Shoving with very little blinds left- ICM

J

jaded848

Rock Star
Silver Level
Joined
Jul 5, 2009
Total posts
325
Chips
0
While playing around with SnG Power Tools, I tried a scenario that happens often. I put myself in the CO with about 5 BBs left and gave myself a hand like K2o. I gave all the positions ahead of me either equal or more chips than I had (4 players in total including myself). Surprisingly, I found that to shove with a hand like K2o, I would have to be down to a mere 2 big blinds!!

I was always under the impression that it is better to shove earlier than later, and in the past with a hand like K2o I might have shoved with my 5 BBs left- but apparently that play is very -EV.

Alas, I am still a big noob when it comes to ICM :(
 
Jillychemung

Jillychemung

Stacks & Stacks
Loyaler
Joined
Feb 7, 2008
Total posts
8,252
Awards
1
Chips
145
What kind of calling ranges did you give to your opponents?
 
J

jaded848

Rock Star
Silver Level
Joined
Jul 5, 2009
Total posts
325
Chips
0
Well that's part of my problem- in fact, it might be my biggest problem. I have a real hard time assigning ranges to people. How can the HUD help me with this? Is there a way to make generalizations based on VPIP?

In regards to the example I stated above, I gave them fairly wide ranges of 22+, A7s, KTs, etc with the person with the most chips having the widest range.
 
T

The_Pup

Rock Star
Silver Level
Joined
Oct 27, 2009
Total posts
254
Chips
0
I am not familiar with 'Power Tools' so I may be talking out of my hat, please forgive me if this is nonsense.

ICM gives the value of your stack as a percentage of the total prize available. If there are say 10,000 chips in circulation and the prize pool is $100 it doesn't mean that your stack of 1000 chips is worth $10; and if you double up you do not double the value of your stack in terms of the prize pool (it will usually be less than double). So, if you assign a calling range of top 30% (roughly the range you say) to the other 3 opponents then on average one of them will call, and your K2 will be a 60/40 underdog. From this we can see that you are taking a 60/40 gamble to gain less than double the value of your stack. This can make sense when you are down to a blind or two as you are going all in either this hand or the next two and K2 is as good as anything.

Perhaps the way out of this is to change your opponents' calling ranges according to the size of your stack. If you are at 2BB their range will be a lot wider than if you have 5 or 10BB.
 
cjatud2012

cjatud2012

¯\_(ツ)_/¯
Silver Level
Joined
Mar 12, 2009
Total posts
3,904
Chips
0
While playing around with SnG Power Tools, I tried a scenario that happens often. I put myself in the CO with about 5 BBs left and gave myself a hand like K2o. I gave all the positions ahead of me either equal or more chips than I had (4 players in total including myself). Surprisingly, I found that to shove with a hand like K2o, I would have to be down to a mere 2 big blinds!!

I was always under the impression that it is better to shove earlier than later, and in the past with a hand like K2o I might have shoved with my 5 BBs left- but apparently that play is very -EV.

Alas, I am still a big noob when it comes to ICM :(

Two things:

1) I don't think applications like SNG Wiz or SNG Power Tools take into account the upcoming blinds. That is, it doesn't know that it would be better to shove any two cards with 5 BB's, when you have more fold equity, than it would be to wait for decent cards, but then you leave yourself little chance of stealing the blinds. So in this case, although it may be a -$EV move to ship in this spot, it is almost always better to do that than wait.

2) Your pushing range gets much more narrow as your position becomes worse. In your example, we are in the CO, so there are 3 players left to act behind us. If each of them is calling the top 25-30% of their range, the majority of the time, we can expect to be called. Compare this to when we are in the SB with the same stack. We have only one player left to act behind us, and therefore the chances of us are getting called are much less. I've run both examples through Hold'em Resources' ICM Calculator, and this is what is spits out. Basically, our push range is much wider from the SB than it is from the CO, with the same stacks.

HERO IS CO
ICM Nash Calculator Results

HERO IS SB
ICM Nash Calculator Results
 
J

jaded848

Rock Star
Silver Level
Joined
Jul 5, 2009
Total posts
325
Chips
0
Two things:

1) I don't think applications like SNG Wiz or SNG Power Tools take into account the upcoming blinds. That is, it doesn't know that it would be better to shove any two cards with 5 BB's, when you have more fold equity, than it would be to wait for decent cards, but then you leave yourself little chance of stealing the blinds. So in this case, although it may be a -$EV move to ship in this spot, it is almost always better to do that than wait.

2) Your pushing range gets much more narrow as your position becomes worse. In your example, we are in the CO, so there are 3 players left to act behind us. If each of them is calling the top 25-30% of their range, the majority of the time, we can expect to be called. Compare this to when we are in the SB with the same stack. We have only one player left to act behind us, and therefore the chances of us are getting called are much less. I've run both examples through Hold'em Resources' ICM Calculator, and this is what is spits out. Basically, our push range is much wider from the SB than it is from the CO, with the same stacks.

HERO IS CO
ICM Nash Calculator Results

HERO IS SB
ICM Nash Calculator Results


Thanks for the response- I am trying to interpret those results. Now, those percentages are (i'm assuming) based on the fact that all players are aware of ICM and will play their hands perfectly according to ICM- is this correct?

