Betting for value
Jan. 21st, 2006 09:13 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
peterbirksOver the past six months there’s been an increased tendency for players to bet for value on the river. As far as I am concerned, this is all for the good. Knowing when to bet for value, when to check (and let your opponent, who has missed, try a bluff) and when to risk a check-raise are techniques that come with experience. I’m seeing a lot of “bets for value” that, as far as I am concerned, is money thrown away.
((For the non-poker cognoscenti out there, to “bet for value” on the river is to make a bet with a hand that is neither very good nor very weak, but which you think will make you more money on in the long run by betting than you would by checking. Pot Limit players find this one of the hardest parts of limit to master.)
However, it also set me wondering, after this hand, whether more might be made of this. In particular, if you are getting a significant number of bets for value, might we have reached a tipping point where you should raise for value or even check-raise for value? For a start, it blows up a lot of the “bet for value” mathematics. The usual principle with a bet for value is that your opponent will throw away (or occasionally raise with) garbage, raise with something brilliant, and call with all the hands that might be winning but might not. If your opponent starts raising on the river with hands that might be winning but might not, your bet for value begins to look less attractive.
$3/$6 Hold’em'
Seat 7 is the button
Seat 1: HoldemStever ( $69.49 )
Seat 2: csmith945 ( $0 )
Seat 3: EdwardGrover ( $258 )
Seat 4: scooter_c ( $221 )
Seat 5: mryumyum22 ( $107 )
Seat 6: (Posting) rubymuc ( $150 )
Seat 7: (Button) e8KXmBFk ( $245.50 )
Seat 8: (Small Blind) AA______ ( $256.52 )
Seat 9: (Big Blind) Birks ( $317 )
Seat 10: isley ( $183 )
AA______ posts small blind [$1].
Birks posts big blind [$3].
rubymuc posts big blind [$3].
Dealt to Birks 3♦, Q♦, (those are diamonds, by the way, my browser seems to be showing all the suits but diamonds. weird)
isley folds.
HoldemStever calls [$3].
EdwardGrover folds.
scooter_c calls [$3].
mryumyum22 calls [$3].
rubymuc checks.
e8KXmBFk folds.
AA______ calls [$2].
Birks checks.
Dealing Flop 9♠, Q♠, Q♣
AA______ checks.
Birks checks.
A check is a standard play here with crap, with two pair and with trips. Many weak players will check their sets, but bet out with draws and two pair. Thanks for the tell, guys. Sometimes I will bet out with this hand, against “aware” opponents. But this was Friday night US time. Standard plays to beat them are best.
HoldemStever checks.
scooter_c checks.
mryumyum22 checks.
rubymuc checks.
Whoops, no-one bit. Tends to indicate there is no four-flush out there, no medium pair out there.
Dealing Turn K♣
AA______ checks.
Birks bets [$6].
Some players go for the deep slow-play here, but at this time of night US time, there isn’t much point. You are likely to get called by a King on turn and river, whereas a turn-check-raise might lead the king to walk away. In fact I prefer a turn check-raise with a bad King or a four-straight. In addition, if there is a JT out there for the made straight, it will raise. A non-made hand is very unlikely to raise here, with the sole exception of a KQ that has been slow-played on the flop. Very unlikely.
HoldemStever folds.
scooter_c calls [$6].
mryumyum22 folds.
rubymuc folds.
AA______ calls [$6].
Hmm, two customers In the face of any evidence to the contrary and there being no raise, I’m going to assume two King-x hands here
Dealing River J♠
AA______ checks.
Birks checks.
Even though I don’t think that there is a flush draw out there, many passive players do not bet their draws. But the rogue 10 is the nightmare. I think I’ll check-call.
scooter_c bets [$6].
AA______ folds.
Birks calls [$6].
scooter_c shows K♥, J♦, two pairs, kings and queens.
Birks shows Q♦, 3♦, three of a kind, queens.
Birks wins $46 from the main pot with three of a kind, queens.
So, scooter_c bet for value on the river. Now, let’s think about my play here.
If we can assume that an opponent with a decent King will bet for value, what is my likely gain from a check-raise? Is there, indeed, any gain at all?
Assuming these players will come in on KQ, KJ, KT half the time, KTs, K9s and K8s, Kxs half the time, QJs, QTs, QJ off half the time, we get something like this:
About 12 hands give the guy a winner where he will bet for value, while 24 or thereabouts (I’ve rounded that up a bit to make the maths simple) give the guy a loser. Surely, therefore, I should check-raise?
