Ed Miller's book "versus" WLLH

[Glenn]

Now you are starting to get it! [img]/images/graemlins/smile.gif[/img]

[Ed Miller]

Hi Lee,

First, I want to make clear that I have "picked on" your book only because it is so influential. Countless terrible hold 'em strategy books have been written, and your book is better than almost all of them. In fact, in the category of "hold 'em strategy books for beginners" (which is where I feel your book belongs), it may well be the very best available. I cannot say that for certain, because I have not read them all, but I have read no other book IN THAT CATEGORY that is better.

But being possibly the best available in its category says nothing about its quality in absolute terms. A book can be the best, but still leave much room for improvement.

Just so you understand, I have harped on the errors in your book because SO MANY PEOPLE have read your book, and many of these routinely make SERIOUS ERRORS that I trace back to concepts they learned from your book. If your book were ten times worse, but read by only a few people, I would ignore it.

Having said that, your view expressed in this post is incorrect (and yes, that is a fact). We are arguing about FACTS, not OPINIONS. "Chocolate is the finest flavor of ice cream," is an opinion. So is, "It should be legal for me to own a shotgun." Debating correct limit hold 'em strategy is a debate of FACTS, not opinions.

For example, say we walked to a football field. I said, "I think there are no more than one million blades of grass in this field." You said, "Fiddlesticks, there are easily one million blades of grass in this field, probably many more." We have both made statements of FACT, not opinion. If one so desired, he could count every blade of grass and come to a conclusion about who is correct. Of course, counting each blade is far too time-consuming to be worth doing (unless a lot of money was riding on this particular argument, of course [img]/images/graemlins/tongue.gif[/img]). So to settle this debate, we might divide the field into small square chunks, perhaps a square foot, count the blades in that smaller area, and multiply by the number of square feet on the football field.

That would give us an approximate answer... for we cannot not know with complete certainty who was right or not without counting each blade. However, if that calculation came to 10 million blades, you could be quite sure (though not 100%) that the number of blades is greater than 1 million. Similarly, if the calculation came to 100 thousand blades, you could be quite sure that the number of blades is less than 1 million.

If we calculated that the field contained 1.1 million blades, we could not know with any certainty at all who is correct... that falls close enough to "the line" that it is still arguable.

Now the poker stuff... where I point out errors in your book, we are not close to "the line." My estimation methods (I do not claim to have exactly calculated the EV for almost any poker decision) have come to 10 million blades, yet you are claiming fewer than 1 million. When that is the case, I say you are "plain wrong." Yes, I am not correct with 100% certainty, but I am so likely to be correct that for practical purposes, I simply assert it to be the case.

I have more to say, but I'll end this post now before it gets too long. I'll continue later.

Ed

[MarkD]

I want to go on record saying that I hope any "discussion" between these two men, Ed Miller and Lee Jones, can remain civil unlike discussions between authorities in the past.

[Lee Jones]

[I've read all the posts in this thread so far. I'm just replying to this one]

First, I note that Ed has been careful to disagree with points in my book, not attack its author. I'm sure that with that as a starting point, we can have huge arguments and yet remain civil at all times. As long as we avoid serious topics like "What is the best flavor of ice cream".

Furthermore, I don't make (and never have made) the assumption that WLLH was the alpha and omega of low-limit hold'em books. I still remember the stunned look on my 18-year-old son's face when I pointed out to him that Homo sapiens is not any particular evolutionary endpoint. We're just a point on a Darwinian continuum. WLLH is a point on the evolution of hold'em books.

Ed is asserting that I've said there's 1 million blades of grass on the football field. He asserts that not only is that not the correct number, but that I'm off by such a large amount (let's say an order of magnitude) that it's proveably wrong.

Here's my point: let's say we're going to argue about the blades of grass on a football field, even if we're unwilling to count them all, and choose instead to use Ed's idea of counting the blades in a square foot of turf and then multiply. Fine. That should indeed tell you whether one million or ten million is a better number.

There's only way analogous way I could see to show with near-certainty that my statements are "wrong". You'd have to mine a database of a few zillion hands actually played, and see whether (for instance) it's better to bet or check the turn with QQ when the flop comes T-8-8 and you get called in one or two places.

Look, when anybody starts discussing how blackjack basic strategy is flawed or wrong or whatever, we all laugh. Because we know it's simple enough to Monte Carlo the hell out of it (and the Peter Griffins of the world can actually compute it) and figure out the Right Answer.

Until you do that with poker, isn't it presumptuous to say you know what is "right" and "wrong"?

Now, I'm sure somebody will say "Er, Lee, shouldn't you be the one doing this research?" Yes, I'd love to. And some day I may take on such a project - I think it would be fascinating. And I have actually discussed with PokerStars management the possibility of doing such a thing. But right now, I'm busy helping to run the store, and simply don't have the time to do this.

