But if he does that it will give false results for archived games.Bamaman wrote:Now Robert has to spend his weekend reprogramming the archive to allow for tiebreakers in regular games.
Ties on Jeopardy! RIP
Moderators: alietr, econgator, dhkendall, trainman
- opusthepenguin
- The Best Darn Penguin on the Whole JBoard
- Posts: 10130
- Joined: Thu Aug 11, 2011 2:33 pm
- Location: Shawnee, KS
- Contact:
Re: Ties on Jeopardy! RIP
-
- Loyal Jeopardista
- Posts: 242
- Joined: Fri Mar 28, 2014 6:59 pm
Re: Ties on Jeopardy! RIP
He could program it so the new rules apply only to regular games on or after 2014-11-24. He'd also need to reprogram or add a note to the wagering calculator tree since the suggestions for the breakpoint scenarios may no longer be accurate. There are also several definitions that'd need to be updated, and since this is a major rule change maybe also put a note in the prizes list (like there are for when sky's the limit, cash for the runners-up, and the doubling of the clue values took effect).opusthepenguin wrote:But if he does that it will give false results for archived games.Bamaman wrote:Now Robert has to spend his weekend reprogramming the archive to allow for tiebreakers in regular games.
Re: Ties on Jeopardy! RIP
I agree with golf, econgator, and opus. This is a stupid change. Even more stupid because the current tiebreaker rules (assuming they apply the same rules as in the tournaments) are anti-Jeopardy! in nature (nothing to lose by buzzing in).
Really, the only bug in the rules was that the possibility of a tie sometimes made it optimal to intentionally throw a late DD. But has that ever happened in an actual game? Even once?
Introducing unnecessary element of luck is a much worse bug. There is already a lot of luck in J!, but all of it as a necessary consequence of the way the game is fundamentally conceived. Forcing tiebreakers is a completely gratuitous addition of another element of luck.
It will be fun, though, to work out the consequences for wagering strategy. (But I'm sure glad I've already been on the show. I'd hate to get a major rule surprise when I show up for the taping, and try to work out the strategic implications in the Green Room...)
Really, the only bug in the rules was that the possibility of a tie sometimes made it optimal to intentionally throw a late DD. But has that ever happened in an actual game? Even once?
Introducing unnecessary element of luck is a much worse bug. There is already a lot of luck in J!, but all of it as a necessary consequence of the way the game is fundamentally conceived. Forcing tiebreakers is a completely gratuitous addition of another element of luck.
It will be fun, though, to work out the consequences for wagering strategy. (But I'm sure glad I've already been on the show. I'd hate to get a major rule surprise when I show up for the taping, and try to work out the strategic implications in the Green Room...)
- opusthepenguin
- The Best Darn Penguin on the Whole JBoard
- Posts: 10130
- Joined: Thu Aug 11, 2011 2:33 pm
- Location: Shawnee, KS
- Contact:
Re: Ties on Jeopardy! RIP
Dave Leach. January 3, 2012. He threw the 29th clue DD and dropped to half the leader's score. Fortunately for him, the 30th clue was a TS (or at least a DS with Dave wisely keeping his mouth shut), he got FJ correct, and the leader also got FJ correct but had wisely wagered nothing. Without the gambit, Dave retires as a 3-time champion. With it, Dave went on to become a 6 time champ who won his quarterfinal in the TOC but lost in the semis.gnash wrote:Really, the only bug in the rules was that the possibility of a tie sometimes made it optimal to intentionally throw a late DD. But has that ever happened in an actual game? Even once?
One more reason I think the new rule sucks.
-
- Loyal Jeopardista
- Posts: 242
- Joined: Fri Mar 28, 2014 6:59 pm
Re: Ties on Jeopardy! RIP
It'd be the same as the semifinals in a regular tournament, except that the winner wins the money and the 2nd/3rd place considerations.gnash wrote:It will be fun, though, to work out the consequences for wagering strategy. (But I'm sure glad I've already been on the show. I'd hate to get a major rule surprise when I show up for the taping, and try to work out the strategic implications in the Green Room...)
