I did not want to forget to answer this one.

By "depended on" I mean (and I think the law means) that which is used to determine the result. By "influenced by" I mean (and I think the law means) all the other factors present which go into establishing that which is used to determine the result.

In all sports, for example, the outcome depends on the score. The winner of every football game is the team with the highest score. In getting that score numerous factors come into play: yards run, yards passed, time of possession, interceptions, kicking, etc.... All of those things influence the score, but the score determines the outcome. As has been seen many times, one team can beat the other team on all the relevant factors but still lose because the other team had one or two great plays that led to a higher score. So how well the teams played certainly influence the score a football team gets in a game, but the outcome of the game depends on the score alone, not on how well they played.

Similarly, in poker the outcome of a hand with a showdown depends on who has the best hand/cards. How folks played influenced that hand (and determined the amount of the pot), but who won or lost depends on the cards revealed. OTOH, the outcome of a hand where all have folded to the winner depends on the players all folding to the winner. That the other players chose to fold to the winner was likely influenced by their cards, but the outcome did not depend on their cards. It does not matter what their cards were once they chose to fold, so the outcome did not depend on their cards.

IOW, the human decision was a determining intervening fact. The decision was the fact that determines the outcome. That the decision was influenced by the cards, that the decision may have even been so obvious that it would be okay to say the decision depended on the cards, does not make the outcome dependent on the cards - it still depended on the decision.

Skallagrim

It certainly would be quite a coincidence if poker was literally the only multiplayer strategy game with random elements which has ever been found to be "predominantly luck" by any court. However, it seems quite possible that this is the case, from what I've encountered.

I expect there have been a few odd court cases where a trivial-strategy slot-machine-variant (video poker, etc.) was examined by the predominance test, but would be interested in seeing details of these decisions, as I don't think I ever have. Not that any assessment of these types of games will necessarily inform any issues surrounding poker.

This is very good post on the issue.

I would simply add that it is quite possible for superior skill to dominate in even a single hand. Correctly reading your opponent as scared and likely to fold if you bet does not always require a history. Sometimes a good player can just read it (especially against a novice not adept at hiding emotions) even against someone they have never met before. And so the more skilled player wins regardless of whether the cards that better player holds could ever win if the hand is played to showdown.

"Why this hasn't been done" is because it is actually really difficult to show non-players that the player getting lucky by hitting his hand is in reality the least likely result in poker. Since we have to admit it happens, what we also have to show is that it is not the more important part of the game overall.

Proving that has been the difficult part, as this thread surely attests.

Skallagrim

Would you agree that in order for a test like this to be meaningful it must produce reasonable results when applied to any, or at least almost any, game?

Please describe in a bit more rigorous detail how you determine that chance predominates in blackjack. I promise I can describe a game in which it is intuitively obvious that skill predominates to at least the degree that it does in many forms of poker which will fail the test.

You can play poker against the house. In all of the biggest casinos in Vegas there are now limit hold'em bots which you can play real, full scale poker against. Aria will now let you play 200/400 against it.

So you think that hands that hands that go to showdown are "luck hands" and hands that don't go to showdown are "skill hands" and you can just count them up and see which there are more of?

I did not say that.

I think hands that go to showdown have an outcome that DEPENDS on the cards dealt.

I think hands that do not go to showdown have an outcome that DEPENDS on the decisions made by the players.

Whether "skill" or "luck" was the more important factor influencing how the hand played out is a different question.

Skallagrim

Ike has a better legal mind than most lawyers in this country.

I do not know what additional detail you are looking for. In blackjack every single hand (in which there is not a surrender) ends by consulting the cards. Cards are randomly dealt. Decisions as to whether to hit or stand have a clear effect on the outcome of the hand. There are mathematically correct decisions and mathematically incorrect decisions. Over time the ability to make the correct decisions will increase your odds of being a winning or near winning player. But in actual fact whether you achieve the mathematical result that correct play says is possible will be determined by whether or not you are dealt a statistically representative series of cards. If you are dealt nothing but winners or nothing but losers no amount of skill can change that result. It is impossible for skill to defeat the deal of the cards in blackjack. That, effectively, is the definition of "game of chance" for all games that involve a deal of the cards.

OTOH, being dealt a series of bad cards in poker makes it difficult to win but not impossible. The cards do not always determine the result. In fact, as has been noted before, most of the time the cards are not even consulted to determine a result.

