This is a very common spot that comes up a lot early on hypers. (I play the $3.50's on PokerStars)
First or one of the firsts hands on match (playing readless)
Stacks: Villain t500, Hero t500, blinds t10/20 Villain minraises to t40 on button
Hero holds Ax, what to do?
My line here is that I 3bet jam A2-A9 and 3 bet to 90 with AT+, is this a good deafult strategy, what would be the best play here?
Readless strategy should look to exploit the typical population tendencies after considering:
- Range population opens their button @ 25BB?
- Range population is calling a 3b jam @ 25BB?
- Range population will flat a 3b NAI @ 25BB?
- Range population will 4b jam over a 3b NAI @ 25BB?
- Do you gain enough expectation by 3b NAI to justify not simply 3b jamming with your value?
- Are you ever 3b bluffing (does population fold too much vs NAI 3b)?
- What's your plan for the rest of the hand if you end up in a 3b pot OOP (properties of particular hands vs call 3b NAI range)?
Answer these questions and you can design a strategy which maximially exploits the poplation.
In the meantime, rag Ax (A7-A2) need pretty wide opening ranges to be +EVs jams @ 25BB and are often better flatted until you have reads down to about 20 BB (+ you can exploit villain tendency to barrel on A high boards post flop).
Good Ax hands typically are better jammed, because you get more value from the inelastic (never folding) part of villain's range than the marginal expectation of letting villain come along with some of his implied value holdings which he's only putting more money in when he flops equity anyway.
Coleman open shoves off the blocks with A2o if that helps.
Not sure if it was readless though as villain called with QTo in a thousand dollar HT.
@ high stakes games it is fair to assume population is opening very wide from button making 3b jamming Ax2y pretty standard I would imagine.
@ micros there are tonnes of fish who only raise the top of their opening range, limping middling hands making 3b jamming rag Ax less profitable since:
- less fold equity
- lower pot equity when called (dominated or up against PP with only 3 outs more often)
FYI below is the expectation of 3b jamming Ax2y @ 25BB for various villain min raise [R] and call [C] ranges:
[C]
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
[R]
10%
-7.0
-7.0
-7.0
-7.0
-7.0
-7.0
-7.0
-7.0
-7.0
-7.0
20%
-2.0
-3.9
-3.9
-3.9
-3.9
-3.9
-3.9
-3.9
-3.9
-3.9
30%
-0.3
-1.6
-2.5
-2.5
-2.5
-2.5
-2.5
-2.5
-2.5
-2.5
40%
0.5
-0.4
-1.1
-1.2
-1.2
-1.2
-1.2
-1.2
-1.2
-1.2
50%
1.0
0.3
-0.3
-0.4
0.1
0.1
0.1
0.1
0.1
0.1
60%
1.3
0.7
0.3
0.2
0.6
1.1
1.1
1.1
1.1
1.1
70%
1.6
1.0
0.6
0.6
0.9
1.4
1.7
1.7
1.7
1.7
80%
1.8
1.3
0.9
0.9
1.2
1.6
1.9
2.3
2.3
2.3
90%
1.9
1.5
1.2
1.1
1.4
1.7
2.0
2.3
2.7
2.7
100%
2.0
1.6
1.4
1.3
1.5
1.9
2.1
2.4
2.7
3.5
Note that 3b jamming Ax2y is generally not +EV @ 25BB until villain is opening more than 50% of hands.
When deciding whether to 3b jam Ax2y readless you would want to know the population is opening their button fairly wide.
You could model various Ax holdings @ various effective stack sizes to find inflection points to make adjustments.
For example, 3b jamming AxKy @ 25BB is basically never going to be -EV play:
[C]
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
[R]
10%
4.5
4.5
4.5
4.5
4.5
4.5
4.5
4.5
4.5
4.5
20%
3.7
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
6.9
30%
3.5
5.6
7.7
7.7
7.7
7.7
7.7
7.7
7.7
7.7
40%
3.4
4.9
6.6
8.1
8.1
8.1
8.1
8.1
8.1
8.1
50%
3.3
4.6
5.8
7.1
8.2
8.2
8.2
8.2
8.2
8.2
60%
3.2
4.3
5.4
6.4
7.4
8.5
8.5
8.5
8.5
8.5
70%
3.2
4.1
5.0
5.9
6.7
7.7
8.5
8.5
8.5
8.5
80%
3.2
4.0
4.8
5.5
6.3
7.1
7.8
8.5
8.5
8.5
90%
3.2
3.9
4.6
5.3
5.9
6.7
7.3
7.9
8.6
8.6
100%
3.1
3.8
4.4
5.0
5.6
6.3
6.9
7.4
8.0
8.6
Because of the low opening frequencies of the population best is to shove A7+ (both suited and off suit) and flat the rest. Basically we only shove for thick value because his range is so strong. A7 is definitely good enough to shove though because you will often get called by weaker Ax which you dominate (unlike with A6 or A5 where you don't dominate weaker Ax because they chop vs you most of the time).
However, as soon as you get a read that your opponent is minraising decently more than average (so in the 60%+ region), shove the rest of the Ax. The analysis done above looks quite solid fwiw :)
As for large Ax, readless it's definitely better to 3b shove than 3b nai as villains tend to call 3b shoves much wider than they 4bs so you're not successfully inducing and just allowing villains to get a cheap price to flat you which they often take :)
Hey coffeeyay,
Do you have a robust method for estimating flatting expectation?