Well, no, because I still have to get called by a losing hand. How often will the guy call with a losing hand? Well, at this time of night, players find it very hard to lay down for a single bet. During the day, they find it easier. So, assume at this time of night I will get called 75% of the time, but that during the day, during the week, I’ll get called 25% of the time.
We also have the problem of the re-reraise. For simplicity’s sake, we’ll assume that this is not a bluff-three-bet on the river and that I lay my hand down.
FRIDAY NIGHT USA TIME
33% of the time, he has winner and check-raise costs me $6 = minus $2
50% of the time (.67*.75) he has loser and calls: = plus $3
17% of the time he has loser and folds:= no impact.
DAYTIME DURING WEEK:
33% of the time he has winner = minus $2
17% of the time he has loser and calls =plus $1
50% of the time he has loser and folds: = no impact.
From this we can see that the inflection point (where it makes no difference whether you check-raise or check-call) is when your opponent will lay down his hand half the time and call you half the time. But this number is dependent on the number of likely holdings with which your opponent will be beating you compared with the number of likely holdings with which you will be beating him. There is also the problem that Friday night USA time is such that you really have to call the three-bet reraise (because it might be a nutty bluff). However, we can assume that this is zero-sum for simplicity's sake. In fact, it might actually be plus EV.
Conclusion: a check-raise on the river for value is a viable play when the game is loose and wild, but makes little sense in the tight games that you find during the week.
((For the non-poker cognoscenti out there, to “bet for value” on the river is to make a bet with a hand that is neither very good nor very weak, but which you think will make you more money on in the long run by betting than you would by checking. Pot Limit players find this one of the hardest parts of limit to master.)
However, it also set me wondering, after this hand, whether more might be made of this. In particular, if you are getting a significant number of bets for value, might we have reached a tipping point where you should raise for value or even check-raise for value? For a start, it blows up a lot of the “bet for value” mathematics. The usual principle with a bet for value is that your opponent will throw away (or occasionally raise with) garbage, raise with something brilliant, and call with all the hands that might be winning but might not. If your opponent starts raising on the river with hands that might be winning but might not, your bet for value begins to look less attractive.
$3/$6 Hold’em'
Seat 7 is the button
Seat 1: HoldemStever ( $69.49 )
Seat 2: csmith945 ( $0 )
Seat 3: EdwardGrover ( $258 )
Seat 4: scooter_c ( $221 )
Seat 5: mryumyum22 ( $107 )
Seat 6: (Posting) rubymuc ( $150 )
Seat 7: (Button) e8KXmBFk ( $245.50 )
Seat 8: (Small Blind) AA______ ( $256.52 )
Seat 9: (Big Blind) Birks ( $317 )
Seat 10: isley ( $183 )
AA______ posts small blind [$1].
Birks posts big blind [$3].
rubymuc posts big blind [$3].
Dealt to Birks 3♦, Q♦, (those are diamonds, by the way, my browser seems to be showing all the suits but diamonds. weird)
isley folds.
HoldemStever calls [$3].
EdwardGrover folds.
scooter_c calls [$3].
mryumyum22 calls [$3].
rubymuc checks.
e8KXmBFk folds.
AA______ calls [$2].
Birks checks.
Dealing Flop 9♠, Q♠, Q♣
AA______ checks.
Birks checks.
A check is a standard play here with crap, with two pair and with trips. Many weak players will check their sets, but bet out with draws and two pair. Thanks for the tell, guys. Sometimes I will bet out with this hand, against “aware” opponents. But this was Friday night US time. Standard plays to beat them are best.
HoldemStever checks.
scooter_c checks.
mryumyum22 checks.
rubymuc checks.
Whoops, no-one bit. Tends to indicate there is no four-flush out there, no medium pair out there.
Dealing Turn K♣
AA______ checks.
Birks bets [$6].
Some players go for the deep slow-play here, but at this time of night US time, there isn’t much point. You are likely to get called by a King on turn and river, whereas a turn-check-raise might lead the king to walk away. In fact I prefer a turn check-raise with a bad King or a four-straight. In addition, if there is a JT out there for the made straight, it will raise. A non-made hand is very unlikely to raise here, with the sole exception of a KQ that has been slow-played on the flop. Very unlikely.
HoldemStever folds.
scooter_c calls [$6].
mryumyum22 folds.
rubymuc folds.
