The true value of tells?

In Poker Essays II Mason writes:


"Tells, in my opinion have a value of approximately $2 per hour to a highly skilled $20-$40 hold'em player who is able to make slightly more than $40 per hour. Some of you might disagree with this. And I do concede that there are probabaly a few players who are able to win this much who might actually make a little more from tells -- perhaps as much as $5 per hour."


I would guess that a "highly skilled 20-40 hold'em player" is what many of us here aspire to. So let's use the $2 per hour mark for this discussion.


This is the equivalent of 1/10 of a small bet per hour for an expert player.


I am finding that as I play and gain experience, tells are probabaly closer to 1 SB per hour. This is a significant difference. I play mostly 6-12 and am gradually moving up with occasional forays into 10-20.


I am assuming that the difference lies in the fact that the value of tells decreases as the limits increase. Players play better, and are most certainly more aware and self conscious. In fact, I would estimate that in good 3-6 and 4-8 games tells are worth nearly .75 to 1 BB per hour. These are tells that actually impact how I play the game. Checking rather than betting, raising rather than folding, folding rather than calling. These tells can not only save me BETS, but they can win me POTS which I would not have otherwise won.


Tommy Angelo (who is a wonderful contributer here) had a recent post in the medium stakes forum regarding a hand where a second 4 hit the board on the turn and both of his opponents froze up momentarily. Acting on this tell could have saved him one BB in this instance (in addition to giving himself a free card at a potential win). Am I to understand that he would not be able to earn or save an additional BB for 20 more hours of gameplay (on average)?


Clearly each tell does NOT have the value Caro attibutes in Caro's Book of Tells. In my admittedly limited experience however, they are worth substantially more than 1/10th SB per hour.


Is my personal observation skewed? Does the value decrease as the limits increase? What am I missing?


Note I am talking strictly about tells, and not about "reading/knowing your opponent" and "what level are you thinking on". This is strictly seeing a physical tell, and using that information by acting differently than you otherwise would have.


Sorry for the length, and thanks for your thoughts.


Clark

Doesn't Mason go on to explain why the tells are not worth as much as they seem, or is that in another essay? Essentially, if you have good hand reading skills, tells become less important. Also, most decisions in poker don't actually cost/make as much as you think. That is, your decision to fold on the flop in a certain situation, which you might claim saved you a small bet, probably only saved you a fraction of a bet, since you would have had some small chance of drawing out (think of when a tell makes you realize you are outkicked, for instance - if you had called you did have 3 outs into a medium sized pot, so you didn't actually gain a bet of expectation by folding). Therefore, even though it may seem like tells are helping you a lot, they really only impact your game slightly.


On the flip side, if I could add $2-5 to my hourly rate I would be quite happy, so don't think that tells are inconsequential. Just not as important as other things.

The value of a tell is your EV when you see the tell AND change your decision vis-a-vis your EV when you don't see it or don't change your decision: If you are going to call anyway seeing a "weak" tell doesn't do you any good.


But, I believe the actual value is between the 2+2 authors and Caro.


- Louie

I don't think the word "tell" is sufficiently definable or distinguishable to be assigned a number. When facing a subtle physical tell, I think we instinctively factor in the degree of reliability of the tell, and the pot/cards/situation, and, and this is big, the betting, to come up with an answer. On this topic we make the needed distinction between "reading hands" based on betting and trends and such, and physical tells. I don't think that line can be precisely drawn as to worth because both things come into play, with ever-shifting weighting, during the moment of truth.


I would think that the value of physical tells would depend largely on the kind of games we play in because there is much variance in the number and degree of tells.


Then there's the rake, in there drastically skewing the percentages. If I make, say, 40K at $20-40 over some stretch of time, and someone put a gun to my head and demanded to know how much of it I thought came from tells, I'd say 10K. But that is NOT 1/4. It's more like 1/8 if I payed $40,000 in collection and tokes during that time frame, which is typical.


Netting 1BB/hour often means playing 2BB per hour better than the opponents. This feature throws a wrench in most earnings/hour discussions, IMO.


