KK flops a ... good draw

[Nate tha' Great]

Everyone at the table is too loose before the flop, and generally too passive afterward except for the Button (but he's not involved in this one). MP plays weirdly and kind of badly after the flop.

Party Poker 10/20 Hold'em (6 max, 6 handed) converter (http://www.selachian.com/tools/bisonconverter/hhconverter.cgi)

Preflop: Nate is CO with K/images/graemlins/club.gif, K/images/graemlins/heart.gif.
UTG folds, MP calls, <font color="CC3333">Nate raises</font>, Button folds, SB calls, BB calls, MP calls.

Flop: (8 SB) 7/images/graemlins/club.gif, A/images/graemlins/club.gif, J/images/graemlins/club.gif <font color="blue">(4 players) </font>
SB checks, BB checks, <font color="CC3333">MP bets</font>, Nate calls, SB folds, BB calls.

Turn: (5.50 BB) 7/images/graemlins/diamond.gif <font color="blue">(3 players) </font>
BB checks, <font color="CC3333">MP bets</font>, <font color="CC3333">Nate raises</font>, BB folds, MP calls.

River: (9.50 BB) 3/images/graemlins/club.gif <font color="blue">(2 players) </font>
MP checks, <font color="CC3333">Nate bets</font>, MP calls.

Final Pot: 11.50 BB
<font color="#990066">Main Pot: 11.50 BB, between MP and Nate.</font>

I planned to check behind on the river without improvement.

[stripsqueez]

the issue is raise the flop or raise the turn - so why the turn ?

raising the flop can become a bit predictable - sometimes raising the flop is almost expected resulting in getting 3 bet a lot by either boring players with good hands or good players who see you do it a lot in this position - if the aim is to lose players then lots of players need the extra intimidation of the turn raise

i cant see an obvious technical reason for choosing the turn on this hand

stripsqueez - chickenhawk

[chezlaw]

I had a similar hand today and I just called on the turn, I was wondering what the reason is for raising.

The raise seems likely to get the BB to call or raise if he has you beat and to fold when you want him to call with few outs - it would be a shame if you got him to fold a club.

Did you think you could get them to lay down an ace or am I missing something completely?

[manpower]

I believe he's raising for the free showdown. At the start of the turn if he just calls down to the river, he will see the showdown for 2 bets when he doesn't improve as the bettor will just keep betting into him, if he does improve on the river, the bettor may very well check to him and then call the hero's bet. Netting the hero 2 bets. However, if he raises the turn he will likely get a free showdown when he doesn't improve (still costing 2 bets) and will be able to net 3 bets when he does improve.

Of course this assumes a few things about the opponent, and there's certainly a case to be made for trying to keep the 3rd player in.

[Ulysses]

I think calling the turn is better in a game where people are pretty liberal about 3-betting the turn.

[Ulysses]

That's all fine and good except for the fact that you're not taking into account the high likelihood of getting 3-bet on the turn.

[manpower]

very true, just trying to explain the mentality to chezlaw. I can't say much about the 10/20 game, as I don't play it, but moves like this work quite well at low limits against the right players.

[King Yao]

[ QUOTE ]
That's all fine and good except for the fact that you're not taking into account the high likelihood of getting 3-bet on the turn.

[/ QUOTE ]


I think thats a great point, I did some expected value analysis.

Assumption: Nate has 10 outs if he is behind. He has 9 club outs and half the time a K will be an out, the other half the time, it won't be (a guess of course), so average it out to 10 outs. Nate has 10/46 chance of catching his hand and a 36/46 chance of not catching. When he does not catch, he has a 90% chance of losing the hand, and a 10% chance the other player is bluffing. I will assume there is zero chance the opponent will fold on the Turn. There is 5.5 Big Bets going into the Turn

Scenario 1: Nate just calls the Turn. Nate calls the River if no club or K shows up and has a 10% chance of winning. If a club or K shows up, on average, Nate wins 1 bet on the river when a club or K shows up (the King could be a loser if the opponent already has a flush of course, but lets leave that alone for now), once in a while he’ll win 2 bets as he will get bet into, but once in a while he’ll win no bets as the other guy will fold with 4 clubs on the board.

EV no club/King = 10% x 36/46 x 7.5 big bets + 90% x 36/46 x -2 big bets = -0.82
EV club or King = 10/46 x 7.5 big bets = 1.63
Total EV = +0.81

Scenario 2: Nate raises the turn and gets what he wants, a possible free showdown if he misses. The opponent calls the Turn and checks the River, and calls if Nate bets (I’ll assume he won’t bet out or checkraise even if he has Qc when a club hits the River).

EV no club/King = 10% x 36/46 x 7.5 big bets + 90% x 36/46 x -2 big bets = -0.82
EV club or King = 10/46 x 8.5 big bets = 1.85
Total EV = +1.03 (if the turn raise for a free showdown works, the EV is higher than just calling)

Scenario 3: Nate raises the turn, but his opponent 3 bets him, at which point Nate calls the turn, and calls the river if he misses, but gets 1.5 expected bets on the river if he hits (assuming half the time his opponent will bet out and call a raise from Nate, the other half of the time, he will check and call a bet from Nate).