Also, if anyone has an answer regarding how to estimate calling ranges using a HUD (or any method, really), please feel free to comment!
 
cjatud2012

cjatud2012

¯\_(ツ)_/¯
Silver Level
Joined
Mar 12, 2009
Total posts
3,904
Chips
0
Thanks for the response- I am trying to interpret those results. Now, those percentages are (i'm assuming) based on the fact that all players are aware of ICM and will play their hands perfectly according to ICM- is this correct?
Yes, this ICM calculator does not take into account player tendencies. However, I still believe these results illustrate why tools like SNG Wiz and SNG Power Tools disagree with your CO/UTG shove-- we have too many players left to act behind us. But, like I said, it is always better to shove when we have a little bit of fold equity than wait for when we have no fold equity at all, although our tools may disagree.

Also, if anyone has an answer regarding how to estimate calling ranges using a HUD (or any method, really), please feel free to comment!

Well, we can't really use a HUD to say "this person will call with 66, but not 55". Rather, based on whether a person's VPIP (voluntarily put $ into pot percentage), we can assign a general range for each player. For example, if a player's VPIP is 30%, we can expect this player to call with all pocket pairs, most aces, and various broadway hands.

Stack sizes is another important factor to consider for calling, probably more important than VPIP stats. If we have 3-5 blinds, and we are shoving into the BB who has 25-30 blinds, we can expect this player to call more often than a villain who has 12 blinds, because the big stacked player will recognize that he can knock out a short stack and not put a lot of his stack at risk, not to mention we're offering good pot odds. Basically, we want to prevent our stack from getting this small, and we also want to target players with medium stacks, say between 15-20 blinds.
 
J

jaded848

Rock Star
Silver Level
Joined
Jul 5, 2009
Total posts
325
Chips
0
Yes, this ICM calculator does not take into account player tendencies. However, I still believe these results illustrate why tools like SNG Wiz and SNG Power Tools disagree with your CO/UTG shove-- we have too many players left to act behind us. But, like I said, it is always better to shove when we have a little bit of fold equity than wait for when we have no fold equity at all, although our tools may disagree.



Well, we can't really use a HUD to say "this person will call with 66, but not 55". Rather, based on whether a person's VPIP (voluntarily put $ into pot percentage), we can assign a general range for each player. For example, if a player's VPIP is 30%, we can expect this player to call with all pocket pairs, most aces, and various broadway hands.

Stack sizes is another important factor to consider for calling, probably more important than VPIP stats. If we have 3-5 blinds, and we are shoving into the BB who has 25-30 blinds, we can expect this player to call more often than a villain who has 12 blinds, because the big stacked player will recognize that he can knock out a short stack and not put a lot of his stack at risk, not to mention we're offering good pot odds. Basically, we want to prevent our stack from getting this small, and we also want to target players with medium stacks, say between 15-20 blinds.

Thank you, this is good advice. My problem with using the VPIP is that it is an average based on every hand from the start of the match (assuming I never played with the person before). So, it is likely to often be lower than what the person's actually raising/calling range would be, assuming they are a good player and understand they need to play more loose aggressive as the sng goes on. How do you deal with this problem??

If I were to play only 1 sng at a time, it would probably be easier to estimate opponents ranges. However, like many other players, I like to play more than one at once, which I why I'm very concerned with using the HUD.
 
cardplayer52

cardplayer52

Cardschat Elite
Silver Level
Joined
Mar 28, 2009
Total posts
1,232
Chips
0
i'll usually put the villians calling range closer to there pfr than there vpip. ofc this depends on how the villian has been playing. also if i'm very short or they are i'll make their calling ranges a little looser than their vpip. another thing i'm finding is the smaller stacks seem to be calling looser than the bigger ones. seems people fear doubling up the short stacks while the short stacks figure it's now or never.
 
cjatud2012

cjatud2012

¯\_(ツ)_/¯
Silver Level
Joined
Mar 12, 2009
Total posts
3,904
Chips
0
Thank you, this is good advice. My problem with using the VPIP is that it is an average based on every hand from the start of the match (assuming I never played with the person before). So, it is likely to often be lower than what the person's actually raising/calling range would be, assuming they are a good player and understand they need to play more loose aggressive as the sng goes on. How do you deal with this problem??

The short answer is that it depends. It's not always true, but our HUD is the often the best thing we have to predict our opponents actions. If a player has stats of 46/3/0.5, then you pretty much know that he is not a good player. In high blind situations, he might over adjust and play too tight, or he might continue his loose calling tendencies. It's hard to know for sure. You know if a player is a 4/2/1, he generally won't adjust his calling range when the blinds escalate. If a player has stats like 12/10/4, he might actually be a good player. If he also thinks you are a good player, he might widen his calling range when you push, if the stack sizes are right, since he knows you are shoving wide.

There's really a lot to consider, and there's no clear cut answers. The HUD gives us a good guideline to follow, though, for how players will tend to play. That's really the best we can do.
 
V

volpereira

Enthusiast
Silver Level
Joined
Jan 20, 2010
Total posts
78
Chips
0
In this situation I would put my opponents in any2 hand...
 
Top