Preflop stacking ranges

ChuckTs

ChuckTs

Legend
Silver Level
Joined
Feb 2, 2005
Total posts
13,642
Chips
0
This needs a rehash. (So do my math skills, but let's pretend I know what I'm doing for a moment).

Regardless of what villain's 3betting and/or 5betting range is made up of, we can show a profit by 4betting and calling it off when his 3bet is high enough.

This chart shows what 3bet range is needed from villain in order to 4bet and call it off profitably with said hands 100bbs deep:

* KK+: Good against any range
* QQ: vs 3% or higher
* AKs: vs 3% or higher
* AKo: vs 5% or higher
* JJ: vs 6.5% or higher
* AQs: vs 8% or higher
* TT: vs 8.5% or higher
* AQo: vs 9% or higher
* 99: vs 10.5% or higher
* 88: vs 12.5% or higher
* AJs: vs 13% or higher
* 77: vs 14% or higher

Let's do some math to demonstrate why. We're going to be ignoring when villain calls a 4bet here which is a little unrealistic (to assume they won't), but just for the sake of demonstration. (It's fair to say that we've still got good equity vs a 3bet/calling range, but to even approach this scenario with an EV equation is just not gonna happen).

100bb effective stacks. You open TT for 3bbs in MP, and villain 3bets to 9bbs on the button (the blinds fold). He's someone you have tons of hands on and you know has a BTN 3bet of %9.

1) Let's assume he only 3bets for value, and always calls it off once we 4bet. What's our EV of 4betting and calling it off? There are two possible scenarios; he either folds to the 4bet, or shoves.

Pot is 13.5bb (3bb open + 9bb 3bet + .5bb SB + 1bb BB). We 4bet to 22bbs (risking 19bbs; 22-3=19). If he shoves 100bbs, that gives us 78bbs to call into 22+100+1+.5=123.5bbs.

Our equity vs his stacking range is %51:

equity win tie pots won pots tied
Hand 0: 49.201% 48.53% 00.67% 548490306 7535467.50 { 88+, ATs+, KTs+, QJs, AJo+, KQo }
Hand 1: 50.799% 50.13% 00.67% 566559399 7535467.50 { TT }

We ignore the cases he folds since he never does, so:

EV = (amt in pot)(prob of winning)-(amt to call)(prob of losing)
=123.5(.51)-78(.49)
= ~25bbs



What about if he always folds?

Easier EV calculation.

EV = 13.5(%100)-0
EV = 13.5bbs



What about if his range is polarized and made up of some hands that stack and some that fold?

Let's say he only stacks JJ+/AK, the rest he folds. That's %3 which is 1/3 of his entire 3betting range. We have %34 equity vs that range.

EVtotal = EV1+EV2

1) %66 of the time we win 13.5bbs.

2) %33 of the time:
EV = 123.5(.34)-78(.66)
= ~-9.5bbs

Then EVtotal = 13.5+(-9.5) = 4bbs

Any other range makeup is going to yield a positive return.

Again, regardless of what villain's 3betting and/or 5betting range is made up of, we can show a profit by 4betting and calling it off when his 3bet is high enough.

***Worth noting***

Villain's 3bet percentage on your hud is NOT necessarily indicative of what his true situational 3bet will be. You may not have a good sample size, or he could be adjusting based on a miriad of factors. He could be tilting. He could be distracted and playing super nitty. Adjust your ranges accordingly.
 
T

themosthigh

Rock Star
Silver Level
Joined
Dec 22, 2012
Total posts
343
Chips
0
I just came across that chart the other day. Thanks alot for breaking it down. Great post!
 
acky100

acky100

Legend
Silver Level
Joined
Apr 22, 2010
Total posts
3,523
Chips
0
chart is sound and lets you make unexploitable decisions, think you should add a caveat though, there are often times where people justify stacking off with this chart because the chart says its fine where in reality there are much better lines to take (just because 4bet calling is > 0 EV doesn't mean calling isn't >> 0 EV ).

