COTW: Thinking About Combos
09-27-2010
, 04:39 AM
“Why do we care about combos?”:
Many people do not fully understand combos, yet alone thinking in terms of combos while making real-time poker decisions. Thinking about combos can help us visualize a player's range, which ultimately can help us with hand reading. When we think about reading combos we are considering the detailed range of hands that a player can have. Let's look at a few examples:
The Preflop Usage:
Here are the bare basics of combos if we think about things from preflop:
Let's look a hand:
Poker Stars $0.5/$1 No Limit Hold'em - 7 players
The Official 2+2 Hand Converter Powered By DeucesCracked.com
CO: $58.90
Hero (BTN): $142.10
SB: $68.75
BB: $105.20
UTG: $102.50
UTG+1: $94.50
MP: $262.90
Pre Flop: ($1.5) Hero is BTN with ::X:: ::X::
UTG raises to $3, 3 folds, Hero raises to $9.4
Say UTG an opponent who raises with 22+/AJ+/KQ and we are considering 3betting him. We think he would fold everything but QQ+/AK and we want to run the outright $EV of our 3bet. Let's think about his range of hands based on combos:
Pairs (22-AA): 78 combos
Suited (AJs, AQs, AKs, KQs): 16 combos
Unsuited (AJo, AQo, AKo, KQo): 48 combos
His opening range is 142 combos of hands. Now let's think about how often he continues if we 3bet him:
Pairs (QQ-AA): 18 combos
AK: 16 combos
The range of hands that would continue has 34 combos. Simple math says he would be folding 100% - (34/142) = 76%
Well that's obviously great...but let's change the situation slightly and say that we have A4s rather than two blank cards. Because we have A4s we are now blocking his Ax hands, and also blocking some combos of 44 and AA. Now let's run the math:
Pairs (22-AA): 72 combos
Suited (AJs, AQs, AKs, KQs): 12 combos
Unsuited (AJo, AQo, AKo, KQo): 36 combos
His opening range is 120 combos of hands. Now let's think about how often he continues if we 3bet him:
Pairs (QQ-AA): 15 combos
AK: 12 combos
The range of hands that would continue has 27 combos. Simple math says he would be folding 100% - (27/120) = 77.5%
That is of course better than a $1.16 profit (and even though a .21bb probably doesn't seem like a massive difference, this kind of situation arrises often and any extra profit can always pad our bottom line. This is why considering blockers is so important, because they can help our outright $EV quite a bit by using hands like Axs, Kxs, etc (and this is all just outright, before we even talk about the added value of equity postflop)
The Flop Usage:
Postflop we want to be thinking about combos to really nail down our opponent's range. Once we are able to appropriately assign a range of hands, we will able to take more optimal lines postflop. While we should use our opponent's postflop actions to help us visualize range more correctly, we can use combos to help give backing to our lines. Let's look at a few things to think about:
Let's look at a few different ranges versus boards assuming that players will always continue (by either raising or calling) when they have big draws or middle pair+:
A few things to note here:
1.) Notice how often draw are something to be feared.
2.) Notice how often TP was hit on an A high board versus a K high board.
3.) When ranges are broadway heavy, they miss low boards often
Let's run this against a cold call range of a fish. We will give the fish a range of lots of suited cards, broadway hands, and assume that he continues with any pair and any draw better than a gutshot:
A few things to note here:
1.) Notice how often draws are something to be feared? Even with massive amounts of draw-possible hands in his range only make for a max 15% draws on the flop
2.) Notice how often he misses low boards when his range is so Bx heavy
3.) Notice how often he tends to hit the Axx board when so much of his range PF is made with Ax hands
While our cards would certainly matter when thinking about combos, especially if we have a hand like AQ and block out a fish's possible Ax hands and Qx hands, this should help us visualize a few different boards against some standard ranges. We want to do a lot of combo work off the table, so that when we are playing we are able to just naturally understand “if he has X in his PF range, then he would be hitting this board Y% of the time.” Here are a few hints for reading postflop combos:
“Can I fold XXX hand?”:
We see these kinds of threads in the forum all of the time. “Can I fold middle set?” “Can I fold this boat?” “Do you see a fold here?” Many of these threads are results-oriented thinking that do nothing for the progress of the OP. So let's take a hand and talk about it.