Good point as I had forgotten about this recent development. But doesn't the fact that a computer company has devised an algorithm it thinks is good enough to defeat most folks at HU Limit Hold 'em tend to prove precisely what I am saying? IOW, does not the existence of this machine mean that how the cards are played is more important than how the cards are dealt (at least in HU Limit) and thus now a company/casino is actually willing to put up money based on this point?

Skallagrim

In my opinion people are thinking about this the wrong way in terms of skill and luck and that is because the difference in skill between any two people who know the rules is not that great (yes even between Ike and some noob). The better way to think about this is if two people are playing and one is trying to win and the other is trying to lose and they both follow the rules of the game will the person trying to win usually/always win. The answer for poker is obviously yes. If I raise/fold 100% of the time and my opponent is not a moron and knows that I am doing this, he invariably wins. This is true for all games of skill, (basketball, tennis, etc...) but is not true of games that involve skill (i.e., sports-betting, stock market investing, etc...). While following the rules of these "games" even if you try to throw a match you could not be guaranteed to do so. I think this demonstrates the predominance of skill in poker over luck, however, the problem is more complicated as skill edge is so small and thus the likelihood that an individual wins any given match or over any given period of time devolves into a statistical argument but that seems to be completely irrelevant for whether poker is a predominantly skill game and thus differs from sports betting. I hope someone lets the PPA know about this argument because even though it seems trivially obvious to me, I haven't heard it yet (admittedly though I haven't read much on the subject).

This needs to be clarified. For the sum, i.e counting # of heads, standard deviation increases, but it gets smaller relative to the total sum as you pointed out.

By the same concept, standard deviation does decrease for the average, i.e #heads/#total.

Let me know if you want the math.

Caveat: The predominance test, contrary to the assertions of the defense attorneys in the case and the poker lobby, has nothing to do with this case. The materiality test in the ny statute quoted by ike, as he said, leaves no doubt in the mind of a reasonable person that poker is gambling under NY law.

The predominance test, for the reasons ike so eloquently stated, is a nonsensical standard to apply to games in general and poker specifically. However, it does happen to be the legal test in quite a few states, so we do have to put forth the best arguments we can in that framework.

So here we have it, a lawyer convinced that playing poker online was an illegal activity:

agrees to T&C that refer to local legality

plays on line illegally

now sues the poker site for 3x his deposits for being organised crime

I suppose all that can be said is he is a credit to his profession.

NY's statutory language of "where the outcome depends to a material degree on chance" is also open to interpretation. At the risk of being called an unreasonable person (hardly for the first time), let me note simply that:

1) the leading expert on NY gambling law, Prof. Bennet Leibman wrote an article addressing this precise point of the difference between the predominance test and the NY statute. It is titled: "Chance v. Skill in New York’s Law of Gambling: Has the Game Changed?" It can be found (but costs money to get) at GAMING LAW REVIEW AND ECONOMICS, Volume 13, Number 6, 2009 ©Mary Ann Liebert, Inc. DOI: 10.1089/glre.2009.13603. The Professor's basic conclusion was that the NY statute was pretty much just the predominance test with the nuance that if a game is 50/50 it is gambling under the NY statute, but not gambling under the predominance test (where chance must predominate).

2) A very recent case concerning the game of Mah Jong was heard at the NY trial level and the court stated the following: "While some games may involve both an element of skill and chance, if "the outcome depends in a 'material degree' upon an element of chance," the game will be deemed a contest of chance. (Donnino, Practice Commentary, McKinney's Cons Laws of NY, Book 39, Penal Law § 225.00, at 356 [2008 ed].) "The test of the character of the game is not whether it contains an element of chance or an element of skill, but which is the dominating element that determines the result of the game.(People ex rel. Ellison v Lavin, 179 NY 164, 170-171 [1904].)" The case name is People v Li Ai Hua and you can read the full opinion here: http://www.courts.state.ny.us/report...2009_29241.htm

In the end it may well be that a court uses the "material degree" language to sink poker as a "game of skill." But to say it is unreasonable to think any other outcome is possible goes way to far in my opinion.

Skallagrim

let's play "spot the incorrect statement"

I counted 3

Ok, this is sorta what I was getting at. However, you're no longer talking about predominance per se, which is the only thing I was interested in in the first place. No one in their right mind doubts the statement "Poker is a game of skill."

Your argument doesn't really seek to show that there is more luck than skill in blackjack, i.e. that luck predominates. It kinda just makes up an arbitrary definition of "game of chance" and then shows that blackjack fits that definition. However, it's also a really bad definition of game of chance and one that pro-poker people seem fond of, so I'll go ahead and point out what's wrong with it anyway.