Just thought i'd throw it out there can't elaborate got exams to go to but people often make sub optimal plays which even though on paper are fine, at the end of the day are less profitable and basically going to make you less profitable than someone who makes the most profitable decision as opposed to the guy who just tries to minimize his losses by solely sticking to the chart.
 
micromachine

micromachine

Legend
Silver Level
Joined
Aug 15, 2011
Total posts
5,770
Chips
0
Very nice post, thanks!

One observation: I would say that with micro regs situation 2 is far more likely and I would even say the stacking off range is more like QQ+/AK with KK+ being more likely than QQ and AK (nitty I know but it's like that) so not sure if there is really profit to be made by 4betting and stacking off according to that chart at 5nl/10nl. At 25nl there is a lot more 3/4betting and regs are stacking off slightly lighter in preflop wars so it probably works there.

One question: How many hands would you say are needed to have reliable 3bet stats for each position?
 
micromachine

micromachine

Legend
Silver Level
Joined
Aug 15, 2011
Total posts
5,770
Chips
0
^^ another amazing comment from g_spot
 
youregoodmate

youregoodmate

Cardschat Elite
Silver Level
Joined
Jun 22, 2012
Total posts
2,683
Chips
0
Okay in all seriousness. You did nothing wrong on this hand. Online poker isn't rigged, you don't lose everytime you have AA or KK. You're just remembering the time where you bubbled a tournament with AA. There's literally not one thing you could have done differently.You have the ****ing nuts on the flop. Don't ever consider folding this even if you are on the bubble of any tournament, no matter the buy in.Sorry it didn't work out, but for god sakes this hand has no strat value whatsoever.

Ill have whatever this guys having.
 
Cafeman

Cafeman

Legend
Silver Level
Joined
Jan 12, 2011
Total posts
3,200
Chips
0
That might be a half decent answer to an unspecified question... maybe. Well done g_spot.
 
S

ScottishMatt

Legend
Silver Level
Joined
Sep 20, 2012
Total posts
2,394
Chips
0
Seriously though, nice chart. However as machine said, a lot of guys at 2/5/10NL are only stacking KK/AA and will either fold or call a 4-bet with QQ and AK. I suppose 25NL is probably the level where this becomes more applicable, I suppose if you are against a decent reg at lower levels then you could adopt the chart as a guideline.

Going to keep this bookmarked, thanks for the post Chuck.
 
benevg

benevg

Legend
Silver Level
Joined
Jul 22, 2007
Total posts
1,267
Chips
0
Seriously though, nice chart. However as machine said, a lot of guys at 2/5/10NL are only stacking KK/AA and will either fold or call a 4-bet with QQ and AK. I suppose 25NL is probably the level where this becomes more applicable, I suppose if you are against a decent reg at lower levels then you could adopt the chart as a guideline.

Going to keep this bookmarked, thanks for the post Chuck.
it is practically impossible to get reliable data on opponents' 3bet tendencies on 2/5/10 nl anyway?


Okay in all seriousness. You did nothing wrong on this hand. Online poker isn't rigged, you don't lose everytime you have AA or KK. You're just remembering the time where you bubbled a tournament with AA. There's literally not one thing you could have done differently.You have the ****ing nuts on the flop. Don't ever consider folding this even if you are on the bubble of any tournament, no matter the buy in.Sorry it didn't work out, but for god sakes this hand has no strat value whatsoever.
best non-sensical comment ever? bot-generated poker-themed reply? or maybe he replied to some thread in another part of the forum? ffs.
 
youregoodmate

youregoodmate

Cardschat Elite
Silver Level
Joined
Jun 22, 2012
Total posts
2,683
Chips
0
best non-sensical comment ever? bot-generated poker-themed reply? or maybe he replied to some thread in another part of the forum? ffs.

