Please turn on javascript in your browser to play chess.
Posers and Puzzles

Posers and Puzzles

  1. Subscriber sonhouse
    Fast and Curious
    18 Jan '08 01:12
    I saw this Jeopardy show the other day, at the end player #1 had 13,000,
    player #2 had 10,200, and player #3 had 10,000.
    Is there a betting strategy that would give anyone an edge? If #1 doubles, and misses, he is at zero, but if he wins he has more than anyone else but say if he bets zero, he has 13K at the end no matter what he answers so the other players knowing that, what is their best bet?
  2. 18 Jan '08 01:30
    I know nothing about Jeopardy. Can you bet as much as you like? If so the first guy should bet 279 bucks then if he fails he's still in front but he's forcing the others to risk more money if he wins I guess.
  3. 18 Jan '08 02:14 / 3 edits
    Originally posted by sonhouse
    I saw this Jeopardy show the other day, at the end player #1 had 13,000,
    player #2 had 10,200, and player #3 had 10,000.
    Is there a betting strategy that would give anyone an edge? If #1 doubles, and misses, he is at zero, but if he wins he has more than anyone else but say if he bets zero, he has 13K at the end no matter what he answers so the other players knowing that, what is their best bet?
    player #1, since he has the most money, MUST bet enough money to allow for himself to win if he gets the answer correct, but should generally only bet JUST enough to go above the next highest possible value (double player #2). in this case, player 1 would bet 7401. This gives him the best possible scenario in case everyone gets their response wrong as well.

    differing strategies now present themselves for player #2 and #3, depending on whether they trust player #1 to bet the "correct" amount. that is, they should assume that player #1 will put himself in the best possible position to assure his victory if he gets the question right. the following ideas are based on that assumption, and may not work in practice, but are my personal takes that i can logically defend.

    if player #3 were out of the picture (as is sometimes the case) i would, as player #2, bet just enough to go ahead of player #1 (this is dependent upon our relative scores). So I COULD bet 2801. This FORCES player #1 to get the right answer to beat me if i get the right answer, and assures my victory over him if we both get it wrong.

    however, since player #3 is very much "in the fight" since his score is comparable to the other players... player #2 (in my opinion) MUST bet enough to go ahead of player 3's score doubled. this makes player #2's bet at a minimum 9801. But then you lose the "double wrong answer" victory over player #1, and so you might as well bet to double your money (as player #1 has already assumed you would do).

    Now player #3 actually has the interesting choice, since player #2 must bet at least 9801 to play for a victory. if player 3 chooses to double his score, he gains little. his only chance to win is if the others get the answer wrong, and he gets it right. Instead, if player #3 bets 3001 (my preferred choice), he clearly wins if the others get it wrong and he gets it right, but he also wins if EVERYONE gets it wrong. he also wins if the other two make a strategic blunder and bet zero, (regardless of whether they get it right). So, if he gets it right, he is forcing the other players to ALSO get it right in order to reap the advantage they earned from the earlier rounds of play.

    I think many would tell me in practice this isn't a great choice, but I think that depends on the quality of strategist you are playing against, and CERTAINLY depends on the relative scores of the three players. The easier method is definitely to just try and double your money to secure the highest possible dollar amount available to your score... but I think you throw away the very plausible "victory by default" if final jeopardy is a real stumper.

    This is all based upon "optimal" choices, and clearly could fail if any of the players either make a miscalculation; or if they fail to "play for a victory," instead assume they will get it wrong, and try to position themselves in such a way that will benefit them if everyone gets it wrong.

    any comments/other strategies? i always enjoy analyzing these things, and frankly am befuddled by players who make strategic blunders in final jeopardy. it happens far too often that the person who was in first place doesn't bet enough to win, and then there is an easy final jeopardy, so the person loses.
  4. Subscriber coquette
    Already mated
    18 Jan '08 06:58
    i don't see how anyone could explain the strategy more clearly. nicely done.
  5. 18 Jan '08 07:24
    I've noticed that quite often the scores going into final Jeopardy are something more like 10000, 9000, and then 2000.

    If I were in the lead, I would be confident that I would get Final Jeopardy correct (I feel this would be a safe assumption since if I were in the lead going into Final Jeopardy I must be pretty smart) and bet 8001.

    What I find interesting is what the person in second place bets. I would bet based on the assumption that the person in the lead did as I would have above. Therefore, the only way the person in second place can win is if the person in the lead gets Final Jeopardy wrong. So I would bet 4999. That's enough to pass the leader if I get Final Jeopardy correct, I would also pass the leader if we both got Final Jeopardy wrong, and would still have enough to stay ahead of the third place contestant, regardless of their answer.

    It astounds me how often the person in second place doesn't follow this strategy. But then, there's probably a lot of pressure, so I would probably screw the whole thing up too!
  6. 18 Jan '08 09:52
    Originally posted by smomofo
    I've noticed that quite often the scores going into final Jeopardy are something more like 10000, 9000, and then 2000.

    If I were in the lead, I would be confident that I would get Final Jeopardy correct (I feel this would be a safe assumption since if I were in the lead going into Final Jeopardy I must be pretty smart) and bet 8001.

    What I find interes ...[text shortened]... then, there's probably a lot of pressure, so I would probably screw the whole thing up too!
    i completely agree - my earlier strategy was based on the case where person #3 has a score comparable to that of person #2.

    In your case you've bet just the right amount with 4999 - if you win you maximize your dollar output, but you assure yourself second place if the leader guesses correctly (which you must assume he will bet enough to win), and most importantly if the leader guess WRONG, you will win regardless of whether you are right or wrong.

    in practice, so many people (who are smart enough to be the leaders after single and double jeopardy) in first place don't play to win! they often bet less than 8001 and leave the door open for a 2nd place player to take the lead if he/she doubles their money! that would totally make me second guess my "strategically more sound" choice of betting 4999 if i worried that the person in first place might not play to win.
  7. 18 Jan '08 09:52
    Originally posted by coquette
    i don't see how anyone could explain the strategy more clearly. nicely done.
    thanks coquette!
  8. Subscriber sonhouse
    Fast and Curious
    18 Jan '08 10:38
    Originally posted by Aetherael
    thanks coquette!
    Kudo's from me too!. Well done.
  9. Standard member PBE6
    Bananarama
    18 Jan '08 19:03
    Different perspective (but only for a 2-player game):

    http://wizardofodds.com/askthewizard/tvgames.html
  10. 18 Jan '08 20:17
    Originally posted by PBE6
    Different perspective (but only for a 2-player game):

    http://wizardofodds.com/askthewizard/tvgames.html
    really cool analysis - would be interesting to keep a pseudo-random number generator on your person for use right before final jeopardy haha. though from an egoist perspective, i would not want to rely on the law of large numbers for a specific one-day appearance on jeopardy - if i were the leader i'd always play to win if i got final jeopardy right. even if in the long run, it produces slightly worse results... i guess it all comes down to whether you consider yourself better than or worse than the "average" contestant when it comes to answering final jeopardy