Response to Sklansky's article "Chips Changing Value in Tournaments"

[David Sklansky]

30 comes from Aaron Browns simulations. You need to have approximately enough chips such that you can wait for a hand which is a favorite over the best of all the other hands. (The idea being you will get all in against that hand) Roughly speaking.

[George Rice]

[ QUOTE ]
Anoter point I do not see mentioned JUST ABOUT ANYWHERE is that when you take a coinflip and double-up - your chip's EV may go down - but what about the increase of your $EV due to the fact that there are now LESS people in the tourney that you need to go thru to get to the 1st?... Obviously, the closer we get to the money - the stronger this effect becomes (and it's obv negligible on day 1 of WSOP ME)...

[/ QUOTE ]

If you're suggesting that the chip's individual EVs will increase (above their value before the hand) when you add in the fact that a player was also eliminated with the increase in stack size, that isn't so. The reason is that the EV the eliminated player loses is shared with the other remaining players, not just you. So if you doubled through another player who started the hand with the same number of chips as you, your EV wouldn't double because you are sharing his EV loss with the other remaining players (and you both started the hand with the same EV). This assumes equal skill for all players.

Now if that player was eliminated by another, and your chips remained the same, then the value of those chips would increase, because some of the eliminated player's EV will be shared with you.

[George Rice]

[ QUOTE ]
Obviously you can respond or not as you see fit. What I get from your response is "I don't know what Snyder is saying (since I haven't really read it) but if it disagrees with me then it must be wrong." I think you're severely underestimating the number of people who are paying attention to this discussion. A couple dozen people have posted in this thread, but there are a lot more watching from the sidelines and previous discussions regarding The Poker Tournament Formula have had thousands of views.


[/ QUOTE ]

Speaking for David Sklansky (lol, and old 2+2 joke):

I assume when he referred to wading through a complex dice system he meant some sort of craps system. If so, the analogy is clear. Some players come up with elaborate dice schemes which seem to work when you follow the math they show you. But they always fail to take something into account, that, when finally factored in, shows the results common sense would dictate. You're going to lose 1.4% of the money you bet on the pass line no matter how you slice it.

As for wading through Snyder's article, I don't blame him. It's hard to justify spending the time finding someone else's error. I stated something similar in another post:

[ QUOTE ]
By the way, I can't bring myself to read the rest of your article seriously because you try to explain the ramifications of your discovery (loss of value of chips in tourneys). The whole discussion is based on an error of calucation.

One wrong turn at a fork in the road . . .

[/ QUOTE ]

In an email me, Snyder responded in part:

[ QUOTE ]
Unfortunately, your understanding of my article is incorrect. You missed the point, but then, of course, you say in your post that you didn't read my article.


[/ QUOTE ]

On their site, his wife posted, in part:

[ QUOTE ]
George Rice, for example, put up a post that purports to refute the articles, but he makes exactly the same mistake in his post that Arnold is addressing in the articles. Then he brags that he hasn't read the articles!

[/ QUOTE ]

Note that I didn't claim that I hadn't read the article. I stated I couldn't read it seriously. What's that about?

Once the flaws were identified, it's Snyder's responsibility to find them and fix them. Any time and effort put in by David, Mason, you, me or others is above and beyond our jobs. You, I, and others do what we've done mainly for the learning process. David and Mason have their reasons, which probably include educating us and defending themselves. But none of us is responsible for doing it. And a thorough debunking of Snyder's articles has little up side. It's more than enough to show a few errors to put the ball back in Snyder's court.

But guys like you and I needn't worry. You'll win plenty of money following Snyder or S&M. Their recommendations are close enough for that. It's Snyder's reasons that seem to be off.

[Zim]

I dunno, seems to me we have a lot of posts here ducking Snyder's article. And they all contain words to the effect of ...

Well, we think he's wrong, sorta, but the play is right. Regardless, we're not going to respond because, uhh ... hmm ... we're too busy. And, like, even though we don't read the articles we criticize, or the books for that matter, take our word for it. And can't we just, like, talk about something else ... like the important distinction between stats and math?

I'm giving Snyder this round.

Ball's in your court.

[BigAlK]

[ QUOTE ]
... like the important distinction between stats and math?

[/ QUOTE ]

Thanks Zim. I was beginning to think I was the only one who thought that certain people were responding by nitpicking at the least important part of a post.

[Mason Malmuth]

That's a very important distinction. Snyder said that we, as mathematicians, just put numbers on everything. That's not what we do. It's a very important distinction for having success in the gambling field whether it's poker or womething else, and as far as I know, we were the first to point this out and emphasize it in out work.