Anyway, I haven't been reading the small stakes forum, but I'll start doing that. Ed, I'll ask you to meet me over there, and we can discuss specific areas where you say I'm wrong. It should be an interesting talk. On a semantic note, BTW, I should mention that just because I claim you can't "prove" that I'm wrong without data, that doesn't mean you won't necessarily be able to persuade me that you're right and I'm wrong. I've been persuaded of other (more important) things for which there's no data; it's just that I found the argument compelling.

See you in Small Stakes Hold'em.

Regards, Lee

P.S. You'll see that I'm not signing this with my "PokerStars Poker Room Manager" signature. That doesn't seem to be germane to this discussion, so I don't want to bring it in here. When I speak on behalf of or in relation to PokerStars, I'll use that signature.

[MarkD]

I for one hope that you and Ed both visit the Small Stakes forum often and share your knowledge with us non-authorities.

[TimTimSalabim]

Lee,

I read your book when I was starting to play 3+ years ago, and it confused me so much I put it down and just relied on Krieger's book to learn the basics(which looking back was not great but at least ok for a rank beginner like myself at the time). Just today, I picked your book off my shelf to recall what it was and immediately the confusion came back to me browsing the starting hands... how come I can play KJo/QJo/JTo from middle position and not late position? Looking back on it I assume it was just a typo, but being a beginner at the time, I didn't know enough to assume that.

[sfer]

[ QUOTE ]
Anyway, I haven't been reading the small stakes forum, but I'll start doing that. Ed, I'll ask you to meet me over there, and we can discuss specific areas where you say I'm wrong. It should be an interesting talk.

[/ QUOTE ]

I think I speak for all the SS regulars in saying this would be terrific. I look forward to it.

[Ed Miller]

NB: I have written on this topic before. For the things I generalize in this thread, you can find specific examples of in this thread and this thread. I will not repeat arguments from the prior threads here, so if you want to follow along, familiarize yourself with the old stuff first.

I'm going to keep the discussion in this forum rather than move to Small Stakes because I think it's actually more appropriate here. It's also easier. Also, be forewarned that you may not get as much substance from me as you'd like, as I am now, at least theoretically, on vacation. I'd like to spend most of my time relaxing, not posting. [img]/images/graemlins/smile.gif[/img]

Ok Lee. Here's the fundamental flaw in your book:

You consistently instruct your reader to assume that his strong hands (top pair and overpairs) are BEATEN simply because the flop is somewhat scary and one or two LOOSE players have CALLED (not raised) his bet.

This incorrect assumption systematically taints your postflop advice, generally causing you advocate checking too often on the turn and also folding too often.

We are talking about games where players play very loosely. Specifically, they play too many hands before the flop, and they go too far with those hands after the flop. That means that your opponents will often call one bet on the flop with a VERY WIDE range of hands. In my opinion, that reduces your assumption that a flop CALL indicates that you are likely to be beaten to absurdity.

Let's discuss the QQ on a T88 (two flush) hand. I didn't bring your book with me on vacation, so I'm working only from memory. But it does not matter... I can demonstrate my point without the book. I don't remember how many players you had in the hand, but let's assume that it is a typical loose game, and you have four or five opponents.

So it goes something like this. Four players limp. You raise with Q [img]/images/graemlins/club.gif[/img]Q [img]/images/graemlins/diamond.gif[/img] on the button. The big blind calls, and the limpers all call (12 small bets). The flop is T [img]/images/graemlins/heart.gif[/img]8 [img]/images/graemlins/spade.gif[/img]8 [img]/images/graemlins/heart.gif[/img]. Everyone checks to you. You bet, and two players call (7.5 big bets).

In your book you advise to check the turn if your opponents check to you, because it is too likely that someone holds an eight to bet profitably. (Please correct me if I misrepresent you.)

Here's the problem... it really isn't particularly likely that someone has an eight. Futhermore, it is almost certainly LESS LIKELY that someone has an eight now than it was when you bet the flop!

You are obviously correct that, in a loose game where players play low and high cards alike, and many people see the flop, it is more likely that someone has an eight than it would be in a tight game. I don't know what the exact numbers are, but maybe the chance (before any flop action) that someone flopped an eight in the tight game (with one limper and the big blind) is maybe 15%, and in this loose six-handed hand 40%. I'm guessing... the numbers aren't particularly important to my argument. (They would become important if the 40% number were much higher... say 80%, but I am certain that 80% is far too high.)

Where you run into a (serious) problem is how you assess how the probability that someone holds an eight has CHANGED from this prior probability (the 40% number) after one or two players call your bet. You are implicitly applying Bayes' Theorem, and you are doing it incorrectly.