Tommy

Just to set the record straight here's the essay that appears in POKER ESSAYS, VOLUME II. Also, John Feeney addresses this same subject in his book INSIDE THE POKER MIND.


All comments are welcome.


The Value of Tells and Positive Reinforcement


Tells, in my opinion, have a value of approximately $2 per hour to a highly skilled $20-$40 hold'em player who is able to make slightly more than $40 per hour. Some of you might disagree with this. And I do concede that there are probably a few players who are able to win this much who might actually make a little more from tells — perhaps as much as $5 per hour.


Part of the reason for this slightly higher figure is that tells can make up for some deficiencies in your game. For example, a top card reader who has a good knowledge of how his opponent plays will often come to the same conclusion that the tell gives you. When this is the case, the tell does not have much value. But if you don't "read" that well, tells can have more value. Thus, some top players may do a little better from their interpretation of tells than other top players.


Anyway, after I wrote that I felt tells only had a value of $2 per hand — which is certainly a significant amount to a top $20-$40 limit hold'em player — another poker writer, who has done some excellent work in this area, replied that they were easily worth two big bets an hour, which at the $20-$40 limit amounts to $80 an hour. I strongly disagree with this claim.


First, if this was the case, there would be a few $20-$40 limit players who would now win at a long-term rate of approximately $120 per hour. This comes to almost $5,000 per week which is approximately $250,000 per year. I know of no one and have never heard of anyone who plays this limit that comes close to this figure. (In fact, most of the best players at the larger limits such as $75-$150 [seven-card stud] probably don't do this well although a few of them do come close.)


In my book Poker Essays I estimate that a top $20-$40 hold'em player can make $50 per hour, and there are only a very small number of people capable of doing this well. However, outside of Las Vegas a great player might do a little better than $50 per hour.


The second reason why this claim that tells are worth two big bets an hour does not make any sense to me is that if it was possible for a few players to do better than $100 per hour at this limit, it would mean that some of the bad players would never have a winning night. And, if that was the case, I suspect that almost all of them would quit playing: As I've said many times, if all the bad players quit playing. there wouldn't be any games.


Players need positive reinforcement. For a poker game to thrive, there needs to be a large enough short-term element of luck present for a poor player to have a reasonable chance to leave a winner after a short period of time. In fact, this idea is true for virtually all forms of gambling. For example, consider blackjack or craps. These are both highly fluctuating games that often allow terrible players to leave with a nice win. (In fact, you can argue that anyone who plays craps must be a terrible player because no matter how you play, you cannot overcome the house advantage.)


Thus, if it was possible to beat a game like $20-$40 hold'em for $120 an hour, there would quickly be no bad players, which means there would be no $20-$40 hold'em games. Actually, this sort of thing has already happened. It's very hard to find a no-limit hold'em game.


And why are no-limit hold'em games hardly ever spread anymore? Simply because the good players had such high win rates against the bad players that the bad players got virtually no positive reinforcement since they just about never won. Thus they quit playing. Again, this translates to no more games. At least no more games on a regular basis. (Indeed, a good no-limit hold'em player in a game with a $10 blind and a $20 blind can easily have a win rate of $120 per hour if there are some bad players present. But how long do you think the bad players last without ever winning, that is, positive reinforcement?)


So, yes, the ability to read tells is a very valuable asset to a top poker player. In fact, I believe it is one of the many skills that any new player should strive to master. But notice that I said one of many skills. This means that putting too much emphasis on this area of play, especially if it causes you to neglect some of the other very important skills that are required to become a top poker player, is a mistake.


If you want to become a professional, you have to master many skills. Not just one.

Betting folding and raising are not actually tells.

When you combine the action with the cards its part of your card reading skills.

Yes making correct decisions based on tells is exactly where their value lies.


Apologies to Clarkmeister

Clark


I would have to say that I think tells are in fact NOT worth much more than Mason states, and here's why.