Poker Stars $0.50/$1 No Limit Hold'em - 9 players
The Official 2+2 Hand Converter Powered By DeucesCracked.com
MP1: $130.30
MP2: $88.60
CO: $100.00
Hero (BTN): $246.50
SB: $34.00
BB: $93.25
UTG: $100.00
UTG+1: $20.00
UTG+2: $101.50
Pre Flop: ($1.50) Hero is BTN with 7
7
UTG raises to $4, 5 folds, Hero calls $4, 2 folds
Flop: ($9.50) 2 7
9
(2 players)
UTG bets $6, Hero calls $6
Turn: ($21.50) J (2 players)
UTG bets $13, Hero raises to $35, UTG raises to $90, Hero ???
Let's break this down simply:
Combos of hands that beat us: 6 (3 sets of 99, and 3 sets of JJ)
We are calling $55 to win $146.5 (ignoring rake). This means we need to be good at least 27% of the time to call.
Because there are 6 hands that beat us, and we are getting ~3:1, we would need our opponent to have at least 2 combos of hands that we beat in his range.
Can we think of 2 combos that would do this? Well there are 3 combos of 22 that we beat, and assuming 22 is in his UTG PF raising range, they can definitely be here. What about a few combos of AA or KK that overvalue themselves? What about a shove with AcKc? What about a tilt shove allocating for maybe a partial combo or two?
When we really think about combos of hands that beat us versus combos of hands we beat, we see that we usually cannot be folding sets and boats. Things can get different when we change our hand to something like a baby flush or bottom two pair...but ultimately when thinking about hand reading we think about how many combos can realistically* beat us, and then consider all of the other combos of hands he might take this line with. If the math and combos line up, we can make our action. If they don't, we might need to consider a different line...
*Realistically is important. We want to think about real combos that he could have...not say “omg, 64 makes the nut and he could have that" even though he is a tight guy that raised from UTG
The Cliff Notes Conclusion:
Enjoy, and let's have a conversation
Many people do not fully understand combos, yet alone thinking in terms of combos while making real-time poker decisions. Thinking about combos can help us visualize a player's range, which ultimately can help us with hand reading. When we think about reading combos we are considering the detailed range of hands that a player can have. Let's look at a few examples:
The Preflop Usage:
Here are the bare basics of combos if we think about things from preflop:
Quote:
- Pairs: Any pair we put in somebody's range makes 6 combos (8c8h, 8c8d, 8c8s, 8h8d, 8h8s, 8s8d)
- XXs: Any suited hand makes up 4 combos (AdKd, AsKs, AcKc, AhKh)
- XXo: Any unsuited hand makes up 12 combos (AcKh, AcKd, AcKs, AdKh, AdKs, AdKc, AhKs, AhKc, AhKd, AsKh, AsKc, AsKd)
- AK: Any unpaired hand (suited or unsuited) has 16 combos
Poker Stars $0.5/$1 No Limit Hold'em - 7 players
The Official 2+2 Hand Converter Powered By DeucesCracked.com
CO: $58.90
Hero (BTN): $142.10
SB: $68.75
BB: $105.20
UTG: $102.50
UTG+1: $94.50
MP: $262.90
Pre Flop: ($1.5) Hero is BTN with ::X:: ::X::
UTG raises to $3, 3 folds, Hero raises to $9.4
Say UTG an opponent who raises with 22+/AJ+/KQ and we are considering 3betting him. We think he would fold everything but QQ+/AK and we want to run the outright $EV of our 3bet. Let's think about his range of hands based on combos:
Pairs (22-AA): 78 combos
Suited (AJs, AQs, AKs, KQs): 16 combos
Unsuited (AJo, AQo, AKo, KQo): 48 combos
His opening range is 142 combos of hands. Now let's think about how often he continues if we 3bet him:
Pairs (QQ-AA): 18 combos
AK: 16 combos
The range of hands that would continue has 34 combos. Simple math says he would be folding 100% - (34/142) = 76%
Quote:
outright$EV = .76*($4.5) - .24*($9.4) = 3.42 – 2.26 = $1.16 (or 1.16bb)
Pairs (22-AA): 72 combos
Suited (AJs, AQs, AKs, KQs): 12 combos
Unsuited (AJo, AQo, AKo, KQo): 36 combos
