NC, USA determines poker = chance

[Dire]

[ QUOTE ]
Ok Dire, rather than send you to the various times I have posted it, here is the argument:

....


[/ QUOTE ]

This was the exact argument my initial post was commenting on. You are essentially circularly arguing that poker is a game of skill because the decisions involve skill. This argument falls apart for reasons as mentioned earlier.

[EGO]

I posted this idea in another thread, but it's dropping down the page fast, so I thought I'd regurgitate it here.

I've often heard that the standard deviation of a game is the measurement of luck in poker, and that it resolves fairly quickly to close to it's true value.

If luck can be measured, can't we quantify to what degree a reduction of luck implies skill?

I'm not a math geek, not even close. It just strikes me that if luck and skill are part of poker, and you can measure luck - then you should be able to measure skill by the difference in luck between a terrible player and an expert player.

Quote from TOP
[ QUOTE ]
...expert players do not rely on luck. They are at war with luck. They use their skills to minimize luck as much as possible.

[/ QUOTE ]

It was this quote that led me to this idea.

[popesc]

[ QUOTE ]

This again is not a valid argument. If this were the case, good players would win the vast majority of their sessions. I win just over 60% of my sessions...

[/ QUOTE ]

Winning 60% of your sessions is certainly significant, since no baseball team last year won 60% of their regular season games, and baseball is certainly a game that is mostly skill. (The winner of the world series last year only won 51.6% of their regular season games.) mlb.com

[SGspecial]

[ QUOTE ]
[ QUOTE ]
Is the white elephant in the skill vs. chance argument room the fact that ultimately, it's not about who wins the hand, but who leaves the table with the most money? That is, skill determines how much $$$ you extract from the other players, regardless of your cards; ergo, chance dealing you good or bad cards can be irrelevant.

[/ QUOTE ]

This again is not a valid argument. If this were the case, good players would win the vast majority of their sessions. I win just over 60% of my sessions and I'm a large winner at relatively small stakes. I'm sure the win rate for high stakes players, who are presumably the most skilled, quickly converges very close to 50%. And if 'leaving the table with the most money' proves skill, wouldn't the best players leaving the table stuck nearly 50% of the time therefore prove the predominance of chance?

[/ QUOTE ]

Slim margins do not imply a small amount of skill is needed for a game. All they imply is that there is a small DIFFERENCE in skill levels among the competitors. In a .05/.10 NL game players may all be equally (un)skilled and therefore no one has much of an edge. If Chris Ferguson sits down with $10, he is a huge favorite to win at the table because his skills are so much more advanced than the competition. If he were on HSP against Todd, Daniel, and Barry, he is no longer a big favorite, and perhaps an underdog.

[questions]

[ QUOTE ]
This again is not a valid argument. If this were the case, good players would win the vast majority of their sessions. I win just over 60% of my sessions and I'm a large winner at relatively small stakes. I'm sure the win rate for high stakes players, who are presumably the most skilled, quickly converges very close to 50%.

[/ QUOTE ]

How does that refute my argument that poker is not about winning hands, but winning more money than others at the table, i.e., busting their stacks? (By the way, the reason I referred to making money at poker as a white elephant is because it's obvious that's why we all play it, but nobody wants to argue that because that would be going down a road leading to questions about taxation.)

[ QUOTE ]
And if 'leaving the table with the most money' proves skill, wouldn't the best players leaving the table stuck nearly 50% of the time therefore prove the predominance of chance?

[/ QUOTE ]

If I were arguing this in court, I'd examine experts' win rates over not just one table against other experts, but play against experts CONTRASTED WITH play against amateurs. The resulting stark contrast would demonstrate skill in that while experts of similar skill consistently keep their chip stacks intact (minus rake) over time, against amateurs, they win big.

[Dire]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Is the white elephant in the skill vs. chance argument room the fact that ultimately, it's not about who wins the hand, but who leaves the table with the most money? That is, skill determines how much $$$ you extract from the other players, regardless of your cards; ergo, chance dealing you good or bad cards can be irrelevant.

[/ QUOTE ]

This again is not a valid argument. If this were the case, good players would win the vast majority of their sessions. I win just over 60% of my sessions and I'm a large winner at relatively small stakes. I'm sure the win rate for high stakes players, who are presumably the most skilled, quickly converges very close to 50%. And if 'leaving the table with the most money' proves skill, wouldn't the best players leaving the table stuck nearly 50% of the time therefore prove the predominance of chance?

