Welcome.

Take a tour. Enjoy some free sample content.

How it works

Free Video: CLP Video No. 287: Home Game Bart Reviews His Splashy At $1-$3 Deep Part 2

Free Podcast: CLP Podcast No. 54: Time Warp And Turn Value
New to Crush Live Poker?

Article 116: Embrace Variance and the chess analogy

jeffnc Posts: 57Member
edited March 2014 in Crush Live Poker Articles
Many poker players like to make this analogy. While some of these players seem to actually be chess players, they don't seem to understand how chess actually works.

Here are some quotes from the article.

"The purest example of a game with no variance is chess."

This is simply false. In fact, the formula that calculates a chess player's rating is more or less the very definition of variance. In a typical chess game, you are playing a player with a rating within 100 points of yours. If you're playing someone 50 points below you, you have a 35% chance of winning and 21% chance of losing (the rest is chance of a tie.) At 100 points, it goes to 46% and 18%. (The draw percentages are actually a little misleading, since games among lower rated players end in draws much less often than games among higher rated players.) In fact, meeting certain percentages is how your rating is determined. In other words, the entire rating system is based on variance. It's not more unlikely that you lose 3 games in a row against an equally skilled player than it is that you flip 3 heads in a row (not accounting for ties.)

"If you played Gary Kasparov and you were not an international grandmaster, it would be next to impossible for you to win unless he made an uncharacteristically large blunder."

Well, yeah. This is about as interesting as saying "If you flop the bottom end of a straight flush in Holdem, it would be next to impossible for you to lose. There is basically no luck and no variance in the game." In that scenario, true. But how realistic is that scenario? How much of your actual playing time does that scenario represent? How many players that are not international grandmasters has Kasparov even played in the last 20 years?

"There is basically no luck in the game."

This is another misconception. Just because chess is a game of complete information, it doesn't mean that the players can actually employ that information. Theoretically, it can be done. However no human or computer has accomplished that feat.

Until everyone can, chess will remain a game with luck, variance, and yes... even bluffing.
«13

Comments

  • BartBart Posts: 5,955AdministratorLeadPro
    While I understand the concept of winning probability based upon rating--and there can be some predictability variance--I still fail to see the luck involved with chess unless you are talking about the luck of your opponent making a blunder.

    Bart
  • JasonPadgett Posts: 10Subscriber
    I came in here to post what Bart said. I don't understand how there can be variance in a game with complete information.
  • ThehammahThehammah Posts: 7,086Subscriber
    My issue w variance is in the big
    Pots. There my variance has been brutally negative for years. And yes it really bugs me. The sample is small because u dont get into 400 bbs pots often but this is another aspect of variance. I am probably winning more than my equity in the smaller pots and losing in the bigger ones.

    Well all i can do is keep plugging away and hopefully it should geavitate to the norm eventually.
  • eyedunno Posts: 215Subscriber
    As someone who played tons of chess from the ages 14-20, I can tell you the main reason I stopped playing regularly is because I got to a level (around 1800 rating) where I couldn't for the life of me beat players rated 1875 and up while players rated around 1700-1750 simply lost to me nearly all the time. It was frustrating that no matter how hard I studied and worked on my game, a certain level player was simply beyond my understanding to the point that I gave up.

    Is there some variance? Yeah I guess. But IMO comparing it to the variance seen in poker is pretty laughable.
  • JCW Posts: 591Subscriber
    edited March 2014
    What we have here is a basic fallacy of logic. It is called Equivocation. It is defined as the fallacy of the same word being used with two different meanings. And to be perfectly honest, I had to look it up in Google, because well, college was a long time ago for me.

    Here is how it breaks down in this example:

    Chess
    1. A game played between 2 people using a board, pieces and yada yada yada.
    2. A competitive game where people are ranked in a system of yada yada yada.

    So to really see the Fallacy of Equivocation it is easy if you re-write it as... there is no variance in chess1. But this is not true because when you are playing chess2 your opponent...

  • AesahAesah Posts: 1,048Pro
    jeffnc said:

    Well, yeah. This is about as interesting as saying "If you flop the bottom end of a straight flush in Holdem, it would be next to impossible for you to lose. There is basically no luck and no variance in the game." In that scenario, true.
    You're comparing it to something arbitrary (the chess equivalent would be like "if the other guy started with only pawns, you will probably win").

    The purest comparison is if a rookie were to play Phil Ivey in a 100bb HU freezeout, he has a chance of winning, whereas a rookie has close to zero chance to beating even like a 1500 rated chess player.
  • jeffnc Posts: 57Member
    Bart said:
    I still fail to see the luck involved with chess unless you are talking about the luck of your opponent making a blunder.
    It's more complicated than that. For example, you can happen to use an opening that your opponent just happens to know extremely well. Or not at all. Let's say we rate every move you make on the chess board on a scale of 1 to 10, with 10 being the best you can possibly achieve given your skill level, and 1 being the dumbest blunder you can imagine. You and your opponents are constantly making moves between 1 and 10. If you or your opponent happen to see something brilliant on some particular move, or miss something obvious, it results in much variance.