MM

[George Rice]

[ QUOTE ]
I dunno, seems to me we have a lot of posts here ducking Snyder's article. And they all contain words to the effect of ...

Well, we think he's wrong, sorta, but the play is right. Regardless, we're not going to respond because, uhh ... hmm ... we're too busy. And, like, even though we don't read the articles we criticize, or the books for that matter, take our word for it. And can't we just, like, talk about something else ... like the important distinction between stats and math?

I'm giving Snyder this round.

Ball's in your court.

[/ QUOTE ]

I can only assume you don't understand what your reading.

Things Snyder has claimed have been disproven. PERIOD.

This whole angle taken by some reminds me of something that happened years ago. James Randy debunked a famous faith-healer on the Tonight Show. He had audio tape of the faith-healer's wife sending personal information about the audience members to him on a radio frequency. The faith-healer was pretending to get this information from God or something of that sort. The televison stations were flooded with telephone calls from people who wanted contact information on the faith-healer. Care to guess why? Because they wanted to make donations to him (the exposed crooked faith-healer).

I know that Snyder calls himself the "Bishop," but you don't have to take everything he claims on faith. I don't think he's really a member of the clergy. But the way some follow him, he might as well be.

[David Sklansky]

"Regardless, we're not going to respond because, uhh ... hmm ... we're too busy"

I already said I would respond to a specific question regarding my contention. That contention being that if you are a winning player with a certain stack sizes or higher, such that you are more likely to double up than go broke from that point on, than your EV in a tournament, once you reach that stack size, does not increase proportionally, in any kind of normal tournament structure including winner take all. If Snyder says that's wrong he must be wrong.

Notice however that I said nothing about strategy. Other than you should not risk a large amount of chips, above this stack size, with a tiny positive expectation (unless you are trying to increase hourly rate rather than EV).

[BigAlK]

[ QUOTE ]
I know that Snyder calls himself the "Bishop," but you don't have to take everything he claims on faith. I don't think he's really a member of the clergy. But the way some follow him, he might as well be.

[/ QUOTE ]

On this point you and I are 100% in agreement. I've thought this entire discussion has had religious overtones from the beginning for people from both camps. In fact you aren't the first one to make this comment in this forum and it's been discussed elsewhere as well.

I'm not a well known authority on poker like Mason. I'm not considered to be an expert in poker theory like David. Nor am I an acknowledged expert in other forms of gaming who is expanding his area of expertise like Arnold. One claim I can make is a long history of questioning ideas put forth by authority figures if they don't make sense (hopefully in a respectful, if still challenging manner).

Here's how I see the status of these discussions.

The whole valuation of chips matter is a non-issue. Snyder may be right, Malmuth and Sklansky may be right, they may each be partially right. Until I can cash in my stack mid-tournament the only thing that matters to me is what will enable me to finish tournaments in the highest position possible (maximize my ROI). The items that have been discussed over several "Snyder" threads and my thoughts on each are:

Optimal rebuy/add-on strategy

If given the choice between the opinion of one or two "authorities" versus the consensus opinion of multiple people who are actually showing success in an area, practitioners rather than ivory tower thinkers, I'll go with the successful practitioners opinion without compelling evidence that they're wrong. For the successful practitioner opinion I'd point to item #15 in the FAQ of the MTT forum which clearly agrees with Snyder.

Utility Value of Chips

As Mornleth pointed out in this thread, the strategy suggested by Snyder's chip utility theory is also in keeping with the concensus of the more successful participants in the MTT forum.

The Impact of Tournament Speed on Strategy

Frankly I can't believe it isn't considered obvious by everyone that you have to make strategy adjustments based on tournament speed. I also believe that (at least in spirit) it is in agreement with Harrington's use of inflection points. Since Harrington makes adjustments for M based on a short table (because you will go through the blinds more quickly) it makes sense to me that if M is supposed to reflect the impact of the blinds grinding away at your stack that an adjustment for the blinds grinding quicker is logical. A simple adjustment for effective M due to quickly raising blinds is not enough because (as Mason pointed out) this could lead you to some incorrect decisions (going all-in with a small pocket pair before you should consider yourself that desperate for example). However I'm convinced that some styles of play that are successful in slower tournaments will only result in you getting eaten alive in faster tournaments. In the limited discussion of this in the MTT forum there wasn't what I could call a consensus. Some felt, as I do, that speed obviously called for adjustments. Others didn't.

Al

[George Rice]

[ QUOTE ]
The whole valuation of chips matter is a non-issue.

[/ QUOTE ]

It should be. It remains an issue because Snyder hanged his hat on it for the basis of his theory. He’s been proved wrong and refuses to admit this. So he’s the one keeping it an issue.

