - some villains
will bet less, some will bet pot, some will just shove. But at least I
got a process that simulates what goes on in many cases and I can
assign an EV value for hero's call of the 2x open.
So for what is worth - results are attached. For each hand the two
numbers represent 'ev for shoving / ev for calling' and both should be
compared with ev of folding which is -1. Of course for AA and such if
the ev for shove is higher than the ev for calling doesn't mean we
have to shove, we can just 3bet
------
Images are here:
range 20: http://imgur.com/m8YUL
range 40: http://imgur.com/4cU7R
range 60: http://imgur.com/f8nJt
range 80: http://imgur.com/y33K5
------
Very interesting - thanks again for sharing.
The problem of estimating expectation when calling OOP is the most difficult part of 3bet decision math. I think it's really cool to see those results after running that kind of simulation, but I still worry about it's accuracy in terms of actual games. The best approximation there is is just historical expectation once you have a big enough sample. I'm happy to share some of my numbers with you if you'd like to see them.
mers
-----
Hi Mers,
I agree about OOP expectation, my method is way too brute to pretend
any kind of high accuracy. Would be nice to see your historical
expectation for hands like QJs, KTo, JTo for let's say 20BBs if opp
min opens and you just call ... I wonder how far off are they from my
brute values ... ( I don't suppose there is a way for you to filter
only those opponents that a) open < 40% b) open > 70% etc ?)
-----
This is 15-25bb deep calling a raise OOP with some common hands:
http://i.imgur.com/nt0b1.png
PT3 counts in big bets so -0.12 in overall expectation is -0.24 big blinds.
----
Thank you for the results! So in practice for offsuit hands
K9-KT,Q8-QJ,J9+,T8+ work well and for suited - Q7s+,J7s+,T7s+,97s+.
How about 76s and 86s ? How low could you go with Kx suited ? K7s ?
Also any chance you could post same results for 12-15 BBs - that is if
you do just call some of that stuff at that depth
----
I think you can and should call pretty wide especially 20-25bb deep, it's harder to get a good sample on the suited hands especially, but this is my flatting range of a minraise first hand of a super readless: You can call a little tighter depending on your postflop skills, and a lot of the hands in the bottom of this range make for great 3bet bluffing hands once you have some reads your opponent is opening a wide range and has a fold button.
Hi Mers,
Thank you for reviewing the hand histories I sent, spot on comments. Still a few hands where a few more clarifications would be nice:
1) http://www.pokerhand.org/?6121441
"This is the third hand, he jammed over your first button, I think it's too big of an overreaction to fold such a strong hand here."
should I just min raise / fold it ? Or limp and stab flop … but that seem too weak for me at this point... I have position here but that's about all I can see positive about this hand, if we catch a pair on flop it's going to be tricky to continue vs an unknown opponent that showed aggression… that is if he lets us see a flop
2) http://www.pokerhand.org/?6121446
"Probably a little too small on the river."
Wanted to get a call from 6/8/K with weaker kicker or a stray T, on the other hand do you think that such a small bet would induce a bluff raise ever from this kind of player (seems nittish so far, 14% open/limp, 3bet 20% OOP and folded rest) ? Definitely I see getting called by Kx but if I bet bigger do I lose 6/8 calls ?
3) http://www.pokerhand.org/?6121450
"This is a call, especially given that he'd shown some previous jam instincts at deeper stacks."
not sure I follow the logic here, eff stack is around 10.5 so we need around 45% odds in order to have EV better than EV fold. Can we be sure at this point he is not just shoving strong-ish ? He has shown TAGish tendencies and around 10BBs he might just shove strong as I've seen such players do. If I assume he will min raise or limp AA-JJ in order to trap or induce a shove from me and he will openshove 22-TT,Ax and broadways I get something like 43% vs this range.
4) http://www.pokerhand.org/?6121453
"This is also a call."
… same question as before, can you see him shoving a lot of hands we have > 45% odds vs them ? Because I can't, he opened like 14% of hands so far and limped < 30%.
Yep, I'd just minraise/fold the 96o. I think you're overreacting to one shove - 96o is a decent enough, top 2/3 type hand, and should be opened and position taken advantage of. There's a big difference between 96o and 92o, as well. I think this fold is too scared.It's not the worst hand in the world to limp, either, and I think folding is the worst of the three options.
" Wanted to get a call from 6/8/K with weaker kicker or a stray T, on the other hand do you think that such a small bet would induce a bluff raise ever from this kind of player (seems nittish so far, 14% open/limp, 3bet 20% OOP and folded rest) ? Definitely I see getting called by Kx but if I bet bigger do I lose 6/8 calls ?"It's definitely a calculation, but I think closer to half pot is best.
I think you estimate your opponent is shoving too tight of a range 10bb deep. They'll also be shoving connected things and suited things that QJ/Q9s crushes. There's a reason why those hands are NASH calls so deep - they have decent equity when behind, and crush a lot of hands. I don't expect the random opponent to be jamming as light as NASH, but definitely enough that these are calls.btw, your links don't work :) I just remembered which hands you were talking about.
Thank you Mers, those do explain all my questions so far. Very clear and enlightening comments, I do have now a list of leaks to work on. Btw I wouldn't have thought to ask about some of those hands should it had not been hand history review, so I am very glad I chose that path!As for pokerhand.org I think their domain expired in the last 24 hrs, good thing you had those hands in mind ... if anybody is interested I will post them here
Salut ma, ce faci?Nam stiut ca sunt alti romani pe aici :)Ai skype? Eu sunt mrbambocha pe skype si joc $15 ST.Salut!