Consider HU limit hold'em with no folding allowed. Your test identifies this as a game of chance. This is clearly wrong. Good players will beat up bad players just about as badly as they do in traditional poker games.

Alternatively, consider casino war with the following additional rule: In addition to your normal bet, you can choose to bet any amount on a side bet that never wins. We'll call this bet the money burner. If you always bet more on the money burner than on the main bet, "skill" can defeat the deal of the cards in that you can make it such that you lose even if are dealt an endless stream of winners.

To anticipate your objection that these are not real games: Why does that matter? If your test makes sense and works, I ought to be able to describe a plausible game and your test ought to give a plausible answer. The reason it does not is that there is not actually any causal relationship between the presence of the characteristics you identify and the degree to which skill determines the results of the games in question. It is merely a coincidence that games you would like to identify as games of skill tend to have these characteristics while games like blackjack do not. Hence, it is easy to construct a game which any reasonable person ought to conclude is a game of skill which fails the test or a game which any reasonable person ought to conclude is a game of pure chance which nevertheless passes it.

I think you've lost track of what I'm arguing. I'm not arguing that poker isn't a game of skill. I'm a professional poker. I've been making a living at poker for the better part of a decade and I don't think it's because I'm exceptionally lucky (I am exceptionally lucky, but that's beside the point).

My point is that the "predominance test" is nonsense. The presence of luck and skill in games are binary traits, not quantities. Asking whether luck or skill predominates in a game that involves both is like asking whether a large, gray rock is predominantly large or predominantly gray.

It is very difficult to prove a negative argument like this, i.e. to prove that a meaningful comparison between two things cannot exist. This is especially so when it is not even entirely clear precisely what "luck" and "skill" mean. The best I've been able to come up with so far is to prove that all efforts so far to make this comparison fail.

Of course I am still talking about predominance. There is no basis fro your assertion to the contrary. In BJ (like in poker) 2 things are happening: cards are being dealt and players are making a decision. Both influence the overall outcome of the game. But in BJ no matter how good you are at making decisions, the deal of the cards will still always determine the outcome and that is why it is predominant.


On this we agree, at least with respect to people knowledgeable with respect to poker. The one thing that still baffles me is why so many people knowledgeable with respect to poker seem to resent my efforts to prove this fact in a court of law or in the court of public opinion.

The definition comes from the courts, not me. Consider the North Carolina case of Joker Club v. Hardin decided in 2007. In that case, despite otherwise being presumed to be in their "right minds," the judges ruled poker a game of chance. That did so on the following basis: ""...in poker, a skilled player may give himself a statistical advantage but is always subject to defeat at the turn of a card, an instrumentality beyond his control. We think that is the critical difference...." You can read the whole opinion here: http://groups.google.com/group/tpjou...4956076345fb18

I cannot imagine how a game with no folding could be called poker.

I also have not said that the question of skill v. chance is resolved by the issue of whether good players can still beat up bad players, assuming such is the case. That is your test, not mine or the one required by the courts.

Your side bet is neither part of the game nor a factor in the outcome of the game. It is not in anyway influencing or determining whether you win the game of war. It is only determining, as your example is posited, that you will leave the table with less money than you started with. Why not just assume that you will tip the dealer twice what you win on every hand you win?

It is a key aspect of poker that betting is inside the play of the game. As such it influences, sometimes even determines, the outcome of the game (winning or losing) and not just whether you go home with money or not.

I disagree with this conclusion. As it is only a conclusion, I do not know what else to say other than that I have previously identified numerous ways in which "the presence of the [specified] characteristics" determine the results of the game. If you go to showdown and your opponent hits his 2-outer the deal of the cards has determined the outcome of that hand. If, with the same cards, you bet and your opponent folds, the decision to fold determined the outcome of the game (and it is a different outcome from what the cards would have produced).


This has been your point from the beginning, but you are wrong that I do not understand it. It is simply that you have hardly convinced me that establishing which of the "binary traits" is more responsible for the outcome of a poker game is impossible. Especially since, unlike you, I see the distinction between the two "binary traits" as very clear and precise: the deal of the cards is luck (actually "chance" in legal terms) and the decisions made by poker players are skill.

Skallagrim

Really nice post.