Re: Ties on Jeopardy! RIP
OK, based on his wager, I have to agree that he probably planned to tank the question, but I don't know if he actually did it. It isn't easy to remember the name of Nixon's AG. Plus he was lucky that the last question (very easy one IMO) was a TS. With the remaining clue in the 3rd row, it may not be advisable to play the drop-to-lock-tie gambit. If he hated the category, a wager of $100 (instead of $700) prevents him from being locked out, and baits the leader to go for the lock-tie.opusthepenguin wrote:Dave Leach. January 3, 2012. He threw the 29th clue DD and dropped to half the leader's score. Fortunately for him, the 30th clue was a TS (or at least a DS with Dave wisely keeping his mouth shut), he got FJ correct, and the leader also got FJ correct but had wisely wagered nothing. Without the gambit, Dave retires as a 3-time champion. With it, Dave went on to become a 6 time champ who won his quarterfinal in the TOC but lost in the semis.gnash wrote:Really, the only bug in the rules was that the possibility of a tie sometimes made it optimal to intentionally throw a late DD. But has that ever happened in an actual game? Even once?
One more reason I think the new rule sucks.
ETA: While the lock-tie situation gives Nicholas an incentive to stay clam on the last clue, Sally has nothing to lose by buzzing in and can break the lock tie (which in turn means that Nicholas isn't safe by clamming either).
Re: Ties on Jeopardy! RIP
Would you feel at all better if tied players split their winning total?gnash wrote:I agree with golf, econgator, and opus. This is a stupid change. Even more stupid because the current tiebreaker rules (assuming they apply the same rules as in the tournaments) are anti-Jeopardy! in nature (nothing to lose by buzzing in).
Really, the only bug in the rules was that the possibility of a tie sometimes made it optimal to intentionally throw a late DD. But has that ever happened in an actual game? Even once?
Introducing unnecessary element of luck is a much worse bug. There is already a lot of luck in J!, but all of it as a necessary consequence of the way the game is fundamentally conceived. Forcing tiebreakers is a completely gratuitous addition of another element of luck.
It will be fun, though, to work out the consequences for wagering strategy. (But I'm sure glad I've already been on the show. I'd hate to get a major rule surprise when I show up for the taping, and try to work out the strategic implications in the Green Room...)
Re: Ties on Jeopardy! RIP
I certainly would. It's an equitable solution, and would be very likely to eliminate ties except in rare circumstances.lieph82 wrote:Would you feel at all better if tied players split their winning total?
Re: Ties on Jeopardy! RIP
The motivation for this form of tiebreaker must have been a (misguided?) belief that tiebreakers will be exciting television. And yeah, the "based on what happens in the next ten seconds, one person will leave with $25K and the chance to win millions and one person will leave with nothing" is pretty exciting on its face, but it's hard for me to adopt that attitude knowing that J! once would have just paid both deserving players.alietr wrote:I certainly would. It's an equitable solution, and would be very likely to eliminate ties except in rare circumstances.lieph82 wrote:Would you feel at all better if tied players split their winning total?
Re: Ties on Jeopardy! RIP
The game is called Jeopardy! for a reason. I have no problem with adding one more jeopardy.
Re: Ties on Jeopardy! RIP
"A tie is like kissing your sister." - Navy football coach Eddie Erdelatz
Re: Ties on Jeopardy! RIP
I am not sure what makes a solution "equitable". It would be better than a tiebreaker because it wouldn't introduce an additional, unnecessary element of luck. It would also be sufficient from the producers'/S&P perspective because it would eliminate any incentive for collusion. It still seems like a solution in search of a problem, though. Like voter ID laws.alietr wrote:I certainly would. It's an equitable solution, and would be very likely to eliminate ties except in rare circumstances.lieph82 wrote:Would you feel at all better if tied players split their winning total?
-
- Loyal Jeopardista
- Posts: 242
- Joined: Fri Mar 28, 2014 6:59 pm
Re: Ties on Jeopardy! RIP
It looks like Ryan Alley and Allison Solomon will get the honor of being the last set of co-champions (like John Beck got the honor of being the last retired 5-timer).
- Silverfox
- Don't Step on My Tail!
- Posts: 279
- Joined: Sun Nov 16, 2014 1:27 pm
- Location: Deptford, New Jersey
Re: Ties on Jeopardy! RIP
Ties are OK with me. It makes the game more exciting and you get to see two players that you may like, the next day. Really how often does a tie happen? But I can see how some others may have different opinions on this.