[ QUOTE ]
Optimal rebuy/add-on strategy

[/ QUOTE ]

This seems to more a debate of what the debate is. Snyder vs. S&M or Snyder vs. Snyder’s Interpretation of S&M. Mason had to correct Snyder more than once about what he (Mason) recommends. There may be some genuine disagreements here, but I think they derive from the chips value issue.

[ QUOTE ]
Utility Value of Chips

[/ QUOTE ]

I haven’t read that post, nor Snyder’s book. I’ve only read Snyder’s articles. He’s clearly wrong when he claimed utility value in his heads-up equal skill example, which I proved. So you can imagine how confident I am about this theory in other situations. I’ve previously stated that I find this concept interesting and that it may have merit. But that’s a far cry from agreeing with it. I need some proof, which hasn’t been forthcoming. And here’s an example of something Snyder wrote in his third article, which troubles me. First he quoted Sklansky’s Magazine article:

[ QUOTE ]
Notice however that this proof assumes that the great player figures to double up before going broke. That is usually true. But there are exceptions. One rare one is the player who plays a lot better with a big stack. In other words he is not a favorite to double up until he has gotten a lot of chips (almost inconceivable for limit tournaments). This is sometimes the case for psychological reasons, either in his mind or his opponents. Or it might simply be that he is weak playing shorter stacks. Such a player would be well advised to gamble early in a tournament including even calling all-in bets. Meanwhile there is a more common situation where a good player is not favored to double up before going broke. I speak of those times where his stack is very short.

[/ QUOTE ]

Then he wrote the following regarding that:

[ QUOTE ]
Sklansky’s Opinion of “Chip Utility” in Poker Tournaments
In the same 2+2 article, Sklansky does at least acknowledge that there may on rare occasions be a utility value to chips for some “rare” players. He says:

(above quote)

Truly skilled players may indeed be “rare,” but I find it amazing that Sklansky thinks that a player skilled enough to play “…a lot better with a big stack,” might be “weak playing shorter stacks.” Every truly skilled player is weaker with a short stack than with a big stack, because his skill options are so limited with fewer chips. The fact is there is never much skill involved in playing a short stack. Any player’s skill options on a short stack are limited to hand selection and an occasional kamikaze shot at the pot, which is why short stack strategies have been ably reduced to simple formulas in a number of books, including The Poker Tournament Formula.
It’s optimal big stack play that takes real skill, because you have so many options.

[/ QUOTE ]

First, Sklansky hasn’t acknowledged utility value to chips under any occasion. What he’s acknowledging is that chips may gain EV on rare occasions (when added to a stack). He then gives an example of a player whose skill is at a much higher level with a lot of chips than it is with fewer chips. Such a player might triple his EV when doubling his stack. Second, this type of player is rare, probably because most players who demonstrate a lot of skill with large stacks also have a lot of skill with smaller stacks. Third, that sudden increase in EV is due to the skill anomaly of a unique individual, not a universal utility value effecting everyone (although EV is EV, and it works the same no matter how you get it). And I suspect that any EV Snyder accounts for using his utility value theory would already be factored in by Sklansky if he were to cite a player’s skill level (which might be a average of all skills measured under various conditions).

Snyder chose to interpret Sklansky’s observation as an acknowledgement of Snyder’s utility value. Snyder then chose to interpret the “rare” meaning truly skilled players are rare. Skilled players may be rare, but Sklansky was stating something else. Finally, Snyder confuses “shorter stack” with “short stack.” Sklansky cited the player’s skill differential with “lots of chips’ and “shorter stacks.” Shorter stacks could mean average size stack, medium size stack, below average stack, and possibly even short stack. Snyder chose “short stack” so he could make his “never much skill in playing a short stack” observation, and then use that observation to, in effect, ridicule Sklansky for mentioning it. Even if Snyder’s observation is true, Sklansky was saying something else (Perhaps a player plays large stacks well but tries to play average stacks the same way, costing himself EV).

Does Snyder really have that much trouble understanding what Sklansky is saying? I have a hard time believing that. If I’m right, then Snyder is intentionally misleading his readers. If so, then why? That would be an integrity issue, in my opinion (what’s his motive?). If not, then a competence issue, as David suggested in an earlier post. In any event, it causes me, and I suspect others, to question the legitimacy of other things Snyder has written.

[ QUOTE ]
The Impact of Tournament Speed on Strategy

[/ QUOTE ]

Mason has addressed this already. And similar to the re-buy issue, there’s disagreement on what S&M, or in this case Harrington, recommend. My impression is that the two strategies are similar in Mason’s opinion, and that he had issues with how Snyder is arriving at his recommendations. I haven’t read Snyder’s book, and don’t have Harrington’s advice memorized, and cannot comment myself.