His opening range is 120 combos of hands. Now let's think about how often he continues if we 3bet him:
Pairs (QQ-AA): 15 combos
AK: 12 combos
The range of hands that would continue has 27 combos. Simple math says he would be folding 100% - (27/120) = 77.5%
Quote:
outright$EV = .775*($4.5) - .225*($9.4) = 3.49 – 2.12 = $1.37 (or 1.37bb)
The Flop Usage:
Postflop we want to be thinking about combos to really nail down our opponent's range. Once we are able to appropriately assign a range of hands, we will able to take more optimal lines postflop. While we should use our opponent's postflop actions to help us visualize range more correctly, we can use combos to help give backing to our lines. Let's look at a few things to think about:
Let's look at a few different ranges versus boards assuming that players will always continue (by either raising or calling) when they have big draws or middle pair+:
A few things to note here:
1.) Notice how often draw are something to be feared.
2.) Notice how often TP was hit on an A high board versus a K high board.
3.) When ranges are broadway heavy, they miss low boards often
Let's run this against a cold call range of a fish. We will give the fish a range of lots of suited cards, broadway hands, and assume that he continues with any pair and any draw better than a gutshot:
A few things to note here:
1.) Notice how often draws are something to be feared? Even with massive amounts of draw-possible hands in his range only make for a max 15% draws on the flop
2.) Notice how often he misses low boards when his range is so Bx heavy
3.) Notice how often he tends to hit the Axx board when so much of his range PF is made with Ax hands
While our cards would certainly matter when thinking about combos, especially if we have a hand like AQ and block out a fish's possible Ax hands and Qx hands, this should help us visualize a few different boards against some standard ranges. We want to do a lot of combo work off the table, so that when we are playing we are able to just naturally understand “if he has X in his PF range, then he would be hitting this board Y% of the time.” Here are a few hints for reading postflop combos:
- Pocket Pairs: 6 combos of missed sets, 3 combos of sets, 1 combo of quads if the board is paired
- Draws: are almost never as big of a concern as you think they are
- Big card boards: Say the board is Axx, it drops their possible combos per Ax hand from 16 down to 12
- 2pears?: Say the board is AQx, it drops their possible combos of AQ from 16 down to 9
- Think about PF: Hand reading should be linear and logical.
- Actions indicate range: The actions that a player makes should give us hints about their specific range.
“Can I fold XXX hand?”:
We see these kinds of threads in the forum all of the time. “Can I fold middle set?” “Can I fold this boat?” “Do you see a fold here?” Many of these threads are results-oriented thinking that do nothing for the progress of the OP. So let's take a hand and talk about it.
Poker Stars $0.50/$1 No Limit Hold'em - 9 players
The Official 2+2 Hand Converter Powered By DeucesCracked.com
MP1: $130.30
MP2: $88.60
CO: $100.00
Hero (BTN): $246.50
SB: $34.00
BB: $93.25
UTG: $100.00
UTG+1: $20.00
UTG+2: $101.50
Pre Flop: ($1.50) Hero is BTN with 7
7UTG raises to $4, 5 folds, Hero calls $4, 2 folds
Flop: ($9.50) 2
7 9 (2 players)UTG bets $6, Hero calls $6
Turn: ($21.50) J
(2 players)UTG bets $13, Hero raises to $35, UTG raises to $90, Hero ???
Let's break this down simply:
Combos of hands that beat us: 6 (3 sets of 99, and 3 sets of JJ)
We are calling $55 to win $146.5 (ignoring rake). This means we need to be good at least 27% of the time to call.
Because there are 6 hands that beat us, and we are getting ~3:1, we would need our opponent to have at least 2 combos of hands that we beat in his range.
