Calling station tendency

[PeteX]

I'm trying to make a "calling-station tendency"-stat on the flop, where I wish to calculate a percentage that reflects the times when the player was facing a bet/raise and just calls on the flop. He doesn't fold or get to see a free turn; he must only call (once or more) and not bet+call and not raise+call or any such combination.

Would these conditions work as numerator:
not flg_f_fold and cnt_f_call>0 and not flg_f_bet and cnt_f_raise=0

and then amt_f_bet_facing>0 as denominator as this would imply that the player saw the flop (hence I think flg_f_saw is not required here) and he was faced with a bet (I guess this would also cover the situation where someone bets, someone else raises and then it is the players turn, I guess he is still facing a bet eventhough there was already a raise after the bet) and was thus given a choise whether to reraise (if not capped), fold or call.

Hence, if the above calculation gives e.g. 50%, the player will, when faced with a choise of calling or folding on a flop, call 50% of the times and the remaining 50% is divided between folding, checking (and getting a free turn), betting (+possibly calling), raising (+possibly calling). Obviously we don't know what the player would have done if the pot was capped before his turn and he calls; he might have wanted to raise, but I think those situations are rare enough so they won't mess this figure up that much.

EDIT: A short summary: I want to know the % how often the player goes from flop to turn without showing any agression and not being able just to see a free turn. That's the point.... ;)

[kraada]

Would you want the following to be included:

Multiway pot, three players: A, B and C. C is on the button, A acts first.

A bets, B calls, C raises, A calls, B folds.

Does B's "calling station tendency" go up or is it unaffected?

[PeteX]

Would you want the following to be included:

Multiway pot, three players: A, B and C. C is on the button, A acts first.

A bets, B calls, C raises, A calls, B folds.

Does B's "calling station tendency" go up or is it unaffected?

It should go down basically as there is 1 more case for the denominator where the player is on the flop and facing a decision whether to be "weak-tight" or a "calling-station", but he does fold and hence, does fall more to the "weak-tight"-side and the numerator doesn't increase.

This stat should reflect the "tenacity" of the player to stick in the pot when having to put money into to pot, but shows no agression.

The whole point of this stat is to say to me: Should I try to bluff this guy with any two cards or not? Will he be folding often (low value) or calling often (high value), when facing a need to put money into the pot.

[kraada]

The whole point of this stat is to say to me: Should I try to bluff this guy with any two cards or not? Will he be folding often (low value) or calling often (high value), when facing a need to put money into the pot.

You have to be very careful with this sort of thing; If you have a person who has a VP$IP of 5% then they are going to be much more sticky, but it's not going to be because they're a calling station it's going to be because they just have better hands when they put money in the pot.

Personally I display a lot of data on my HUD and I think seeing all of it helps me make a better decision; I'm honestly not sure that you can build just one stat to answer your basic question. It's too complex an issue to be boiled down to one stat. You can find my HUD in this thread.

I think you might be happy enough with the Fold % stats, just don't forget to take into account board texture, and the person's range given their preflop play.

[PeteX]

The whole point of this stat is to say to me: Should I try to bluff this guy with any two cards or not? Will he be folding often (low value) or calling often (high value), when facing a need to put money into the pot.

You have to be very careful with this sort of thing; If you have a person who has a VP$IP of 5% then they are going to be much more sticky, but it's not going to be because they're a calling station it's going to be because they just have better hands when they put money in the pot.

Personally I display a lot of data on my HUD and I think seeing all of it helps me make a better decision; I'm honestly not sure that you can build just one stat to answer your basic question. It's too complex an issue to be boiled down to one stat. You can find my HUD in this thread.

I think you might be happy enough with the Fold % stats, just don't forget to take into account board texture, and the person's range given their preflop play.

Yes, VPIP% is definitely on my HUD, but I'm trying to tone down the amount of stats on the hud as I used to have a ton of numbers (like 16 or so) visible and I just don't think I can take advantage of them all, so I'm trying to make a bit more "powerful" stats that incorporate more info into the one figure.

Fold% is actually the stat I was thinking before and it may be that I will still use it, I'll have to think about it a bit.

[kraada]

Just be aware that Fold % is out of total actions.

So if on a given street a player's actions were check then call then fold (which can happen in multiway pots), he'd have a 1/3 (or 33%) fold% for that instance.

[PeteX]

Just be aware that Fold % is out of total actions.

So if on a given street a player's actions were check then call then fold (which can happen in multiway pots), he'd have a 1/3 (or 33%) fold% for that instance.

Ok, that doesn't quite sound what I want. I would expect the low Fold % to reflect the tendency of a player to get out of the hand when faced with some action, i.e. more bluffable.

Basically I guess I'd then have to define a new "Fold2 %"-stat that would basically take the ratio between a street seen and folded on that street. I.e. if a player NEVER folds on the flop then he'd have 0% on and if he ALWAYS folds, the value would be 100%. That way it would be easy to add e.g. colorings: Red like <35% (i.e. NEVER bluff this guy), yellow 35% < and < 75% and green >75% (always bluff this guy).

[kraada]

What's wrong with the Fold Flop stat then?

That's calculated as: (# of times someone folds the flop / # of times someone has seen the flop) * 100.

There's also Fold Turn and Fold River. You could easily make a custom stat that is "Overall Fold" or some such if you wanted, though honestly I do think people vary by street so much that it might cause problems.

[PeteX]

What's wrong with the Fold Flop stat then?

That's calculated as: (# of times someone folds the flop / # of times someone has seen the flop) * 100.

There's also Fold Turn and Fold River. You could easily make a custom stat that is "Overall Fold" or some such if you wanted, though honestly I do think people vary by street so much that it might cause problems.

:) Nothing actually, I didn't think of that one. Yep, I'll think that one is the one I'll be using.

Yeah, I was thinking about making an overall Fold% by using weighted averaging like this:
(FlopFold%*100 + TurnFold%*10 + RiverFold%) / 111

So, this way I could put lot more emphasis on the folding the earlier it happens making me more confident to win the pot as early as possible.

But I guess there are enough players who peel the flop light and fold on turn without improving and then there are those who call 3 streets and then fold on the river as the eagerly chased gutshot doesn't materialize. And for those I guess you'd need to really look at the separate values for 3 streets to see the individual tendencies.

[PeteX]

Ok, just got an idea that might wreck the usage of Fold Flop to my purposes.

What about this case: Player sees the flop 100 times and nobody ever bets and the players gets a free turn and then possibly folds or whatnot. The value of Flop Fold would be then 0%, wouldn't it? This would indicate that this player just can't be bluffed off the pot at the flop, which doesn't quite feel right...

So, wouldn't the checks mess up the usage of this stat for the real "fold when facing a bet"-tendency?

Should it not be calculated like:
(# of times somenone folds the flop AND faced a bet on the flop / # of times someone has seen the flop and faces a bet on the flop) * 100

or more simply (as facing a bet on the flop implies in itself that he sees the flop)

(# of times somenone folds the flop when facing a bet / # of times someone faces a bet on the flop) * 100

and wouldn't this also work for this purpose:

(# of times somenone folds the flop when facing a bet / # of times someone saw the turn) * 100

Wouldn't that be a more accurate indication whether the player is more likely to fold to a bet on the flop or call a bet on the flop?