A relatively simple statistical question

[skullturf]

Lots of data to help answer this question is already available in the archive, and maybe one or more of you has already done the number-crunching.

In what percentage of regular Jeopardy! matches does a defending champion win, and in what percentage is there a new champion?

People colloquially say that if you get on the show, you have a 2/3 chance of losing, which is close to being true in a broad overall way. However, I would guess that in more than 1/3 of games, the defending champion wins.

[slam]

I haven't looked at this recently, but when I studied it a few years ago, the defending champion won approximately 50% of the time. Presumably some of this advantage was due to familiarity with the signalling device and some of this advantage was due to probably being an above average contestant (as evidenced by at least one victory).

[the_phil]

Presumably some of this advantage was due to familiarity with the signalling device

Not as much as you'd think. There's a good chunk of practice before the tapings start in the morning, so as a contestant I don't think you're ever unfamiliar with it. You may start cold, because you haven't used it in a few hours, whereas the returning champ just got done 15 minutes prior. But with nerves and everything else happening, I'd say it's not much of a factor.

[Rex Kramer]

Presumably some of this advantage was due to familiarity with the signalling device

Not as much as you'd think. There's a good chunk of practice before the tapings start in the morning, so as a contestant I don't think you're ever unfamiliar with it. You may start cold, because you haven't used it in a few hours, whereas the returning champ just got done 15 minutes prior. But with nerves and everything else happening, I'd say it's not much of a factor.

Perhaps "familiarity" isn't exactly the right word, but I think slam's basic point -- that the confidence and lack of self-consciousness in using the buzzer that comes from experience makes a difference in gameplay -- is entirely accurate.

Rex

[slam]

There's another factor which follows from Rex's comment which I neglected to mention earlier. Simply being more comfortable out there due to experience should provide some advantage (but beware the "sophomore jinx"). That comfort applies to areas other than just the use of the signalling device.

And Rex is certainly correct that the word "experience" better describes the effect I was mentioning than "familiarity".

[cheezguyty]

In what percentage of regular Jeopardy! matches does a defending champion win, and in what percentage is there a new champion?

People colloquially say that if you get on the show, you have a 2/3 chance of losing, which is close to being true in a broad overall way. However, I would guess that in more than 1/3 of games, the defending champion wins.

Returning champs have won 46.6% of the regular-play games since November 6, 1993. 72.8% of contestants who appeared on the show during that time did not win.

[Edited to fix typo]

[pdano]

Presumably some of this advantage was due to familiarity with the signalling device

Not as much as you'd think. There's a good chunk of practice before the tapings start in the morning, so as a contestant I don't think you're ever unfamiliar with it. You may start cold, because you haven't used it in a few hours, whereas the returning champ just got done 15 minutes prior. But with nerves and everything else happening, I'd say it's not much of a factor.

Well, you only really get game-relevant buzzing experience when you get used to Alex's cadence, and you only get that through gameplay. My buzzing was pretty mediocre the first couple of games until something clicked, and then it turned into a big advantage. I think it's a pretty big deal.

[MDaunt]

Presumably some of this advantage was due to familiarity with the signalling device

Not as much as you'd think. There's a good chunk of practice before the tapings start in the morning, so as a contestant I don't think you're ever unfamiliar with it. You may start cold, because you haven't used it in a few hours, whereas the returning champ just got done 15 minutes prior. But with nerves and everything else happening, I'd say it's not much of a factor.

Well, you only really get game-relevant buzzing experience when you get used to Alex's cadence, and you only get that through gameplay. My buzzing was pretty mediocre the first couple of games until something clicked, and then it turned into a big advantage. I think it's a pretty big deal.

The buzzer is the game. Well, the buzzer and Final Jeopardy. Actually, the buzzer, Final Jeopardy and wagering strategy. Wait...the buzzer, Final Jeopardy, wagering strategy and knowing the airdate. So, the buzzer is definitely part of the game.

(But seriously, the buzzer is the game. Master that and nothing else matters.)

[TomKBaltimoreBoy]

Presumably some of this advantage was due to familiarity with the signalling device

Not as much as you'd think. There's a good chunk of practice before the tapings start in the morning, so as a contestant I don't think you're ever unfamiliar with it. You may start cold, because you haven't used it in a few hours, whereas the returning champ just got done 15 minutes prior. But with nerves and everything else happening, I'd say it's not much of a factor.

Well, you only really get game-relevant buzzing experience when you get used to Alex's cadence, and you only get that through gameplay. My buzzing was pretty mediocre the first couple of games until something clicked, and then it turned into a big advantage. I think it's a pretty big deal.