EV no club/King = 10% x 36/46 x 9.5 big bets + 90% x 36/46 x -4 big bets = -2.07
EV club or King = 10/46 x 10 big bets = 2.17
Total EV = +.10

So if he does get three bet, the EV goes way down (there’s the complication that a three bet by this player may allow Nate to make a good fold on the river, which I do not adjust for her). Using these three EV numbers, we can figure out how often a three bet by the opponent will have to happen in order for Nate’s Turn raise to be correct or incorrect, or indifferent.

Essentially we want to solve for Z in this formula:
+0.81 (the EV of just calling) = 1.03 (the expectancy that a free showdown happens) x Z + .10 (the expectancy that a free showdown does not happen) x (1-Z)

Z in this case equals 76.3%, which means that Nate’s Turn raise would have to work more than 76.3% of the time in order for it to have positive value over just calling. In online poker, I would guess this is a bit too high, but not as high as I would have originally thought without this analysis. The numbers will be different if you tweak the assumptions a bit, but I think it shows that this play needs to work a very high percentage of the time in order for it to have value. If there is a chance that your opponent will fold when you raise on the Turn, it greatly reduces the percentage of time this play needs to work for it to have value, but in this hand, the odds of him folding to a turn raise seems fairly low.

Overall, I don’t see a problem with Nate’s play. I think its pretty close, probably indifferent between raising on the Turn and just calling.

[Ulysses]

[ QUOTE ]
very true, just trying to explain the mentality to chezlaw. I can't say much about the 10/20 game, as I don't play it, but moves like this work quite well at low limits against the right players.

[/ QUOTE ]

Yeah, your reasoning is exactly correct in a less aggro game where it is much less likely for you to get 3-bet w/ a hand like one-pair. The big difference between the 10/20 and all the other Party 6-max games it the level of aggression. A decent Ace will very often 3-bet here, which is very different than even the 5/10.

[Your Mom]

U is right. An Ace without a club will nearly never 3 bet in the Party 5/10. The 5/10 is full of bad players. I think this is a direct result of Party having no 6 max game between 1/2 and 5/10.

[James282]

Hey Yao, your analysis is very flawed if you think all the club outs are always good but only 1/2 the kings are good? I would much, much rather have a king on the river than a club - the board is paired, after all.

I very rarely raise the turn in spots like this for a "free" showdown as well - but will do so against players who are too wuss to three-bet without a flush or a 7 here.
-James

[King Yao]

James, your right, in my rush to do the analysis, I missed the fact the board is paired. This would mean Nate has fewer outs which means all the expected values of each scenario is lower, which means putting in more money is not advisable. My analysis is better suited for a situation when the Turn does not pair the board as it did in this spot. My apologies, but I hope I got my point across in what to think about when it concerns free showdown raises.

By the way, just an FYI, my first name is King, its not a handle or a nickname, its my real name. Please feel free to call me that if you wish, Yao is my last name. Thank you.

[Nate tha' Great]

1. There are two players to act behind me on the flop. In these games, I'd usually going to get an overcall from someone who held something like the J /images/graemlins/club.gif or the Q /images/graemlins/club.gif and was drawing virtually dead against me if it's only one bet to him.

2. There is one player left to act behind me on the turn. Suppose that he holds a weak ace, with no club draw. Suppose that MP is betting with a draw, or something a pair of jacks with a club redraw. If there's even a small chance that I could fold out the ace by representing some kind of big hand and forcing him to call two big bets cold on the turn, that would be a huge plus for raising.

3. Alternatively, if I induced him to fold something like the Q /images/graemlins/club.gif from which I would have gotten an overcall, that would be a minus.

4. If an opponent is aggressive enough to 3-bet the turn (and, by the way, I don't think this opponent was), he'll frequently be aggressive enough to bet into me again on the river, even if a club or a K hits. If that's the case, then I'll often make an extra bet from him on the river when my draw hits.

[Nate tha' Great]

[ QUOTE ]
The raise seems likely to get the BB to call or raise if he has you beat and to fold when you want him to call with few outs - it would be a shame if you got him to fold a club.

[/ QUOTE ]

The guy who bet into me wasn't likely to fold, unless he was on a pure bluff.

[ QUOTE ]

Did you think you could get them to lay down an ace or am I missing something completely?

[/ QUOTE ]

I thought I could maybe get the overcaller to lay down an ace.