We can easily calculate the fair value of our equity in the pot immediately after we call, but flatting theoretically needs to also incoporate the expectation of future post flop actions.
I watched a Mersenneary video where he employed a "from start of hand" h% vs f% methodology for comparing 3b to flatting.
It was an extremely thought provoking video which I enjoyed very much.
But, unless I'm misunderstanding the methodology, this appears to simply lower the 3b expectation by the value of the big blind.
Essentially it adds a margin of safety to your 3b jamming expectation and assumes if you don't satisfy the risk-adjusted hurdle for 3b, then by deduction:
flatting > 3b jamming > folding
However this doesn't actually calculate the expectation of flatting..
I've recently been studying a little PLO and there are some really cool analytical techniques which approximate expectation of different actions pre-flop by considering the implied post-flop value by numerically integrating [H] v [R] flop distribution curves.
I'm not sure if Mersenneary is still around the site, or whether you can steer me in a direction that will translate into real $$ regarding analysing when flatting is better than 3b?
ie. calculating the marginal expectation of flatting vs 3b?
Cheers in advance, big fan of your work :)
What does integrating H & R flop distribution curves mean??
You can model implied value vs nutty NAI 3b ranges by taking the integral of a flop distribution curve across [o,m] where m is bottom of your range which has the required post flop equity to stack off.
eg. Take a villain who 3b NAI with [AA,KK,QQ] and will get all in postflop every time, can you call a NAI 3b with JTs?
http://www.propokertools.com/simulations/graph_hvr?g=he&h1=JxTx&h2=AA%2CKK%2CQQ&s=generic
For effective stacks S = 25.0 BB
SB = 0.5 BB
BB = 1.0 BB
P0 = 1.5 BB
R = 1.5 BB
P1 = 3.0 BB
3b = 5.0 BB
P2 = 8.0 BB
C = (5.0+1.0) - (1.5+0.5) = 4.0 BB
for share of
P3 = 12.0 BB
We need 4/12 = 33% equity to call a 3b.
We only have 19.75% equity, but do we have enough implied odds?
http://www.propokertools.com/simulations/show?g=he&h1=JxTx&h2=AA%2CKK%2CQQ&s=generic
How much equity do we need to stack off on the flop?
There is (S-P3/2) left behind going to the flop = (25-6) = 19.0 BB
which we assume will be going in.
So effectively we need (19.0/50.0) = 38% equity to stack off on the flop.
We can see from flop distribution curve that we hit enough equity about 10% of the time vs villain's assume 3b NAI range.
We can calculate the average equity we will have by taking an integral (area under curve) from [0,10] and dividing by the length (10).
There are numerical techniques for this such as trapezoidal method & simpsons.
For the sake of the forum post, crudely we can see area under the curve is a (38x10) rectangle + a (62x10) triangle which gives an area of:
A = 38 * 10 + 0.5 * 62 * 10 = 690
Divided by length 10, we can see on average when we have enough equity to stack off we will have (690/10) = 69% equity.
So what is our EV of calling a NAI 3b given these assumptions (keep in mind this is an implied odds model)?
It costs us 4.0 BB to call and see a flop.
10% of the time we will flop enough equity to stack off and on average will have 69% equity when we do so in a pot worth 2 * S = 50.0 BB the rest of the time (90%) we simply fold).
EV(call NAI 3b) = 0.9 * (- 4.0) + 0.1 * 0.69 * 50.0 = -3.6 + 3.45 = -0.15 BB
In this case we don't have enough implied odds to call the NAI 3b against villain's assumed range and tendencies.
However if the stacks were deeper or villain had a wider range (so we had a less polarised flop distribution) we could imagine this becoming a +EV play (net of the increased equity required to stack off as stacks get deeper).
Anyway, I'm really trying to develop a robust way to model the expectation of flatting vs 3b to incorporate properties of particular hands OOP and exploitable postflop population tendencies. Due to the disadvantage of position and popularity of cbetting we typically want to play hands that flop well. Hence my interest in [H] v [R] flop distribution curves.
I'm sure some of the more analytical guys around the site have given it some thought?
cdon nice work, Thankyou.
did you post any other text about this? (Quote: Anyway, I'm really trying to develop a robust way to model the expectation of flatting vs 3b to incorporate properties of particular hands OOP and exploitable postflop population tendencies.).
I'm sorry for my bad timing, but I find this very interesting.
Sacrifice minus profit equals zero
There was similar analysis posted here which will probably interest you:
http://www.husng.com/content/flatting-nai-3b-marginal-holdings-high-impl...
And the original articles which got me thinking about this methodology was a series of articles called 'PLO from scratch' found here:
http://en.donkr.com/Articles/
Inevitably, I've found that most forum participants are not prepared to put in the required effort to explore these topics thoroughly.
In hindsight I was the "more analytical guy around the site".
I've developed about 1/2 of a project would allow me to run simulations on these types of questions. But this year I've been really busy with work and haven't had much time for poker. I still really enjoy discussing strategy and still read a lot of the posts on this site so if you want to reignite any of my posts which stagnated due to lack of participation please feel free to do so :)
Thankyou for your help. I'm glad to hear what you said.
Sacrifice minus profit equals zero