AA______ calls [$6].
Hmm, two customers In the face of any evidence to the contrary and there being no raise, I’m going to assume two King-x hands here
Dealing River J♠
AA______ checks.
Birks checks.
Even though I don’t think that there is a flush draw out there, many passive players do not bet their draws. But the rogue 10 is the nightmare. I think I’ll check-call.
scooter_c bets [$6].
AA______ folds.
Birks calls [$6].
scooter_c shows K♥, J♦, two pairs, kings and queens.
Birks shows Q♦, 3♦, three of a kind, queens.
Birks wins $46 from the main pot with three of a kind, queens.
So, scooter_c bet for value on the river. Now, let’s think about my play here.
If we can assume that an opponent with a decent King will bet for value, what is my likely gain from a check-raise? Is there, indeed, any gain at all?
Assuming these players will come in on KQ, KJ, KT half the time, KTs, K9s and K8s, Kxs half the time, QJs, QTs, QJ off half the time, we get something like this:
About 12 hands give the guy a winner where he will bet for value, while 24 or thereabouts (I’ve rounded that up a bit to make the maths simple) give the guy a loser. Surely, therefore, I should check-raise?
Well, no, because I still have to get called by a losing hand. How often will the guy call with a losing hand? Well, at this time of night, players find it very hard to lay down for a single bet. During the day, they find it easier. So, assume at this time of night I will get called 75% of the time, but that during the day, during the week, I’ll get called 25% of the time.
We also have the problem of the re-reraise. For simplicity’s sake, we’ll assume that this is not a bluff-three-bet on the river and that I lay my hand down.
FRIDAY NIGHT USA TIME
33% of the time, he has winner and check-raise costs me $6 = minus $2
50% of the time (.67*.75) he has loser and calls: = plus $3
17% of the time he has loser and folds:= no impact.
DAYTIME DURING WEEK:
33% of the time he has winner = minus $2
17% of the time he has loser and calls =plus $1
50% of the time he has loser and folds: = no impact.
From this we can see that the inflection point (where it makes no difference whether you check-raise or check-call) is when your opponent will lay down his hand half the time and call you half the time. But this number is dependent on the number of likely holdings with which your opponent will be beating you compared with the number of likely holdings with which you will be beating him. There is also the problem that Friday night USA time is such that you really have to call the three-bet reraise (because it might be a nutty bluff). However, we can assume that this is zero-sum for simplicity's sake. In fact, it might actually be plus EV.
Conclusion: a check-raise on the river for value is a viable play when the game is loose and wild, but makes little sense in the tight games that you find during the week.
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
Value Bets
Date: 2006-01-21 09:51 pm (UTC)Regarding value bets, I think it is important to know your customer, in either limit or pot limit. You want to value bet more against calling stations, but regarding tricky players whether of the loose or tight variety, following Mike Caro's advice is usually best, i.e. to bet less, call more and seldom value bet after they check since they often would bet a bluff but be waiting to checkraise when they had a monster all along or the river helped them.
In that hand in question, I would go for the checkraise since the 3rd player folded, although *probably* folding to a reraise from a reasonable player. And if in the middle and the first player had bet, then I would just call and go for an overcall by a looser player yet to act, and also so as not to give too much action to that last player if he had the nuts.
I see a situation in plo all the time where a straight comes on the river and the first player bets into more than 1 other player, and the very next player raises. This is pure stupidity as only calling signifies weakness and might just garner an overcall from another looser player behind you who could donate to both players holding the nuts.
BluffTHIS!
Re: Value Bets
Date: 2006-01-22 01:29 am (UTC)I think that I might write an article for Stan James on "When it is right to just call with the nuts". The situation in limit hold'em doesn't crop up often, but it usually occurs when you are in the middle and there is a potentially loose raiser behind you (or a calling station). I recall one hand where I had the nut flush and I suspected the player on my right was on a steal. Therefore he would not call any raise. However, the player on my left might well overcall my call. With the second nuts, he might even raise.
So I called and the player on my left called as well. First player turned over a value bet, as it happened. I showed the nut flush and the player on my left went into one, calling me a complete moron for just calling when I had the nuts. I kept my counsel, but it really did make my day.
PJ
Bet for value on the river
Date: 2006-01-22 05:58 pm (UTC)(Sorry, this is the image that comes to mind)
-- Jonathan
Re: Bet for value on the river
Date: 2006-01-22 06:12 pm (UTC)Pete