[/ QUOTE ]

Slim margins do not imply a small amount of skill is needed for a game. All they imply is that there is a small DIFFERENCE in skill levels among the competitors. In a .05/.10 NL game players may all be equally (un)skilled and therefore no one has much of an edge. If Chris Ferguson sits down with $10, he is a huge favorite to win at the table because his skills are so much more advanced than the competition. If he were on HSP against Todd, Daniel, and Barry, he is no longer a big favorite, and perhaps an underdog.

[/ QUOTE ]

Again, you really have to be careful what you say. If, as you said, at an average low stakes the game the players will generally be similarly unskiled and thus the average difference in abilities will be small then chance will be the predominately deciding factor. Similarly, if the players are generally similarly well skilled at higher stakes games then the same holds true. The judgement isn't saying that poker isn't a game of skill, just that it's a game that's predominately decided by chance.

Now the expert versus argument idea to prove skill is interesting - but runs into the same problems as with earlier skill arguments. A blackjack 'expert' would obviously do much better than an 'amateur' blackjack player, but it doesn't change the game's classification as a game of chance.

[Skallagrim]

This really has become an intellectual exercise in dissuading you guys from equating skill with good play. If you can get over that you will be able to easily reconcile the points made in this thread, good and bad.

Is my golf swing NOT an act of skill because I have no idea of the correct way to swing having never played golf before?

I say it is an act of skill, just in my particular case it is very bad skill. Thus when I play an equally bad player, skill (how well I hit the ball) is still determining the outcome (other than, of course, the fluke things that do happen in golf), same as when I play Tiger Woods and he trounces me (unless I get that lucky bounce off a duck....) - its just that my skill is so bad you dont want to call it "skill" anymore because you associate the word skill with "good." Even though its "luck" as to which of us 2 equally bad golfers hits the better shot on any particular swing, its still who hits the better shot(s) that detemines who will win the game. Golf therefore is a game of skill.

Good skill, or in better words, good play, is indeed something that manifests over time, can be defeated by the chance to one degree or another, and is less of a factor against players with similar levels of play. I would be hard pressed to prove that "good skill" acocunts for a specific percent of poker results indeed. Just as I would be hard pressed to prove that Tiger Woods' obviously better skill accounts for a specific percentage of the strokes he beats me by.

But the Courts dont require us to prove the difficult thing you guys are trying to prove when you fail to see this distinction. In other words, the Courts dont require us to prove that GOOD PLAY accounts for MOST outcomes. They require us to prove that CHANCE DOES NOT ACCOUNT FOR MOST OUTCOMES.

And we can do that, we can prove that the actions of the players determine most outcomes, not chance.

I call player actions, whether good or bad, actions of skill. From now on here though, I am just going to call them player actions instead of skill so you wont confuse them with only good play.

Skallagrim

[EGO]

Some random musings on skill games.

It's like chess, where one player is more skilled than their opponent. The more skilled opponent will win most of these games. No one would suggest that chess isn't a game of skill, even though one opponent is less skillful than another.

There's even a rating system, with a formula that will show how often a player with a particular rating will defeat a person with a different particular rating, assuming both ratings are accurate.

Maybe poker needs a rating system. Not likely, I know. Maybe a play-money ladder system involving HU matches? If players can rate their expectation over the long term based on ratings differences, then it should be clear that poker is a game of skill.

I think the problem with this debate has been some ambiguity over the meaning of the word "skill". It's got lots of different meanings, and many different parts of speech. I'll throw out a couple of ideas for definitions for the word in the context that it seems to be used here.

Skill:
The ability, in some form of game that involves luck, to produce consistent results over a large sample size.

OK, so I've got only one idea for a definition, and it's certainly not very clear - but I'm just tossing out ideas. This seems to be the crux of Skallagrim's argument, though.

The thing that makes poker a skill game, is that players can play in such a manner that their results, over the long term, are different than what luck would dictate.

Here's an example from my PT database. Over 26,178 hands at .5/1, there's been $5485.30 raked from the pot. If everything else is equal, then everybodies long-term winrate be an even distribution of the rake, or -2.28BB/100 for each player in my database. Naturally, this is false.