It seems like a solution in search of a problem.
It seems like a solution in search of a problem.
Be alert. Stupid never takes a day off!
Re: Ties on Jeopardy! RIP
So for all the talk about Arthur Chu and The Final Wager, the last tie was not because of any sort of strategy, but because they both apparently picked a random number to wager out of thin air that happened to create a tie.GoodStrategy wrote:It looks like Ryan Alley and Allison Solomon will get the honor of being the last set of co-champions (like John Beck got the honor of being the last retired 5-timer).
- makeabeilein
- Jeopardy! Champion/Second-Best Performer
- Posts: 23
- Joined: Fri Nov 21, 2014 2:44 pm
Re: Ties on Jeopardy! RIP
Do you believe those were a pair of random numbers? I'll allow that they might have been, and no one can prove they weren't, but I personally don't believe that tie was organic. If nothing else, the optics of that week of shows are just awful, exactly because people like me can't believe that the second tie wasn't intentional — and intentionally fudged to appear unintentional. Multiple ties, one of them with 2 suspiciously-irrational wagers? It starts to look like the players are competing with the show, not each other.Dudd wrote:So for all the talk about Arthur Chu and The Final Wager, the last tie was not because of any sort of strategy, but because they both apparently picked a random number to wager out of thin air that happened to create a tie.GoodStrategy wrote:It looks like Ryan Alley and Allison Solomon will get the honor of being the last set of co-champions (like John Beck got the honor of being the last retired 5-timer).
(Full disclosure: I happened to be in Culver City this week, but this #hottake was fully-formed as soon as I saw Ryan mention on The Jeopardy Fan chat that he anticipated the death of ties.)
(Also: Even if the Ryan/Allison tie was indeed intentional, crafted through an understanding that the trailer would bet the nearest $1000, or something, I don't think ill of the tie-ers. I can't say how relieved I was to learn that I would never face the moral quandary of deciding whether to offer the tie or not.)
Re: Ties on Jeopardy! RIP
If you think the tie was fudged, you haven't been there to know just what lengths they go to in order to keep things on the up-and-up.
Sorry, guys…
This might surprise you, based on my public persona, but… [rips off mask] I really dislike ties.
There is only one situation – tied scores heading into Final – in which ties are truly unavoidable. The rest of the time – particularly in "wager-to-tie" situations like 2/3 and A = B + C – they're strictly a risk-mitigation technique.
The tied scenario is fairly uncommon, however: in J! Archive, 46 games have first and second tied heading into Final, and 48 have second and third tied in a non-lock game. (There's one overlap: the three-way pre-Final tie featuring ol' Jeeks.) There might be 1-2 players each year who get stiffed out of a co-championship in such a manner. That's a bummer, but them's the rules.
What I don't like is when a contestant gets to play again – and in some cases, continue a streak – because of the "generosity" of an opponent, or because two players have horrible wagers that end up on the same number. Not only does it inflate the number of wins, it also removes a spot for someone else. With several ties in a single Tuesday-Wednesday session, it could cause a logistics issue for Maggie, the contestant staff, and the players who get bumped, which could include some non-local contestants if enough ties happen.
Why now, though, and not at the end of the season? I think the producers know that ties are here to stay, because (thanks in large part to the publicity Arthur gave me) many contestants now use my site to prepare, and that's what I suggest they do. I've received a few messages from in-the-can contestants telling me several players at their tapings were planning to wager for the tie, and got thrown off a bit by the news that they were disallowed.
The only reason I advocated for the tie wager is because it is, mathematically, the best play under the rules as they existed, and I wanted to ascribe all of my techniques to game theory (thus avoiding the question: "Do I want to bring back a player with buzzer experience?").
As for a tiebreaker clue being used to decide the game, I don't have a problem with it. All of the players know the rules and can risk a tie at their leisure (or not, in the rare case of tied pre-Final scores). I'm sure the staff can do a better job of editing the game to make it appear less rushed than the tiebreaker in the last Teen Tournament.
There's actually one thing that makes me somewhat mad about the rule change: I now have to redo all of my tutorials.
There is only one situation – tied scores heading into Final – in which ties are truly unavoidable. The rest of the time – particularly in "wager-to-tie" situations like 2/3 and A = B + C – they're strictly a risk-mitigation technique.
