great 2 articles from Annie Duke here read it..
Skill vs. Luck in Poker
I had a very perplexing and heated debate with a poker pro a few weeks ago about the skill vs. luck argument in poker. The gist of the debate had to do with my making the assertion that poker is a game of skill. I know it sounds weird that that would cause debate with another pro. For sure, the general public might argue the point with me. But a poker pro?
The pro basically asserted the following: Yes, for winning players poker is a game of skill because they are good at determining their mathematical expectation and, thus, are playing with an edge. But most people are losing players. There are always people on the losing end of the transaction and those people are not playing a skill game but are playing to get lucky, playing at negative expectation. I know that on the surface his argument might seem compelling. Sure, if someone is playing at a negative expectation they might be trying to overcome that expectation and win anyway. They rate to lose and they are trying to thwart their EV. That I agree with. It is the leap that those players are not playing a game of skill, that for players playing poker at a negative expected value
that poker is a luck game, that I don't agree with. In fact, I couldn't believe that a pro would ever claim that just because a player is not an expert that they are playing a luck game; that it is only pros who are playing a game of skill. Frankly, it is absurd to assert that the skill level of a player in any game, much less poker, has any effect on whether the game itself is a game of skill or not.
Take golf as an example. I think everyone can agree that golf is a game of skill. And the game has specific skill elements that we can identify, like putting and driving. As well, we can all agree that a player like Tiger Woods applies those skill elements very adeptly. Now what about a golfer who shoots a 120? Clearly, that golfer applies the skill elements very poorly. He is a poor putter and a bad driver of the ball. If we pit Tiger against this bad golfer, Tiger would win every time.
Here is the question I have for you, "Is the game of golf itself more of a game of skill when Tiger is playing it than when the 120 handicap golfer is playing it?" The question sounds completely absurd when I ask it about golf. Of course golf is a game of skill whether or not someone applies those skill elements well or poorly. The quality of the player has no bearing whatsoever on the absolute fact that the game is a skillful one.
What is interesting is that while that logic is so completely transparent to people when it comes to a game like golf or football or baseball or bridge, it is not completely transparent when it comes to a game like poker. I think because there is wagering involved. But think about it, the logic is exactly the same as the logic for the golf example. We can all agree that there are skill elements in poker like whether you bet, fold or raise at any given decision point, or how much you bet or raise if you make that decision. These are elements of skill and how well a player executes these skill elements determines the likelihood of winning or losing. As in golf, we can all agree that the top players, players like Erik Seidel and Phil Hellmuth, apply these skill elements very adeptly. But what about a player new to the game? Obviously, a new player to the game would apply these skill elements poorly, just as the 120 golfer putts poorly. And just as with golf, the fact that a new player applies these skill elements poorly has no bearing whatsoever on the absolute fact that poker is a game of skill. A player who is a poor decision maker at the table is, to be sure, executing the skill elements of the game of poker very poorly but even when a bad player is player the game, he is still playing a game of skill. He is just not playing it well.
So a poker player can be playing at a negative expectation and still be playing a game of skill. That there are bad players in the world actually reinforces the argument that poker is a game of skill in this sense: In a game of pure luck, like Baccarat
, a player cannot be bad or good. The only decision in that game is whether to be player or bank and both decision have the same expectation. So one player cannot be worse than another since the only decision in the game that affects the outcome is neutral to the player's expectation. In a skill game like poker, on the other hand, one player can be worse than another because there are skill elements, multiple decision points that do have a direct effect on a player's expectation. In a pure luck game like baccarat or the lottery, no player can be worse than any other. Only in games of skill can there even be a tangible difference between players. Poker is one of those games.
IS THERE REALLY ANY LUCK IN POKER?
My brother, Howard Lederer, and I have recently been having some really interesting discussions about the luck factor in poker. Now we all know that there is skill in poker but the general consensus has been that there is a preponderance of skill, not that poker is a game that is all skill. Arguments for the preponderance of skill have centered on the fact that good players, in the long run, will come out winners but in the short run anyone can win. So the argument has been that poker is a game with a significant luck factor in the short run but over the long run the law of large numbers will play out and the better players will win.
But is this really true? Is there really any luck in the short run at all? Howard has come up with a very compelling argument that the answer to that question is no.
Let's say we program a machine so that it knows the rules of Texas
Hold'em. It knows that you are dealt two cards. It knows that a flop, and a turn and a river comprise the community cards. It knows that you can check, bet, call or raise on any given street. It knows the rules and mechanics of the game. But let's also say that we program the machine to play with no skill at all. This means that the machine will randomly choose at any given decision point whether to check, bet, call or raise. Now remember, on any given street there are up to 5 possible decisions (a bet and four raises) and our machine will behaving randomly, with no skill, at any of those decision points. How would the machine do? Terribly, obviously.
If you put our machine into a short event, like a sit and go, against 8 skilled players it would lose every time. The skilled players would quickly come up with the most effective strategy against the machine, which would be to raise the minimum against the machine every time. This would always put the decision back on the machine for the lowest risk and 1/3 of the time the machine will fold, 1/3 it will call and 1/3 it will raise, regardless of its hand. And the machine will behave this way regardless of its hand. It will be as likely to fold Aces full as it will to fold 9 high. It will be as likely to call with top pair as I will to call with 5 high. You can see pretty quickly that our unskilled machine would never win, even in the short run.
Howard's argument shows that poker players tend to drastically over estimate the luck factor in poker, mainly because in general we are playing against very skilled players and whenever we close the skill gap between opponents in a skill game it appears that there is more luck involved. We can take baseball as an example to demonstrate this. No one argues that baseball is not a game of skill. And the same thing happens in baseball when we narrow the skill gap that happens in poker. If we take the Yankees and pit them against a little league team where the skill gap is very large then the Yankees will win every time. But if we take the Yankees and pit them against an equally skilled major league team, say the Red Sox, now luck appears to play a much larger factor. Factors like injuries, weather, etc. now play a much larger role in determining the outcome of the game. While the better team will win over a series of games, the outcome of a single game will appear to be determined by luck, by factors outside of the control of the teams.
And poker is no different. Good poker players will overestimate the luck factor in poker because they forget exactly how skilled the opponents they are playing against are. The fact is that most players are very skilled at hand selection and betting theory that you play against, even in the smallest games, compared to a totally unskilled player like our machine. As in baseball, the more skilled the players are that you play against, the more there will be the appearance of luck in determining the outcome in the short run. To take the baseball analogy further, if we take the very best professional player in the world and stick them in a .50/$1 NL game, they will crush the game just as the Yankees will crush the little league team. If we take that same player and pit them against the other top pros, they will win in the long run, but the short run outcome will be largely determined by luck just as in baseball in a similar situation when the Yankees play the Red Sox the outcome will be determined by factors like injuries.
The interesting thing is that if we took our same unskilled machine and programmed it to know the rules of the lottery but gave it no skill it would perform the same as a human being. This is because there is no skill to the lottery. Once you know to purchase the ticket and fill out the appropriate number of circles on the ticket and give it back to the attendant to receive your ticket you are good to go. Once you know the rules there is nothing more to the game. And yet, lotteries are excluded in the current gaming legislation but poker is at risk. Seems illogical to me.
Poker is a game of skill. It is a game where the outcome is as determined by skill as baseball. Once we understand this, it is clear poker should be set aside from gaming legislation that legislates games of chance since it is clearly not a game of chance. It is just a matter of getting people to truly and deeply understand the difference between games of skill and games of luck.