Can we think of 2 combos that would do this? Well there are 3 combos of 22 that we beat, and assuming 22 is in his UTG PF raising range, they can definitely be here. What about a few combos of AA or KK that overvalue themselves? What about a shove with AcKc? What about a tilt shove allocating for maybe a partial combo or two?
When we really think about combos of hands that beat us versus combos of hands we beat, we see that we usually cannot be folding sets and boats. Things can get different when we change our hand to something like a baby flush or bottom two pair...but ultimately when thinking about hand reading we think about how many combos can realistically* beat us, and then consider all of the other combos of hands he might take this line with. If the math and combos line up, we can make our action. If they don't, we might need to consider a different line...
*Realistically is important. We want to think about real combos that he could have...not say “omg, 64 makes the nut and he could have that" even though he is a tight guy that raised from UTG
The Cliff Notes Conclusion:
- Think about combos when hand reading
- Don't be a results-oriented fish and think you should have folded when you know damn well you never should have
- Allocate for “idiocy” in combos, especially against dumb-aggressive and/or fishy opponents
- Understand why fearing draws is silly
- Understand how textures effect combos
- Grind some of this work off the table and internalize it
Enjoy, and let's have a conversation
09-27-2010
, 04:40 AM
09-27-2010
, 06:27 AM
09-27-2010
, 06:48 AM
Very good post.
3 things I'd like to add which are important:
1. While 0.21bb may not seem like much, in every identical scenario we would stand to make 10.5BB/100 which is a huge amount of $$$! I mean most of us would be pretty stoked with a winrate of half that! Note by identical, I mean exact same ranges etc. Insert comment about ranges being dynamic etc.
2. While fishes ranges tend to hit more often in a pure % than a TAG will, what they hit will tend to be much weaker.
Let's compare gutshots, top pair and overpairs and set %'s in the ranges given above:
Gutshot (weak draw otf which still counts as a piece):
Flop 1. TAG=0%, Fish=4.96%
Flop 2. TAG=0%, Fish=1.06%
Flop 3. TAG=13.2%, Fish=16.3%
Top pair and overpairs (strong 1 pair hands):
Flop 1. TAG=22.6%, Fish=21.6%
Flop 2. TAG=29.8%, Fish=22.2%
Flop 3. TAG=24.8%, Fish=14.5%
Set (monster otf):
Flop 1. TAG=6.77%, Fish=2.23%
Flop 2. TAG=7.44%, Fish=1.59%
Flop 3. TAG=7.44%, Fish=1.53%
Fish will also have absolutely "nothing" far more often too (not sure how flopzilla defines nothing).
3. Just because a tight player opens UTG doesn't mean he doesn't have 64. I remember playing against a UTG 5/4 (5/4 overall that is he played 3/3 UTG over around 1200 hands total) with 88 (he overplayed 1 pair type hands, so perfect to set mine especially otb). 100BB effective stacks. 9BB otf 758. He bet 7BB, I raised to 26BB, he shoved for 96BB. Getting 1.86-1 against an almost certain overpair made snap calling a no brainer. There was a moment where my mouth just dropped as he turned over 96 for a flopped straight. I always remember this hand and will to the day I die because everyone makes strange plays. I've looked at this hh 100s of times - no obvious fish at the table, no deepstack play here, nothing significant to note. He just made a "balancing" play, (I've since asked the guy if it was a misclick, he says no - take what you will from that) - point is, even nits make strange plays. If 64 makes the nuts, he could very well have 64 (it's a minute chance of course and shouldn't effect our decision making process too much especially at 100BB deep), but in deep stack play (like 400BB+), if 64 makes the nuts and you face a large raise, you at least need to consider it and see if it is consistent with postflop play and how likely it is etc. Just a point to note for completeness sake.
3 things I'd like to add which are important:
1. While 0.21bb may not seem like much, in every identical scenario we would stand to make 10.5BB/100 which is a huge amount of $$$! I mean most of us would be pretty stoked with a winrate of half that! Note by identical, I mean exact same ranges etc. Insert comment about ranges being dynamic etc.
