Poker Players Alliance Forums » Poker Players Alliance

Visions of the Luckometer

(5 posts)
  • Started 1 year ago by Big Jim Slade v2.0
  • Latest reply from Big Jim Slade v2.0

No tags yet.


  1. Big Jim Slade v2.0
    Moderator
    Visit User Profile

    For years I've heard people debate over the concept of whether poker is a game of skill or a game of chance. I don't know the current state of the debate, but was wondering if anyone has heard if this issue has been put to rest or not?

    I was thinking that if only someone could devise a device of some type, a Luckometer to give it a name, to measure skill versus luck we might be able to get somewhere with the debate. I can envision such a theoretical device, the Luckometer, composed of some combination of rheostats, vacuum tubes, and other smoke filled components of the magical discipline known as electronics.

    The Luckometer would have properties of being able to determine the probabilities of events happening. For instance, if 100 people played in a poker tournament the Luckometer would have the capability of determining what was the random chance of one of the people making it to a place at the final table. If a number could be determined for this event – one in a hundred making it to the final ten, then the Luckometer would provide a baseline for measuring skill.

    Say for instance, the Luckometer determined that the chance of one player out of 100 making it to the final 10 was 10%, then we could postulate that this number could be used as a baseline for determining skill. The question to be determined is will this chance always be 10% for 100 players? Perhaps multiple readings of the Luckometer would validate this for us, but for now I will theorize that this is the case and will proceed with my argument based on this line of thought.

    If we chose one random player and followed his tournament career for 100 tournaments, then we might think he would make it to the final table about 10 times out of 100 tournaments. For a player who makes it to the final table less than 10 times, we can brand him with the arbitrary term of “unlucky”. For a player who makes it to the final table more than 10 times out of 100, we can brand him with the arbitrary term of “lucky”.

    In looking for a scientific explanation of “lucky” and “unlucky” we can determine an unknown talisman is producing this effect for the lucky, and some curse is effecting the “unlucky”. For now, we can label this unknown talisman for the lucky as “skill.”

    Using the Luckometer I can theorize we can create a listing of who makes it to the final table and who doesn't. This distribution can be mapped as a bell curve with the frequent losers to the left and the frequent winners to the right. The top of the bell will be those have an average amount of wins, possessing neither talismans or curses.

    I'm guessing that a science of some sort can be developed to determine probabilities. This new science of “probability” could be used to create another bell curve of expected values for people winning tournaments.

    Comparing the two bell curves could yield evidence of the existence of this previously unknown talisman, “skill.”

    Now to come up with a design for the Luckometer. How to design output from it? Please contatc me if you know where I can find some nixie tubes.

    Posted 1 year ago #
  2. Skallagrim
    State Director & Moderator
    Visit User Profile

    Hello again Big Jim,

    As the Litigation Support Director for the PPA I can tell you for certain that the "skill v. chance" debate regarding poker is very much alive, and very much going in our favor (knock on wood). We have advanced the skill argument in three separate courts in the last six months and we successful with it in all three (the 3 are South Carolina, Pennsylvania, and Colorado - although in SC the defendants still convicted as a result of a different disputed legal issue). Appeals are ongoing in all 3 states, the PPA is still directly involved, and news or developments regarding those cases is usually posted on the main PPA homepage.

    From the main PPA home page you can also read the briefs we have filed. I suggest you take a look at our most recent brief, the one we filed in Colorado, although it is primarily concerned with tournament poker (as that was the issue in that case) all of our basic arguments and all of the latest scientific studies are referenced in that brief. The specific link is:
    /ppa/2009/05/21/co-ppa-amicus-filing-for-raley-vs-state-of-colorado-052009/

    You can also check out older discussions of this issue over at my sub-forum "Ask Skallagrim" on the right hand side of this page. You might be particularly interested in the thread "Best way to prove poker is a game of skill."

    And if you want to read ALL the stuff we have collected or written on this issue, just go to the home page, click "news" and then "headlines" from the pop up menu, and you will find a long list with many articles, briefs, and press releases on this subject (and others of course).

    Thanks for your thoughts and input here Big Jim. But I think if you read through some of the scientific studies we report on in the places I suggested, you will see the scientists have already done a lot of what you are suggesting. Doesn't mean we cant do more however! So keep up the good ideas.

    Skallagrim

    Posted 1 year ago #
  3. Big Jim Slade v2.0
    Moderator
    Visit User Profile

    I'm actually not suggesting to prove that poker is a game of skill. I'm suggesting scientific evidence to prove that poker is not a game of chance.

    As with the Colorado filing, my suggested use of the Luckometer was for tournament poker. One thing I might not have made clear about the Luckometer - I was not suggesting to prove that poker is a game of skill. But rather I was suggesting the groundwork for a scientific test to prove that poker is not a game of luck.

    It is possible to model, exactly, what luck will be in a tournament and how it is distributed. It is then possible to take a corpus of tournament data and test that data to see if it follows the luck model.

