Harrington's %10 bluffing rule

Does it really apply to every situation?

His rule states that, regardless how tight a player is, any bet from someone has at least a %10 chance of being a bluff. He never really explained where he deviated this from or why it's a concrete rule, but I think there have to be exceptions, right?

I'm guessing that's an average for people you have little or no info on.

Some people think I am the exception...

lol

Quote for truth.

I think you're exceptional, Daks.


But yeah, I've always interpreted that as applying in general, and probably particularily at serious levels that Harrington plays at, and just based on his experience rather than any type of research.

I think what's he's trying to point out too is to never totally eliminate the possibility that your opponent might be bluffing.

I bluff around 10-20% of the time but only to a board i would imagine my opponent to have also missed and i normally wait til mid-late in the game so people start to think i only bet when i have a hand, works well most of the time unless you get caught straight away.

I think this is pretty much what he was trying to say. I was very skeptical at first, but as I started to see stuff happen with that rule in mind, it made more and more sense. Many people simply do not understand how to bluff, and simply think it means shoving chips in the pot with sh**. It just means that no matter what you may think, even if you think it would take an idiot to bluff in this situation, you should assign at least a 10% chance of them bluffing, because although unlikely (10% is still very low), there's a chance they're making a move with 9 high.

Do you guys think that we could go so far as saying that every time we are facing a bet giving us 9:1 or better on the river that we have to call with a hand we know is beat but can beat a bluff?

I believe the 10% rule applies a lot more to live play than online play. Online play is a little easier to bluff because you dont have to face the guy you are bluffing, but live you have to stare down, and be stared down on, so you would mostly likely bluff less for that reason...

online i usually go with 15-20%

live i stick with the good old 10%

Harrington also says that if you are not bluffing 10% you are not bluffing enough. Sort of a self-fulfilling number there.

I just read that section, and he said that was from experience.

Would that include standard steals from both the button and the SB?

hmm thats a little to much for my liking, but maybe ill give it a shot, just to test it out

i have a feeling it would be post flop play

And maybe it applies more to tournament play than ring games, especially if you play one table SnGs.

Here I'm thinking of very short table play where I'll take stabs at pots with nothing just because I think the other player can't or won't call. I'll do this a lot more often than bluffing at a full table, although c-betting is a form of bluffing as well.

No.

While there are plenty of times you can correlate event odds with dollar odds, this really isn't one that translates directly.

But, if you do make this call it probably wont be that big of a dollar mistake long term...

Not sure exactly what you mean - are you comparing real dollars to T$?

I meant for my example to be a cash game one, but forgot to add that. Also forgot to say we're HU for simplicity.

No.
While it's all fine and dandy to say that your opponent bluffs at least 10% of the time, you have to remember that that's long term/run. In any given situation it could be 100% bluff or 100% stone nuts.

So to say "if your opponent bets 10% of the pot you should call," isn't exactly true (although as I said, I don't think it's a huge error to do so).

I think that Harrington is saying that 10% is something you have to factor into your calculations as far as putting your opponent on a range of hands. For example, you shouldn't think it's 80% he has top pair and 20% he is on a draw. You should factor in in the 10% bluff factor some how (75% tp, 10% bluff, 15% busted draw?). Regardless, this isn't how you think at the table. The point is that you need to consider the (naked?) bluff as a factor as well, and 10% (long term) is enough to make it a statistically significant factor.

Well that's what I meant.

For example, if we've got %90 hands that beat us and %10 bluff pinned for our opponents range, then we need at least 9:1 odds to call...

Right, but that isn't the same as saying you should aways call if your opponent bets only 10% of the pot...

I think I'm just confusing myself here.

Why not? If we're getting 9:1 or better, and apply Harrington's minimum %10 bluffing rule, then we should always be calling, no?

Not if it's 100% that you are beat in the specific situation.

Example:
we have 42s, and check from the bb to the button limp (sb folds). Flop comes down AK3, giving us the gut shot. Check, check. Turn: Q, check check. River: Q. Opponent bets 10% of pot. Clearly you can't call. [ok he can't actually bet 10% here, since there hasn't been enough action, but don't make me think of a scenario where someone else called and you called a flop bet w implied odds, the point is the same]

Similarly, in any given specific situation you might have a very specific read on your opponents bluffing frequency, either + or - the 10%.

The point is that you need to factor in at least a 10% bluffing frequency when assigning a range LONG TERM. There are time you can be, for example, 93% certain that your opponent isn't bluffing.

I think...

or maybe I'm wrong here... could be...

Well that's true. That's what I meant with the bold part here:

That's what I'm getting at. With enough info, you could come to the conclusion that a certain type of player will bluff-raise <%10 of the time, for example, right?

Well now you've just lost me

I'm not sure why he would mention %10 minimum bluffing frequency in a hand analysis if the %10 is supposed to be meant for long-term applications.

I can't argue always calling a 10% pot bet.

After reading through this Chuck, it looks like you're taking Harrington's Law out of context (based on OP). The 10% is not about calling when the pot is 9:1 odds, it is part of your hand analysis equation for the overall guesstimate of your chances of winning the hand then compared to the pot odds to make your decision. The 10% is not used by itself which is why you can apply it to every decision in your analysis.

Take Harrington's example specifically: Hero is UTG with A♦A♣ and raises to 4xBB. Villian (MP, experience and conservative) flat calls and flop comes out 9♥5♣2♠. Hero bets the pot, villian reraises to put hero all in. Pot odds are 3 to 2 for us to call.