"In any game with any element of skill at all, skill will be the predominant factor in determining the outcome, simply because luck evens out in the long run. More chance doesn't mean less skill. Less chance doesn't mean more skill. It is a stretch to even argue that either is a scalar quantity such that you could meaningfully compare "the amount of skill in poker" to "the amount of skill in chess" or "the amount of luck in backgammon" to "the amount of luck in deep stacked pot limit Omaha." Even if you can come up with some way of measuring things like that, it's just willfully ignorant to imagine that there exist quantifiable relationships between the two that allow a meaningful quantitative comparison of the proportion of luck to skill in two different games or a determination of which of the two elements "predominates" in a given game."

I don't think the above is quite right though. A more reasonable interpretation of the DOJ's position would be to say that 'A game of skill is one in which skill determines the outcome (within a certain level of confidence) within a reasonable amount of time.' Interpreted in this way, poker is game of "predominantly luck" if luck determines who wins and who loses within a (hand? session? week? year? lifetime?).

For example, I think it would be perfectly reasonable to call a game with a standard deviation of 10,000 bb and a range of winrates from -10 bb/100 to +10 bb/100 and hands which take on average 5 minutes, as a game of "predominantly luck."

Again, interpreted in this way, we can have a reasonable discussion with the DOJ while mutually accepting prior precedent. Stock trading is legally considered not to be a game of predominantly luck, and accepted "skilled" stock traders have losing years. We can then show that stocks are not a game of predominantly luck if and only if poker isn't a game of predominantly luck.

I think the stock/financial market comparisons are kind of useless, since they have a critical underlying purpose to our economy. Online poker doesn't. Comparisons to the dozens of online games that you can legally bet on, seem more useful. But this is just an ad hoc observation.

So is Willie and Kenny

What critical underlying purpose to our economy do synthetic CDOs serve? I have no idea what the numbers are but surely a huge $ volume of the securities transactions in this country are speculative rather than having some underlying economic purpose.

I'm not sure why you want to name exotic examples that may be exceptions to my simple factual statement. The "stock exchanges" are the thing named ITT that I responded to.

Virtually 100% of them are speculative and contain risk, including the essential ones I was referring to.

It's widely different from a poker tournament because only terrific golfers enter, unlike poker where many terrible players enter.

MM

No. The luck versus skill argument is extremely important, but it has not been presented correctly. The correct argument is a "luck and skill" argument where the skill can be enough to overcome the luck in a reasonable amount of time.

Mason

Suppose two players, A and B, are evenly matched but A gets lucky and wins the first 10 hands. Since they are evenly matched the expectation for each player from that point on is zero. But for luck to even out, it would need to be -10 for Player A and +10 for Player B. But this would contradict the fact that they are evenly matched.

Mason

Skall, Todd, spadebidder, et al:

I can only imagine how all this will play out in front of a jury once Elie and Campos actually go to trial. If the defense and prosecution get caught up in an endless "skill versus luck" argument - and the judge actually allows it - I can see jurors eyes glazing over. To whose advantage would a trial that devolves into an endless series of technical arguments accrue - the defense or the prosecution? Will jurors be receptive to an argument over whether a crime was even committed since the banks actually profited from these transactions? How many jurors might harbor feelings of resentment toward Wall Street bankers and wonder why the Government is picking on poker sites? Which of these arguments will the judge allow - and which of them will the judge nip in the bud?

I assume the prosecution will attempt to strike any potential jurors who have played poker - especially internet poker - during voir dire; but what if, (despite all their efforts), they wind up with a jury consisting of one or two poker players and ten non poker players? It takes a unanimous verdict to convict in a criminal trial. There's always the possibility of jury nullification - or a hung jury. A "not guilty" verdict would be devastating for the Government while, at the same time, guilty verdicts would be devastating for Elie and Campos. This reality leads to the obvious question: How likely is it that Elie, Campos, and the DoJ will reach a plea bargain agreement with Elie and Campos pleading guilty to much reduced charges? Does the Government really believe their case is so strong that they're willing to take it all the way and risk a total defeat?

Finally, even if Elie and Campos are convicted, what will the Government have gained that they haven't already accomplished? Based on previous comments from Professor I. Nelson Rose (and others), Isai Sheinberg, Nelson Burtnick, Ray Bitar, Scott Tom, et al, (i.e. all the defendants physically located outside the United States), will likely never be extradited and never stand trial. Sure, these individuals have lost money and lost business, but they haven't lost their freedom.

It looks like the "big winner" here is Mr. Sheinberg. He continues running (and will probably continue running) the world's largest internet poker site - regardless of what happens here in the United States. I would sure love to trade places with Isai.

Former DJ