    In the Colorado filing I see maybe two "scientific studies". One involving "robotic" simulated play which I was not specifically familiar with. The other was Paco Hope, Brian Mizelle & Sean McCulloch, Statistical Analysis of Texas Hold’Em at 5 (Jan. 28, 2009). I recall such a study based earlier this year based on hands at an online poker room - I assume this is the same. Both are interesting, but only go to add to a preponderance of evidence that poker is based on skill.

    You point me to your thread "Best way to prove poker is a game of skill." In perusing this thread, I see a response added three weeks ago by someone named "Big Jim Slade." Big Jim suggested, in the thread you proffer, that a mathematical test be preformed to test the null hypothesis - that is to test if a tournament is driven by luck.

    I submit that the distribution of final table winners will fail a chi square test and therefore prove that poker is not driven by chance. (This technically does not prove it is driven by skill - it could be the rings of Saturn - but we can prove SCIENTIFICALLY poker is not a game of chance.)

    To the same degree that expert testimony in court can prove ballistics or DNA tests, I'm suggesting that expert testimony can be given that scientifically proves that poker is NOT a game of CHANCE.

    BTW - the paper showing how 76% of hands are folded was a bit wimpy on the science side. Even if it is more esoteric, I'm suggesting something with a bit more meat.

    If only we could find a person who knows about poker and has a PhD in Mathematics. I'm thinking for expert testimony purposes a PhD could prove to a scientific certainty that chance is not the driving factor.

    Posted 1 year ago #
  4. Skallagrim
    State Director & Moderator
    Visit User Profile

    Well Big Jim Slade, I think you need at this point to have the discussion directly with some mathematics professors. I will see if I can get any to do so (they usually charge for this sort of stuff, unfortunately).

    My training is as lawyer and so that is how I approach issues.

    In my non-mathematically trained mind, I figured that the computer simulation study showing how a program manufactured to act randomly will lose consistently (97% of the time) to a program given even the most basic of betting skills was pretty much similar to what you suggested. Yet this is going well beyond my area of expertise, so I admit I could well be wrong.

    Perhaps you could try your point again in a manner I can send to one of the experts we have had testify for us. Thanks for you effort.

    Skallagrim

    Posted 1 year ago #
  5. Big Jim Slade v2.0
    Moderator
    Visit User Profile

    As to your suggestion that I have the discussion with a Mathematics professor, I am unsure if you are saying *you* would benefit from me talking to *your* professor, or if you are saying *I* would benefit from talking to such a person. Lest you misunderstand me, I'm not asking for advice on how to construct the Luckometer. But I would be happy to talk to your professor if you wish. Hey, I'll even talk to any number of your associated professional poker players who also have PhD in mathematics. Several of them have admitted to me they do have such a degree.

    Since your training, Skallagrim, is as a lawyer, I'll approach it from a lawyer's point of view.

    I recall once going to Federal Court about a technical issue. After interviewing and selecting a group of representative plaintiff's (we were suing the government) I then gathering "expert" witnesses. The term "expert witness" has a specific and special meaning in court. I used a professor from the local technical college in the area. In my estimation, college professors make good expert witnesses.

    I'd recommend having a Luckometer study introduced as "expert evidence" to which the local university Mathematics Phd can render an "expert opinion" on. The expert witness will have to satisfy Fed. R. Evid. 702. Under 702, the expert witness will have to have "scientific, technical, or other specialized knowledge" and be able to "assist the trier of fact," most likely a judge in this case.

    The actual study will be performed by one or more of the professor's students, with the professor putting his name on it. You will need to decide if you want the same Phd to author the study as the one who testifies on it in court.

    The premise I am suggesting is that when dealing with large numbers, probabilities can be modeled precisely. To get more specific, "natural randomness" can be modeled (which actually varies slightly from casino randomness). The Random Number Generators (RNG) at online poker rooms do in fact produce distributions of "natural randomness" as opposed to the randomness at a casino card table.

    The study would first model the distribution of winners in a large number of poker tournaments of "natural randomness." Then the study would take the results of winners from a large number of tournaments and and see how they two distributions compare.

    There are precise mathematical tests to determine if actual events fit the proscribed pattern of natural randomness. They won't because poker tournament outcomes are not based on chance.

    If we were looking at spins of a roulette wheel things should match up perfectly. In roulette, about 1/3 of all players who play for less than two hours will walk away from the table with more money than they started with. Based solely on chance, you expect to see some winners.

    In poker you will see some winners based on chance. But in the long run, with poker you will huge aberrations of people who win an impossible number of times to allow chance to account for it.

    There are specific mathematical tests that can be run to demonstrate that chance is not why they are winning. The tests will not explain why some players win so much, by it can be proven to a scientific certainty that is is not due to luck.

    Posted 1 year ago #

Reply

You must log in to post.