The tied scenario is fairly uncommon, however: in J! Archive, 46 games have first and second tied heading into Final, and 48 have second and third tied in a non-lock game. (There's one overlap: the three-way pre-Final tie featuring ol' Jeeks.) There might be 1-2 players each year who get stiffed out of a co-championship in such a manner. That's a bummer, but them's the rules.
What I don't like is when a contestant gets to play again – and in some cases, continue a streak – because of the "generosity" of an opponent, or because two players have horrible wagers that end up on the same number. Not only does it inflate the number of wins, it also removes a spot for someone else. With several ties in a single Tuesday-Wednesday session, it could cause a logistics issue for Maggie, the contestant staff, and the players who get bumped, which could include some non-local contestants if enough ties happen.
Why now, though, and not at the end of the season? I think the producers know that ties are here to stay, because (thanks in large part to the publicity Arthur gave me) many contestants now use my site to prepare, and that's what I suggest they do. I've received a few messages from in-the-can contestants telling me several players at their tapings were planning to wager for the tie, and got thrown off a bit by the news that they were disallowed.
The only reason I advocated for the tie wager is because it is, mathematically, the best play under the rules as they existed, and I wanted to ascribe all of my techniques to game theory (thus avoiding the question: "Do I want to bring back a player with buzzer experience?").
As for a tiebreaker clue being used to decide the game, I don't have a problem with it. All of the players know the rules and can risk a tie at their leisure (or not, in the rare case of tied pre-Final scores). I'm sure the staff can do a better job of editing the game to make it appear less rushed than the tiebreaker in the last Teen Tournament.
There's actually one thing that makes me somewhat mad about the rule change: I now have to redo all of my tutorials.

Hate bad wagering? Me too. Join me at The Final Wager.
Re: Ties on Jeopardy! RIP
I remember watching the tie-breaker during one of the tournaments this year. (Can't remember which one.)
What sticks with me was that while I was still registering that there was a tie, slap-boom-bang the quickly announced the tie and tiebreaker, cut to AT reading the clue, then the answer and bam! it was over. It just happened so quickly compared to the rest of the show.
I'm sure they don't have much time left after FJ to air the tiebreaker, and boy did is seem abrupt.
What sticks with me was that while I was still registering that there was a tie, slap-boom-bang the quickly announced the tie and tiebreaker, cut to AT reading the clue, then the answer and bam! it was over. It just happened so quickly compared to the rest of the show.
I'm sure they don't have much time left after FJ to air the tiebreaker, and boy did is seem abrupt.
- Linear Gnome
- One Miner Gal
- Posts: 2006
- Joined: Thu Nov 15, 2012 9:55 am
- Location: Missouri
Re: Ties on Jeopardy! RIP
This would never happen, but I've been thinking of how you could work a Prisoner's Dilemma scenario into this. This is a first cut at the amounts, which might or might not be suitable.
I thought about this for a two-way tie but it seems easy enough to extend to a three-way tie. The tied players secretly vote to keep their share (half of the winning total) or play a tie-breaker. If both players vote to keep their share, they each get half and come back tomorrow. If one player votes for a tie-breaker and the other votes to keep their share, they both get half the money but only the one who voted for a tie-breaker comes back tomorrow. If both players vote for a tie-breaker, they play one for all the marbles. Edited to add: the vote would happen after seeing the category for the tie-breaker.
FWIW, I would prefer split-the-winnings-and-both-come-back to the new tie-breaker rule. I think it would disincentive ties enough that they would mainly happen either when they "should" or when it's sort of random and both players feel a little lucky to come back.
I thought about this for a two-way tie but it seems easy enough to extend to a three-way tie. The tied players secretly vote to keep their share (half of the winning total) or play a tie-breaker. If both players vote to keep their share, they each get half and come back tomorrow. If one player votes for a tie-breaker and the other votes to keep their share, they both get half the money but only the one who voted for a tie-breaker comes back tomorrow. If both players vote for a tie-breaker, they play one for all the marbles. Edited to add: the vote would happen after seeing the category for the tie-breaker.
FWIW, I would prefer split-the-winnings-and-both-come-back to the new tie-breaker rule. I think it would disincentive ties enough that they would mainly happen either when they "should" or when it's sort of random and both players feel a little lucky to come back.