[stripsqueez]

[ QUOTE ]
Nate’s Turn raise would have to work more than 76.3% of the time in order for it to have positive value over just calling. In online poker, I would guess this is a bit too high, but not as high as I would have originally thought without this analysis

[/ QUOTE ]

my gut says its less than 24% of the time you face a 3 bet in this game - my gut also says its very rare the turn raise will get an A to lay down even for 2 BB's, although even if this figure is say 5% it has a significant effect on the value of the play - i think its right to raise either the flop or turn nearly all the time - to do less is to not invest appropriately in your position

stripsqueez - chickenhawk

[chezlaw]

I'm still confused here. I'm ok with the turn raise if its head up, but bearing in mind the board has paired and the BB has few outs my feeling was that the advantage gained by raising the turn was more or less balanced by the disadvantage of getting the BB to fold a hand you want him to call with.

If my feeling was correct then any risk of being 3 bet would make calling the better play. I've done some analysis to see how good the raise is if you completely discount the three bet possibility and it comes out as slightly worse than calling.

This backs up my feeling (what else is analysis for /images/graemlins/smile.gif) but that may be because my assumptions are unreasonable or I've made a mistake.

Assume if you raise the turn:
no-one 3 bets even with a very good hand
no-one bets the river (except you, if you get a club or a king)
BB folds to your raise 90% of the time and always calls the river is still around.

and if you call the turn:
MP bets the river
BB calls the turn 50% of the time and calls the river 50% of the time.


For simplicity assume that you can never get a better hand to fold, if you river a king or a club then your hand is good and if BB calls your turn raise then you have to improve to win. (hopefully these compensate each other)

I'm also assuming that if you call, or your raise gets the BB to fold, then you win unimproved 20% of the time.

As you never fold and never get a better hand to fold, the existing pot can be ignored in the calculations.

You have 11 outs, so will make your hand 11/46 of the time

If you call:
You win 20% * 35/46 (your hand wins unimproved) + 100% * 11/46 (you make your hand). So you win about 40% of the time.

MP always puts in 2 bets. BB puts in 1 bet 50% of the time and 2 bets 25% of the time. You always put in 2 bets so:

60% of the time you lose 2 bets
10% of the time you win 4 bets (BB calls all the way)
10% of the time you win 3 bets (BB folds on the river)
20 % of the time you win 2 bets (BB folds on the turn)

that works out as costing you 0.1 of a bet.

If you raise:

60% of the time you lose 2 bets (as before)
90% * 35/46 * 20% of the time you win 2 bets (BB folds and you win unimproved)
90% * 11/46 of the time you win 3 bets (BB folds and you improve)
10% * 11/46 of the time you win 6 bets (BB calls and you improve)

that works out as costing you 0.14 of a bet

Have I made a mistake or are the assumptions out enough to turn raising into the better option?

[King Yao]

A couple of comments:

1. If we go with your assumptions, I'm not sure why you just estimate the chances of you winning when you just call. You estimated it to be 40%, but if we do the math that you have set out, its only 39.1% Using 39.1%, I get an expectancy of losing 0.14 when you do call, which is the same (under your assumptions) as when you do raise. So with your assumptions, you should be indifferent between raising and calling.

2. If you have a lower chance of winning unimproved, you'll see raising is better....if you lower your 20% of winning unimproved down to 10%, I get losing -.27 of a bet if you raise, but losing -.50 of a bet if you call (for others who are reading this wondering why we even want to continue the hand with negative expectancy, the reason is because we have not included the rest of the pot into these calculations. if we did, the numbers would turn positive)

3. but to counteract the benefit of raising, if the original bettor reraises you, he takes away that advantage of raising.

4. all of these numbers all depend on the assumptions. that's why hold'em is not just a game of math, the ability to pinpoint the percentages, i.e., to read other players hands and ability to call/raise/fold is crucial. your assumptions could be right if you have a great read on your opponents, but if you don't read them well, your assumptions will be off, and the calculations will give a bad answer.

[King Yao]

[ QUOTE ]
......assumptions will be off, and the calculations will give a bad answer.


[/ QUOTE ]

actually, an example of that would be my original expected value post in response to ulysses, which James correctly pointed out the assumptiosn were off because of the pair on board.

[chezlaw]

Hi King

I went with 40% because I could do the next bit in my head /images/graemlins/grin.gif. Can I wave my hands a bit and claim that strict accuracy is a bit over the top as I'm just testing whether there is a significant advantage either way and the assumptions aren't very precise.

20% to win unimproved seemed ok to me given that MP played 'wierdly and kinda badly'. Maybe 10% is a better guess. I think the other assumptions favour the case for raising - I have assumed that when you hit you never lose and get called by MP and sometimes BB.

We can play with the assumptions a lot, but it seems a pretty close call between raising and calling even if we could guarantee no 3 bet and a river check from MP.

[King Yao]

[ QUOTE ]
Can I wave my hands a bit and claim that strict accuracy is a bit over the top as I'm just testing whether there is a significant advantage either way and the assumptions aren't very precise.


[/ QUOTE ]

sure! i have no problem with that at all. I think this was a good exercise, the numbers change when our assumptions change, which makes it much more interesting.