Just read a couple more of his recent posts and they make just as little sense. Hes just a troll bot.
 
micromachine

micromachine

Legend
Silver Level
Joined
Aug 15, 2011
Total posts
5,770
Chips
0
it is practically impossible to get reliable data on opponents' 3bet tendencies on 2/5/10 nl anyway?

10nl regs will play more predictably than 100nl regs no?
 
WVHillbilly

WVHillbilly

Legend
Silver Level
Joined
Nov 7, 2007
Total posts
22,973
Chips
0
Very nice post, thanks!

One observation: I would say that with micro regs situation 2 is far more likely and I would even say the stacking off range is more like QQ+/AK with KK+ being more likely than QQ and AK (nitty I know but it's like that) so not sure if there is really profit to be made by 4betting and stacking off according to that chart at 5nl/10nl. At 25nl there is a lot more 3/4betting and regs are stacking off slightly lighter in preflop wars so it probably works there.

One question: How many hands would you say are needed to have reliable 3bet stats for each position?

Seriously though, nice chart. However as machine said, a lot of guys at 2/5/10NL are only stacking KK/AA and will either fold or call a 4-bet with QQ and AK. I suppose 25NL is probably the level where this becomes more applicable, I suppose if you are against a decent reg at lower levels then you could adopt the chart as a guideline.

Going to keep this bookmarked, thanks for the post Chuck.

You guys seem to have missed this:
Again, regardless of what villain's 3betting and/or 5betting range is made up of, we can show a profit by 4betting and calling it off when his 3bet is high enough.
 
micromachine

micromachine

Legend
Silver Level
Joined
Aug 15, 2011
Total posts
5,770
Chips
0
yeah but that was assuming villain only stacks with JJ+/AK and EVtotal in the TT example was a mere 4bb. If villain only stacks off with KK+ it isn't profitable anymore. Or am I missing something?
 
tenbob

tenbob

Legend
Silver Level
Joined
May 16, 2005
Total posts
11,221
Awards
1
Chips
20
I have this chart attached to the wall beside my computer for the last ~2 years, and it makes a massive difference. When decision become close i just refer to it, having said that i will usually err on the side of the lower variance option when it arises. (ie i wont stack 1010 vs 8.4%) Having said that i play FR and im 99% sure that the vast majority of these sort of ranges apply to 6max games. It's pretty rare to see FR regs with a 3bet>8% on nl$25 these days.
 
benevg

benevg

Legend
Silver Level
Joined
Jul 22, 2007
Total posts
1,267
Chips
0
yeah but that was assuming villain only stacks with JJ+/AK and EVtotal in the TT example was a mere 4bb. If villain only stacks off with KK+ it isn't profitable anymore. Or am I missing something?
you are missing something. the +9bb that you win when he folds. which with your stack-off range will happen much more often ;) remember, this person needs to be 3betting over 8.5%. if he only 5bets KK+, then he folds absurdly often to 4 bets. :)
 
micromachine

micromachine

Legend
Silver Level
Joined
Aug 15, 2011
Total posts
5,770
Chips
0
ah yeah, thanks :)

Then if we actually know someone is 3betting like 9% but only stacking with KK+, would it not be better to 4bet with most of the suggested range vs a 9% 3bettor but to flat with our premiums (QQ+) to avoid missing out on value?
 
benevg

benevg

Legend
Silver Level
Joined
Jul 22, 2007
Total posts
1,267
Chips
0
ah yeah, thanks :)

Then if we actually know someone is 3betting like 9% but only stacking with KK+, would it not be better to 4bet with most of the suggested range vs a 9% 3bettor but to flat with our premiums (QQ+) to avoid missing out on value?
VVV
chart is sound and lets you make unexploitable decisions, think you should add a caveat though, there are often times where people justify stacking off with this chart because the chart says its fine where in reality there are much better lines to take (just because 4bet calling is > 0 EV doesn't mean calling isn't >> 0 EV ).

Just thought i'd throw it out there can't elaborate got exams to go to but people often make sub optimal plays which even though on paper are fine, at the end of the day are less profitable and basically going to make you less profitable than someone who makes the most profitable decision as opposed to the guy who just tries to minimize his losses by solely sticking to the chart.
 