  • jeffnc Posts: 57Member
    JasonPadgett said:
    I don't understand how there can be variance in a game with complete information.
    The fact that chess is a game of complete information is a red herring. Since no human being or computer can actually utilize all that information, luck starts to play a large role. Sometimes you or your opponent will see a mate in 2, and sometimes you won't, for example. It's a skill game like poker, but with quite a bit of variance. This is the reason the rating system works the way it does. If it didn't, then you would always lose to someone rated higher than you.

  • jeffnc Posts: 57Member
    eyedunno said:
    As someone who played tons of chess from the ages 14-20, I can tell you the main reason I stopped playing regularly is because I got to a level (around 1800 rating) where I couldn't for the life of me beat players rated 1875 and up while players rated around 1700-1750 simply lost to me nearly all the time. It was frustrating that no matter how hard I studied and worked on my game, a certain level player was simply beyond my understanding to the point that I gave up.

    Is there some variance? Yeah I guess. But IMO comparing it to the variance seen in poker is pretty laughable.
    I might also argue that your particular issue is related to a single chess player in the world, which is a bit laughable. My point is not that poker variance and chess variance are the same. My point is that whenever someone says there is no luck or variance in chess because chess is a game of complete information, they are wrong.

  • jeffnc Posts: 57Member
    Aesah said:
    You're comparing it to something arbitrary (the chess equivalent would be like "if the other guy started with only pawns, you will probably win").
    I was simply comparing it something with approximately the same likelihood, that's all.
    Aesah said:
    The purest comparison is if a rookie were to play Phil Ivey in a 100bb HU freezeout, he has a chance of winning, whereas a rookie has close to zero chance to beating even like a 1500 rated chess player.
    If you want to make that analogy, then you have to compare apples to apples. Luck and variance occur in games like chess and golf, but the timeframe is just different, that's all.

    If you want to compare an amateur beating Kasparov (or let's say Carlsen) in one game, then the equivalent would be to play Phil Ivey heads up for about a month.

    Just because the games even out over different time intervals does not mean that a game of complete information doesn't have luck or variance in the real world.

  • khalwatkhalwat Posts: 997Subscriber
    What exactly quantifies the "luck" element of chess?
  • eyedunno Posts: 215Subscriber
    jeffnc said:


    If you want to make that analogy, then you have to compare apples to apples. Luck and variance occur in games like chess and golf, but the timeframe is just different, that's all.

    Maybe poker players are just biased and jaded because of how big the luck factor can be in any one hand, session, day, month, year etc....

    Are there any examples in the history of a game like chess where a top player consistently lost to an inferior opponent?
    khalwat said:
    What exactly quantifies the "luck" element of chess?
    This...
  • BartBart Posts: 5,955AdministratorLeadPro
    edited March 2014
    I think that this goes back to what JCW posted. The "game" of chess has no luck or variance in it but when you are playing vs an opponent that will make mistakes and has a level of variance in choosing the correct move you can interpret that as luck. The "game" of poker certainly does a high degree of luck involved with it.

    But to be honest, I get these types of non topical responses to my articles on twitter all the time. What does the chess analogy have to with the poker message that is being conveyed? It seems like you came on here just to argue to prove that you were smarter than everyone else who may not know chess as well as you. Is there variance when me with a 1200 rating plays against someone with a 2600? Maybe. I might win 1 out 5,000,000 or whatever the correct number is. But if I play against the best in the world HU in a 200BB blind match I may have a 45% chance of winning. So yes, poker and chess both have luck by your definition. It is just that poker has an unquantifiable amount, almost infinitely times more when two players confront of disproportionate skill levels which allowed me to comfortably say that chess has no luck in it as compared to poker.
  • AesahAesah Posts: 1,048Pro
    CLP chess4rollz tourney anyone???
  • NicholasK Posts: 237Member
    edited March 2014
    i think that the game of chess is pure skill but if you were to bet on chess matches you'd run into variance due to the added unknown of the human element. Ivey has a b game too so because poker involves luck as well as skill there is an exponentially higher rate of variance involved.
  • JCW Posts: 591Subscriber
    edited March 2014
    In chess, the game, when you move your Rook to check the opponent you check the opponent. You do not have a 80% chance to check you opponent.

    In chess, there is no "Fog of War". When you move your Rook to the 7th row you do not discover after the fact that the piece was hung. In poker, when you have KK on the button the blinds combined have roughly a 2% of having AA. You can go all-in only to discover that this was a mistake.

    If you consider that a player playing below their expect performance is variance, then I would argue that you have weaken that word to a point of being meaningless. Then all the world is variance and we have to make a new word to make the distinction between games like chess and poker.