However, many player's winrates might fall within the statistical "boundaries", that is - whatever their actual winrate, even over large samples, it is still feasable for them to be -2.28BB/100 players, even at fairly high confidence levels. That's just a guess... I'm not a math whiz.

Compare this to a game like roulette, or blackjack. A player who plays blackjack perfectly will be making (er, LOSING), over a large number of trials, pretty close to what the actual house edge. Same for roulette... it doesn't matter how you play it, but if you play long enough, then your loss will be close to the house edge.

This might raise a quandry, though. How many people actually play enough hands for luck to have less effect on their game than skill? If 50 million people from the US play poker online, but only 1 million people play enough to overcome luck, then it's still (effectively) a game of luck for 98% of the people involved.

Just for the record, I think poker is a game of skill - but I've played over 200K hands. It is a game of skill - for me. However, most people won't play that many hands in their lives (not counting the jokers here, that is.). For instance, only about 16% of the people I've played against this year have played more than 100 hands against me (that's 84% for people with less), compared to 6% for people that have played more than 1000 hands against me.

If a player doesn't play enough poker, then skill isn't as important as luck - and most poker players don't play enough poker to overcome luck.

Just some food for thought.

[Skallagrim]

Ah, long term, short term again, eh EGO?

"If a player doesn't play enough poker, then skill isn't as important as luck - and most poker players don't play enough poker to overcome luck."

For any given player, in any given short term, what you said MAY be true, but it also may not:

I am playing NLHE and I am not getting good cards (bad luck). Over and over again I look down at 2-7, 3-J etc... I fold a lot and am down by a number of blinds. Finally, I look down to see ... another 2-7. But this time, a guy who has been getting lots of good cards, is winning a fair bit, and I can tell is a perceptive player, makes a big move in front of me. Everyone folds to me (I am sitting one away from the cutoff). I put on an act and reraise even bigger. Everyone folds back to the intial raiser. He looks at me, sees my "tightness" and false determination, looks at his nice stack of chips, and, just like I figured, decides not to risk them against my first big raise of the evening. Then I stand up and leave an overall winner.

Where is the luck in that short term winning result?

My conclusion, any group of short term results MAY be due to chance OR it MAY be due to skill.

Skallagrim

[EGO]

Simarly, a player posts in the micros that they are losing 3bb/100 in 1000 hands, and get numerous responses of "sample size", and "anything can happen in 1000 hands."

I think that skill should be defined close to how I stated it, perhaps adding that the consistancy a player should show should be different than what random luck would determine (-2.28BB/100, in the games I play).

Your winrate for this sample session is probably nothing close to your overall winrate - many sessions are not. Your decision to pull a bluff here is a clear example of skill, and it's good one for the argument, since it should resolve itself fairly quickly into your winrate.

Let's take drawing to a gutshot in limit poker. It can be an awfully long time before you hit your gutshot (getting corect odds, of course). A player could play 30 or 40 or more of these, losing a huge amount that affects his winrate in the negative, or he could hit them a few times in a row affecting his WR in the positive. It doesn't matter which, and we both know that - because we will play enough hands for it to even out.

If I draw to a hand that's going to win 5% of the time, getting $24-$1, then I know that I'll be making a certain amount of money. (.05*24)-(.95*1) = .25 Sklansky$$.

However, if I only play this bet once, then skill doesn't factor into my winrate. I'm willing to bet that me drawing to two-outers in huge pots hasn't evened out to where it should be for those decisions EV-wise, even over 200K hands. Good thing it isn't a routine decsion, since my SD would be huge.

I guess I'm saying that there should be some sort of metric that determines skill. Let's falsely assume that poker is 100% luck - then over time everybodies winrates should be: 0-rake/#players. Since it's not, then it implies some level of skill.

Long term winrate, Standard Deviations, something. I don't see how we can prove that poker is more than 50% skill without quantifying how much luck is involved. It seems that figuring out the luck part should be easier.

I could be wrong.

Edit: I hope I'm wrong. I also see this has been discussed before, as referenced in the Poker is a Market thread, and that looking at it from this standpoint is not new, and probably not productive. Apologies for not researching thourougly.