First off, you have to distinguish exactly what a "tell" is, and what it is not. A tell is an action, gesture, mannerism, or some type of behavioral cue that tells you the strength of an opponent's hand. For instance, the classic "shaking hand tell". If someone who has otherwise been very calm during the game all of a sudden puts his chips into the pot with a shaky hand, it is very likely that he has a big hand. For instance, once, in Atlantic City, I was playing Hold'em and flopped middle set. On the river, I made a full house. It wasn't the nut full house, but my hand was strong. However, there was a bet, then one of my opponents all of a sudden came out raising, and when he did, his hand was visibly shaking when he put his chips into the pot. There was a lot of action on the pot, and another called his raise before it got to me. Most people could NEVER lay this hand down in this spot, and admittedly, I almost called. But I KNEW he had quads, so I folded. It was a very clear-cut case of my opponent giving me a TELL that gave away the strength of his hand. I have probably never made such a fold again. Other less obvious tells are discussed in Caro's book. I believe they all have value, but I would seriously question the monetary value assigned to them by Caro.


Tells must be distinguished from "telegraphs". A telegraph occurs when someone gets their chips ready to call or raise, or indicates they intend to fold, before it is their turn to act. A telegraph only occurs when this person RELIABLY indicates what they are going to do. I intentionally do things to ensure that people do not know what I am going to do, but to the unobservant, I may appear to be telegraphing my intentions some of the time. There is a player in my local cardroom whom I would estimate is 80% reliable on his telegraphs. If he counts out the chips ahead of time, he will usually play them just like he counted them. Since he does this virtually all the time, this provides me with an edge. Although he is an aggressive player, I don't mind him being on my left, in fact, I often purposely position myself accordingly. It provides me with a tremendous advantage to know what an aggressive player intends to do. IMO, telegraphs are much more useful than tells, because lots of people telegraph, and many of them are pretty darn reliable about it. Tells are very specific to individuals, and relatively few can be generically attributed to everyone.


As far as how lucrative tells are, I look at it like this. The specific circumstances where you successfully use a tell determines its value. Say I am last to act on the river (small to medium pot) and the SB bets out, everyone folds to me. He has some sort of mannerism that I am attributing to the strength of his hand. This could work two ways. First, say I have a weak hand, a hand that can only beat a bluff, but will lose to a legitimate hand. If I think the SB's tell indicates that he is bluffing, I would call. If I am right, and I win the pot, then that particular tell was worth the amount of money that was in the pot. On the other hand, say his tell indicates that he is strong. Now, I would fold my marginal hand. In this case, the tell is worth one bet. His tell gave away the strength of his hand, and I saved a bet by correctly reading him.


Another consideration is the reliability of an opponent's tell. Perhaps an opponent bets very forcefully when he is bluffing most of the time. BUT – about 10% of the time that he bets forcefully, he actually has a strong hand. This takes away from the value of the tell. This principle is discussed in Caro's book, and has merit. Precisely quantifying this might not be easy though. The best way to learn what your opponent's tells are, and how accurate they are, is to pay attention to the game while you are playing.


The final consideration I am going to mention has to do with how often you actually USE an opponent's tell to make a decision (or change an otherwise routine decision). In my experience, I really can't say I make very many brilliant strategy decisions based on tells very often. I would not estimate my advantage due to my picking up tells is any higher than Mason's estimates, but most likely it is lower. Most people I play against are so clueless that they are unlikely to do anything consistently enough to give off any accurate cues as to their intentions. In fact, some of them often don't even know what they HAVE. These people are completely unreadable, and they give off few reliable tells. Now if I factor in telegraphs, I probably exceed Mason's estimate of 1/10 a SB/hour as far as my earnings go. However, I doubt I exceed this by much.


Remember, a tell might be worth a pot today, but if you don't pick up another one for two weeks, the AVERAGE value goes way down. Tells are only profitable when you actually have the opportunity to use them. They are only profitable when they are reasonably accurate, you interpret them correctly, and make the best play based on your interpretation. And finally, they are only worth the DIFFERENCE between the money you would have won or lost had you played your hand without the benefit of the tell.


Dave in Cali

You make good points dave. It is also worthwhile to point out that tells that make you fold can easily cost money even in the long run if you fold winning hands, or if they force you to back off a hand and play it passively by missing a raise.


A good example of this is in Ciaffone's book where a student of his missed a few bets with AA because he misread a tell.


Pat