    As far as the chess ranking system, and matching system, etc... Yes that has tons of variance.

    by 1Aesah
  • jeffnc Posts: 57Member
    eyedunno said:
    Maybe poker players are just biased and jaded because of how big the luck factor can be in any one hand, session, day, month, year etc....
    Yes, the timeframe in poker is the main thing that distinguishes it from other games. The "long run" is much longer in poker than many other games. This is what makes it difficult to understand.
    eyedunno said:
    Are there any examples in the history of a game like chess where a top player consistently lost to an inferior opponent?
    Depends on what you mean by "consistently" - what timeframe. Is there anyone who has "consistently" beat Tiger Woods in his first 15 years as a pro? Is there anyone who has beat Tiger Woods on one golf hole?

    The luck in chess is the same as the luck in golf or the luck in poker. It can take many forms. At the Masters last year, Tiger hit a near perfect shot on #15 - far more precise than anyone else had. And yet it bounced off the flagstick and into the water and he bogeyed instead of eagled or birdied. That's one kind of luck. Another kind is simple variance. Golfers birdie holes all the time that Tiger has only parred, etc etc etc It doesn't mean they're better, it means they got a little lucky that hole.

    Luck in chess occurs when your opponent doesn't see something he usually sees, or gets into a line he's uncomfortable with just by coincidence, or gets into a line he just happens to be an expert at just by coincidence, or gets distracted and blunders, or forgets to hit his clock and lets time expire, or........

  • jeffnc Posts: 57Member
    edited March 2014
    Bart said:
    But to be honest, I get these types of non topical responses to my articles on twitter all the time. What does the chess analogy have to with the poker message that is being conveyed?
    I would argue that if it has nothing to do with the message, then it didn't belong in the article. If it has a little to do with the message, then to that little extent my comment is relevant.
    Bart said:
    It seems like you came on here just to argue to prove that you were smarter than everyone else who may not know chess as well as you.
    Don't be silly Bart. I see the chess analogy many times when people are trying to explain poker concepts. It happens to be a bad analogy and I'm just making that comment. It doesn't reflect on quality or usefulness of your articles, nor of my intelligence. It's simply a misleading argument, regardless of good intent.
    Bart said:
    Is there variance when me with a 1200 rating plays against someone with a 2600? Maybe.
    Almost zero variance. But then, 1200 players don't often play 2600 players in the real world. Again, this would be similar to a pro golfer playing a 15 handicapper. The 15 handicapper will basically never win over 18 holes. But he will almost certainly eventually beat the pro on some individual putt at some point.

    That's why I keep making the comment about timeframe. A heads up freezeout match is perhaps equivalent to 1 putt in a golf match. Maybe playing heads up poker for a year would be about equivalent to a round of golf. This does not imply that there is no luck or variance in golf or chess, just that it gets to the long run quicker.
  • jeffnc Posts: 57Member
    NicholasK said:
    Ivey has a b game too so because poker involves luck as well as skill there is an exponentially higher rate of variance involved.
    Phil Ivey once mucked the winning hand when he saw his opponent's hand at showdown. At that point in time, the game was a game of complete information. This sort of thing happens far, far more often in chess than in poker - where someone incorrectly evaluates the situation. In fact in chess, it happens on practically every single move for lower ranked players. The lower your rating, the more luck plays a factor in the game.

  • jeffnc Posts: 57Member
    edited March 2014
    JCW said:
    In chess, the game, when you move your Rook to check the opponent you check the opponent. You do not have a 80% chance to check you opponent.
    ???
    JCW said:
    In chess, there is no "Fog of War".
    You may be speaking purely theoretically, but as chess is actually played, until chess is solved, this is untrue. And even if chess actually gets "solved" by computers, it will still be untrue in human games. In fact, when chess experts evaluate a position, they even have a special symbol (the infinity symbol) which means after their best evaulation, the position is "unclear". That is the "fog of war" equivalent in chess.
    JCW said:
    When you move your Rook to the 7th row you do not discover after the fact that the piece was hung.
    Actually, that is precisely what happens in games every single day.
    JCW said:
    If you consider that a player playing below their expect performance is variance, then I would argue that you have weaken that word to a point of being meaningless.
    That is not what I mean by variance. Each game of chess is not the same. Some games you get into positions or situations that play to your strengths, and some games you don't. This is not by any fault of yours, but just by what your opponent does, or just the randomness of the game. It's not until you have played many games of chess that these situations even themselves out and you can get an idea then of your actual performance level.

    By definition, it's impossible to play to your performance level in 1 game of chess*. The rating system proves that. Just like it's impossible to flip a coin 1 time and match the EV of the event, or draw to a flush 1 time and match the EV of the event.

    *The only exception I can think of is the rare situation where you happen to be playing someone of the exact same rating, and you happen to draw. Maybe this would be like flipping a coin and it lands on its side :-)
Sign In or Register to comment.