HA time: He could be on a set. If he is, then were about 10% to win. He could be on a big pair and we're 92% to win with 2 cards to come. He could be bluffing with a couple big cards making us 97% to win.

Because he's a conservative player, we give him the minimum 10% chance of a bluff. We also make an educated guess that 50% of the time he has the big pair and 40% of the time he has the set. Now with all that established (guessed), we wind up with:

50% (has high pair) * 92% (we win) + 40% (has set) * 10% (we win) + 10% (bluff) * 97% (we win)
= 50%*92% + 40%*10% + 10%*97%
= 46 + 4 + 10 (all numbers rounded)
= 60% chance that we will win the hand overall based on several factors.

Since pot odds are 3 to 2, we make the call. We don't make it because there's a 10% chance he is bluffing, we make it because we guesstimate that we will by river about 60% of time regardless of what he actually has.

I realize I used the #s right out of the book, but I'm hoping that discussing this whole thing in context using the same numbers might help.

What about the same hand with a super rock villian? Let's say he hypothetically makes this move with either a set or a bluff only (even rocks will bluff - it's one of their perks because they get the credit). Still 10% is the minimum for a bluff, but now 90% is a set.
= 90%*10% + 10*97%
= 9 + 10 (rounded again)
= 19% or about 4 to 1. Since pot odds are still 3 to 2, we now have to fold.

I'm rambling now, but I guess the key is that regardless of where the 10% comes from (I'll take Dan's word on it, tbh), it's only a part of the equation (so to speak).

jesus i must of read that book a long time ago because i am just now remembering the example that harrington put out


jesus christo jack good research.

Without trying to sound rude, I'm not quite sure what point you're trying to prove, JD.

While your (Harrington's) example is obviously right, it's different from mine. The difference being that we're being put in a call/fold decision on the river, not the flop. On the river, we don't have to factor in our % to win with each hand group - it's pretty much just win/lose, and we decide whether or not to call based on our odds. (We could factor in multi-level thinking, ie he's betting small enough to give 9:1 to try and induce a call with a monster, etc etc, but I'm ignoring this for simplicity).

I am applying the %10 rule to a specific (and different) example, but I don't think the logic should change. If there's still a %10 chance of him bluffing, then we should be calling the river getting 9:1 or better if our hand can beat a bluff. There's no real structured HA needed.

Example:

You raise with A♠K♠ in late position, and the nitty big blind calls you.

You both see a flop of 6♠7♠2♥. He checks, you bet, he raises, and you call. (again, ignore actions up to river for simplicity's sake)

Turn comes 2♦, and you both check.

River comes J♦, and villain makes a bet that's giving you >9:1 odds.

You suspect he's on a bluff at least %10 of the time, and has something that beats you (8♣8♠ for example) the remaining percent. You should call based on the fact that your pot odds are giving you a good price (right?)

Anyways, I'm not really sure where this is heading, anyways The hand I'm explaining is really theoretical and doesn't apply much. My original question is pretty much asking whether the rule is concrete for every single example, or if it's just a guideline since it applies most of the time.

Sorry, guess I replied to your OP, not what it drifted off into with various hypotheticals tacked on to prove points.

Well simply put then, if you're not looking for hand analysis, then no the 10% rule doesn't simply apply to every hand. My point was that you are taking the 10% rule out of context, plain and simple. It is not a rule in and of itself, it is part of hand analysis. That's why you can use 10% on a conservative opponent and higher % on looser ones.

And yes, you can still apply hand analysis on the river. Running late for work, I'll check back later. Until then, I'd say reread that section again.

Chuck, you used the same analysis tool that Jack did, just with a different situation. Your analysis was 10% bluff, 90% pair or better. So, it is 10%*100% + 90%*0% or 10% to win. 9:1 odds is just break even.

Chuck, I'm in the camp that will call almost any 9:1 bet. My decision would be based on my holdings. If I were holding small suited connectors against the flop you described, but still only a draw, of course its automatic, but if I got JT or QJ I might not call 9:1, but with high card possibilities I think in your example it is automatic call.

Poker has math , but it also has psychology. WIth a flop like that if you haven't seen villain get aggressive with babies or small suited connectors, then it looks to be a bluff.

I took the Harrington section to encourage the reader to give credence to the real, but remote (he's estimating 10%) possibility that you are getting bluffed. With a 9-1 bet, It wasn't much of an effort to pull off that bluff tho.

But that brings up another possibility. With such a situation, could you really even call it a bluff? I can't hardly imagine anyone trying to steal a pot without at least an ACE in hand.... Oh wait, someone we know did recently try to do that..... But he didn't leave 9:1 odds on the table so nevermind.

In your example there are only two sane possibilities, and they are WAWB either someone is betting his paired 2's or they are betting big unpaired unmatched cards. Well you beat the 2nd scenario, so that puts you in the 50% hand range with 9:1 odds. Gee, decisions, decisions.....

Well I think we've all kind of agreed now. It basically all comes back to this:

I think the %10 rule applies to most situations unless you've got a few 1000 hands on a player and can narrow your opponent's range beyond that %10 in a particular spot.

Anyways, thanks for the responses everyone. Didn't expect this thread to go so far. Good stuff

Harrington says that the MINIMUM you should use in your bluff estimate is 10%; that even rocks bluff that much. You could probably use a higher figure from what you see in a single session with a LAG.