mpethypoker.com

by mpethybridge 08.24.2010

River Call Efficiency is a Goldilocks stat.

What I mean by that is that it is a leak for your RCE to be either too high or too low.

Think about the situation when you are on the river facing a bet, either because you have checked out of position or you are in position.

When do you want to call?

Answer:

1. When you are ahead of his range to value bet.

2. When you are behind his range to call a raise.

3. When you are bluff catching.

4. When you are offered attractive pot odds; you know you will lose most of the time, but if you win X% of the time, you break even or profit based on his bet sizing.

Example:

you raise first in on the button with KQs, and get 3 bet from the big blind from a 5% 3 bettor. You flat. the flop comes down Q 7 5 rainbow. he bets, you call.

The turn is a 2 of spades, putting two to a spade flush on the board, he leads again and you call. There are 80bb in the pot at this point, more or less.

The river is the T of diamonds. The villain shoves 60 into 80bb pot. You are thinking about calling. Lets say the board is now Qh 7s 5c 2s Th.

You suspect his range to 3 bet is a polarized range weighted toward value hands. So we have him on the top 3.5% of hands plus 1.5% semi-bluffing hands, some of which could have hit the board.

So his range to bet/bet/bet is his value range of AA, KK, QQ and a few combos of AQ to allow for the possibility that he will ship stacks with that hand. I put in 3 combos of suited AQ. His total value range to bet/bet/bet is 1.6% of combos, or roughly 1/3 of his 3 betting range.

Now you have to allow for the possibility that he is bluffing all three streets. he MIGHT do this if, for example, he 3 bet you light with 8s6s and picked up a straight draw on the flop, added flush outs on the turn, but bricked the river. Or something like that. So lets give him a light 3 betting range of hands that hit the flop, picked up equity on the turn, but didn’t get there on the river. Lets say he does this with all 2 86s combos, to allow for the possibility that he was just double barreling with his straight draw on the turn. then lets say that he bets all of his A3s and A4s combos because he picked up the gutshot on the turn. Then lets add in some triple barrels with AK; we’ll say he does this with 3 combos of AK, representing AK of spades and two “dammit, this is my pot,” combos.

So he has a value range of AA, KK, QQ, 3 combos of AQ.

He has a bluff shoving range of A3s, A4s, 86ss, 86hh and 3 combos of AK.

If you stove this, you see that we have 30% equity–we beat 30% of his shoving range.

The pot is 140bb with his shove, and it is 60bb to us to call. The pot is offering 2.3 to 1, so this spot is very, very close. A combo or two in or out of his bluff shoving range is going to swing it from a call to a fold or vice versa–the rake actually makes a difference here, it is so close. I’d probably fold, but you could actually decide that your equity plus the “note equity” you get by bluff catching him (note equity = knowing he is aggro enough to bluff shove rivers and using that info against him later) makes it a profitable call.

So say we bluff catch him here and win. Our River Call Efficiency on this hand is 2.3–RCE answers the question: “how many dollars do I win for every dollar I put in the pot as a call on the river?” Phrased differently, then, RCE helps measure the accuracy of our ability to make accurate range assignments on villains on the river and to make correct pot odds based calculations on whether to call river bets. But if we played this hand twice against this villain, called once and won, called once and lost, our RCE would be 1.15 (I think a loss gets calculated as $0 won, not -60 here–but I am not sure).

All of this is preparatory to saying that RCE can be either too high or too low. If you have a high RCE, it indicates that you are accepting situations (calling rivers) where you are the clear favorite and passing on marginal spots.

I don’t really want to give a hard and fast number about what I think RCE should be. there is a very interesting post on 2+2 where a guy argues that to maximize the value of your river calling, your RCE should be 1.0. This sounds like a game theory thought process to me, and I would say that this is correct if we are randomizing our calls.

But we should not be randomizing our calls, we should be looking for the spots where we will show a profit from calling by identifying spots where we beat more of his range than the pot odds are offering. So my opinion is that RCE should be well above 1.0.

I see a lot of solid grinders who have a RCE in the range of 1.75 to 2.1. I think this is a fine range to be in. The best grinder I ever did a db analysis had a RCE of 2.2 or 2.3–I forget which now. So I’m not prepared to say that an RCE of 2.3 is too high, especially at mid stakes, where the players are good enough to value bet hands that will win a check/call, thereby removing some marginal calls from the RCE calculation by turning them into river bets.

If you’re in the 1.75 to 2.1 range, I think you are in pretty good shape, and you don’t have much to worry about.

I’ll be writing more on RCE, because it is sort of the Alaska of Hold Em Manager–one of the last unexplored frontiers. Using this stat to analyze river play fascinates me. In fact, it is about all I thought about through either Texas or New Mexico, I forget which.