S

Sori

Rock Star
Silver Level
Joined
Dec 7, 2011
Total posts
256
Chips
0
subbing for reference, seems like a valuable chart
 
WVHillbilly

WVHillbilly

Legend
Silver Level
Joined
Nov 7, 2007
Total posts
22,973
Chips
0
ah yeah, thanks :)

Then if we actually know someone is 3betting like 9% but only stacking with KK+, would it not be better to 4bet with most of the suggested range vs a 9% 3bettor but to flat with our premiums (QQ+) to avoid missing out on value?
If they're 3betting 9% and folding everything but KK+ you should adjust by 4betting WIDE and folding, but that doesn't mean it's not profitable to call. Remember, as acky pointed out, the chart is not about what is necessarily the best play, the chart only shows that certain hands are profitable to 4bet/call against certain ranges.
 
ChuckTs

ChuckTs

Legend
Silver Level
Joined
Feb 2, 2005
Total posts
13,642
Chips
0
Was starting to write responses to all the questions and concerns about hand selection and stuff but noticed WVH and benevg have it covered.

The chart shows how wide a range needs to be to 4bet and stack one of said hands to show a profit. Nothing more. There are times to 4bet a polarized range against a certain villain. There are times to flat the 3bet. Both are often higher EV than stacking one of these hands, but the chart just shows you can't go wrong by doing so.
 
micromachine

micromachine

Legend
Silver Level
Joined
Aug 15, 2011
Total posts
5,770
Chips
0
Was starting to write responses to all the questions and concerns about hand selection and stuff but noticed WVH and benevg have it covered.

One question: How many hands would you say are needed to have reliable 3bet stats for each position?

No-one answered this question :( I normally just look at total 3bet stat but want to start using positional 3bet stats for regs, which I suppose would take approx 3 times as long to converge, so like 600-1K hands to get a reasonable idea?
 
ChuckTs

ChuckTs

Legend
Silver Level
Joined
Feb 2, 2005
Total posts
13,642
Chips
0
I think like 50 hands and I'm starting to group players as aggro or not aggro w.r.t. their 3bet %. A few hundred hands and I actually start relying on the stat, but not leaning on it too much. Over that and we can start actually assigning ranges with some degree of certainty based on their overall 3bet %.

But that's just me feeling it out, I haven't run or read anything about hand samples and converging in a while.

I don't look at positional 3bet %s until I have at least like 3-500 hands on a guy. Even then I don't think it's something to really bank on. I don't really rely on actual percentages, rather just look to see if anything's really out of whack, like they don't 3bet anywhere but the btn, or they 3bet excessively out of the blinds or whatever.

A good rule of thumb for stat convergence is that the more requirements need to be met in order for the stat to register, the longer it'll take to converge. Like river cbet % requires a it to have folded to them pf, them to make a pf raise, flop cbet and turn cbet, and for them not to have been raised anywhere. PFR on the other hand only requires that it's folded to them and that they raise.
 
benevg

benevg

Legend
Silver Level
Joined
Jul 22, 2007
Total posts
1,267
Chips
0
.... Like river cbet % requires a it to have folded to them pf, them to make a pf raise, flop cbet and turn cbet, and for them not to have been raised anywhere. PFR on the other hand only requires that it's folded to them and that they raise.
wait, what? river cbet requires a flop and turn cbets? that doesn't sound right. let me go look it up :)

(edit) darn, apparently it does. :-o
 
WVHillbilly

WVHillbilly

Legend
Silver Level
Joined
Nov 7, 2007
Total posts
22,973
Chips
0
wait, what? river cbet requires a flop and turn cbets? that doesn't sound right. let me go look it up :)

(edit) darn, apparently it does. :-o
Remember that C stands for Continuation and it's not continuous if you stop betting and check a street. Raise/ Bet/Bet/Bet is the only way!
 
Top