2. While fishes ranges tend to hit more often in a pure % than a TAG will, what they hit will tend to be much weaker.
Let's compare gutshots, top pair and overpairs and set %'s in the ranges given above:
Gutshot (weak draw otf which still counts as a piece):
Flop 1. TAG=0%, Fish=4.96%
Flop 2. TAG=0%, Fish=1.06%
Flop 3. TAG=13.2%, Fish=16.3%
Top pair and overpairs (strong 1 pair hands):
Flop 1. TAG=22.6%, Fish=21.6%
Flop 2. TAG=29.8%, Fish=22.2%
Flop 3. TAG=24.8%, Fish=14.5%
Set (monster otf):
Flop 1. TAG=6.77%, Fish=2.23%
Flop 2. TAG=7.44%, Fish=1.59%
Flop 3. TAG=7.44%, Fish=1.53%
Fish will also have absolutely "nothing" far more often too (not sure how flopzilla defines nothing).
3. Just because a tight player opens UTG doesn't mean he doesn't have 64. I remember playing against a UTG 5/4 (5/4 overall that is he played 3/3 UTG over around 1200 hands total) with 88 (he overplayed 1 pair type hands, so perfect to set mine especially otb). 100BB effective stacks. 9BB otf 7
58. He bet 7BB, I raised to 26BB, he shoved for 96BB. Getting 1.86-1 against an almost certain overpair made snap calling a no brainer. There was a moment where my mouth just dropped as he turned over 96 for a flopped straight. I always remember this hand and will to the day I die because everyone makes strange plays. I've looked at this hh 100s of times - no obvious fish at the table, no deepstack play here, nothing significant to note. He just made a "balancing" play, (I've since asked the guy if it was a misclick, he says no - take what you will from that) - point is, even nits make strange plays. If 64 makes the nuts, he could very well have 64 (it's a minute chance of course and shouldn't effect our decision making process too much especially at 100BB deep), but in deep stack play (like 400BB+), if 64 makes the nuts and you face a large raise, you at least need to consider it and see if it is consistent with postflop play and how likely it is etc. Just a point to note for completeness sake.
09-27-2010
, 04:40 PM
Quote:
3. Just because a tight player opens UTG doesn't mean he doesn't have 64. I remember playing against a UTG 5/4 (5/4 overall that is he played 3/3 UTG over around 1200 hands total) with 88 (he overplayed 1 pair type hands, so perfect to set mine especially otb). 100BB effective stacks. 9BB otf 758. He bet 7BB, I raised to 26BB, he shoved for 96BB. Getting 1.86-1 against an almost certain overpair made snap calling a no brainer. There was a moment where my mouth just dropped as he turned over 96 for a flopped straight. I always remember this hand and will to the day I die because everyone makes strange plays. I've looked at this hh 100s of times - no obvious fish at the table, no deepstack play here, nothing significant to note. He just made a "balancing" play, (I've since asked the guy if it was a misclick, he says no - take what you will from that) - point is, even nits make strange plays. If 64 makes the nuts, he could very well have 64 (it's a minute chance of course and shouldn't effect our decision making process too much especially at 100BB deep), but in deep stack play (like 400BB+), if 64 makes the nuts and you face a large raise, you at least need to consider it and see if it is consistent with postflop play and how likely it is etc. Just a point to note for completeness sake.
58. He bet 7BB, I raised to 26BB, he shoved for 96BB. Getting 1.86-1 against an almost certain overpair made snap calling a no brainer. There was a moment where my mouth just dropped as he turned over 96 for a flopped straight. I always remember this hand and will to the day I die because everyone makes strange plays. I've looked at this hh 100s of times - no obvious fish at the table, no deepstack play here, nothing significant to note. He just made a "balancing" play, (I've since asked the guy if it was a misclick, he says no - take what you will from that) - point is, even nits make strange plays. If 64 makes the nuts, he could very well have 64 (it's a minute chance of course and shouldn't effect our decision making process too much especially at 100BB deep), but in deep stack play (like 400BB+), if 64 makes the nuts and you face a large raise, you at least need to consider it and see if it is consistent with postflop play and how likely it is etc. Just a point to note for completeness sake.
09-27-2010
, 04:40 PM