Any number of things could happen after you bet the flop. Obviously, everyone could fold. If that happens, the chance that someone flopped an eight drops from 40% to 0% after the flop action. Or, everyone could raise... so the action could go bet from you, check-raise, check-3-bet, check-4-bet, check-5-bet. While it's not a certainty, that action makes it pretty darn likely that someone has at least an eight. So maybe the probability jumps from 40% to 95% (or more.. again, the number isn't important).

The important thing to note is that neither of these scenarios is likely AT ALL. How often do you expect five loose players all to fold for one small bet in a 12 bet pot on a T [img]/images/graemlins/heart.gif[/img]T [img]/images/graemlins/spade.gif[/img]8 [img]/images/graemlins/heart.gif[/img]? Not very often... like easily less than 5% of the time... probably more like 2% or perhaps even less. So while it tells you a lot about what hands your opponents hold when it DOES happen, the fact that it DIDN'T HAPPEN tells you little.

That is, over 98% of the time, you will get action on your bet. So say you know that the prior probability of someone flopping an eight is 40%. Now I tell you, "You bet the flop and got action." I don't tell you what the action is... just that not everyone folded. Now what's the chance someone has an eight? 41% maybe? Maybe 42%. No higher.

But the fact that you got MINIMAL action IS significant. That is, a decent percentage of the time, you will be check-raised and two or more people will call. Clearly it is more likely that someone has an eight in such a scenario as it is when one or two players simply call. So now once I tell you, "You got action, but it was minimal action... only one or two LOOSE players called," you'd have to conclude that the chance someone holds an eight has DROPPED... significantly I might add. I don't know how much exactly, but probably to 30% or even lower.

So you are completely mistaken in your analysis that getting called here means that an eight is reasonably likely.

Furthermore, you simply must protect your hand in this big, raised pot on this coordinated board with a couple of opponents who probably don't (but still possibly might) have you beaten. If you had A [img]/images/graemlins/diamond.gif[/img]A [img]/images/graemlins/club.gif[/img] and the flop were 8 [img]/images/graemlins/heart.gif[/img]8 [img]/images/graemlins/diamond.gif[/img]2 [img]/images/graemlins/spade.gif[/img], I wouldn't complain nearly as vociferously. I'd still disagree... and I would still recommend betting in a typical game with loose opponents... but your advice to check would not be nearly so bad in such a context.

But in the example given, it is simply wrong. It is based on a flawed assumption about the likelihood of an eight being out, and that assumption leads you to a play that, if you repeat it, will SIGNIFICANTLY lower your earn at small stakes poker.

Simply put, you cannot play small stakes games against VERY LOOSE players and assume that they have your strong hands (top pair and overpairs) beaten. By definition, LOOSE means that they USUALLY DON'T HAVE NUTHIN. If you consistently assume that loose players have big hands when they simply check and call, you will dig yourself into a mathematical hole from which you cannot extract yourself. Your opponents will get "big hands" WAY more often than you will, and you won't win even close to your share of pots. If you take it too far, you will cease to win, even if you play tightly preflop.

[Ed Miller]

Lest it seem like theroetical nitpickiness on my part, I assure you that my criticisms are launched essentially solely because I aim to help intermediate level players reach the next level. I feel that many players find it difficult to progress beyond the "decent player" level in part DUE TO ideas they learned from your book.

In addition to the thread I referenced in the preceeding post (where the fellow complains that he folds his top pair on the turn only to find out that he folded the best hand) which clearly demonstrates the connection, I have essentially two years of experience actively reading and critiquing microlimit and small stakes posts on this forum. I don't know how many hands I've addressed, but it's hundreds. I've read thousands. I see a nearly omnipresent undercurrent of fuzzy thinking, and I can trace some of it directly to your book.

People have absorbed this, "They called me so I must be beaten," mindset, and they CONSISTENTLY complain about folding the best hand, missing bets, allowing people to draw out on them, etc. I see this all the time. That's why I wrote my crushing microlimits post, that's why I point out flaws in your book, and that's why I wrote my own book. I am trying to fix SPECIFIC errors that LOTS of people are making. I have no interest in harming you personally or professionally. I'm just trying to help others, and to do that, I must address the problems in your book. I wish it weren't so...

[deacsoft]

To both Mr. Jones and Mr. Miller,

You both have my respect as noted poker authorities. I learned much from Mr. jones' book and am very much looking forward to the release of Mr. Miller's book. I'm, furthermore, looking forward to the possibility of future debates between the two authors. As we are all adults here, I'm sure that these debates can and will remain civil. I do hope that ego does not become an issue and that either of you will be willing to accept that the possibility that the other is correct my be the case. I hope that both you, Mr. Miller, and you, Mr. Jones, are able to learn from each other as we all have and will continue to learn from you.

Cheers,

deacsoft