The buzzer is the game. Well, the buzzer and Final Jeopardy. Actually, the buzzer, Final Jeopardy and wagering strategy. Wait...the buzzer, Final Jeopardy, wagering strategy and knowing the airdate. So, the buzzer is definitely part of the game.

(But seriously, the buzzer is the game. Master that and nothing else matters.)

Don't forget the fanatical devotion to the Pope...

[slam]

Presumably some of this advantage was due to familiarity with the signalling device

Not as much as you'd think. There's a good chunk of practice before the tapings start in the morning, so as a contestant I don't think you're ever unfamiliar with it. You may start cold, because you haven't used it in a few hours, whereas the returning champ just got done 15 minutes prior. But with nerves and everything else happening, I'd say it's not much of a factor.

Well, you only really get game-relevant buzzing experience when you get used to Alex's cadence, and you only get that through gameplay. My buzzing was pretty mediocre the first couple of games until something clicked, and then it turned into a big advantage. I think it's a pretty big deal.

The buzzer is the game. Well, the buzzer and Final Jeopardy. Actually, the buzzer, Final Jeopardy and wagering strategy. Wait...the buzzer, Final Jeopardy, wagering strategy and knowing the airdate. So, the buzzer is definitely part of the game.

(But seriously, the buzzer is the game. Master that and nothing else matters.)

Well, yeah, but it helps to know a few things, too.

[the_phil]

Well, you only really get game-relevant buzzing experience when you get used to Alex's cadence, and you only get that through gameplay. My buzzing was pretty mediocre the first couple of games until something clicked, and then it turned into a big advantage. I think it's a pretty big deal.

I think it can come and go more quickly than that; at least it did for me. I went up against a 2-time lockout winner, and managed to go 9/10 when trying to buzz in during the J round (only 10? I know, it was a bad board for me). But my buzzer-fu entirely left me in DJ, and I had a stretch of 1/12 that spanned both DD's and doomed me.

[Mathew5000]

What fraction of the time, in non-lock games, does the leader bet less than a cover wager on FJ?

Has anyone mined the archive to answer this type of empirical question, and published the results?

Another question, what is the percentage breakdown of FJ wagers by second place, in cases where B>(2/3)A and B>2C ?

[Randy Kaplan]

A slightly different take on the champion's advantage. Stage fright can be a factor for some players, but if you have any nervousness or stage fright, the game is absorbing enough, and its rhythms so familiar to anyone who gets to be a contestant, that I really doubt that this is a factor for more than a couple of minutes for a challenger. That's a pretty thin "advantage" for a returning champ.

I had long experience as an actor and singer, so stage fright had never been an issue for me. I also had substantial television experience, including regular appearances as a talking-head expert panelist on the spinoff from Australia's version of the Today Show that was called Business Today. I appeared on the weekend Business Sunday edition for about a year, as well as on any number of other TV news shows Down Under.The difference I noticed immediately when I was on Jeopardy! is that the shows that I was on NEVER had a live audience. To me, the feeling was that the audience was coming into MY place to have a good time. The presence of an audience INCREASED my comfort level; I have no idea if that is true for any other contestants, but it was for me.

It takes very little time to become accustomed to coordinating your signaling to the light that goes on around the board to indicate that you can buzz in, and the practice rounds seemed more than enough to me. But perhaps that's another minuscule advantage for the defending champ. To me, though, it seemed that you could get into a state where you can take control of the rhythm of the game with a run of correct questions early on, so those minuscule advantages could possibly be parlayed into a larger one by achieving that control from the early going. I had resolved to Forrest Bounce if the going got tough, but I leaped out into an early lead in all three of my games, so it never seemed necessary, and I was a conventional down-the-column player. I never ran a single category, though, something I've always regretted.

[John Boy]

Lots of data to help answer this question is already available in the archive, and maybe one or more of you has already done the number-crunching.

In what percentage of regular Jeopardy! matches does a defending champion win, and in what percentage is there a new champion?

People colloquially say that if you get on the show, you have a 2/3 chance of losing, which is close to being true in a broad overall way. However, I would guess that in more than 1/3 of games, the defending champion wins.

I haven't collected data long-term, or collected it at all for a while. But when I did, the data were in strong agreement with your guess. Over a 2-year stretch I found that 34% of people who won a game went on to win a second (or more).

[Bob78164]

What fraction of the time, in non-lock games, does the leader bet less than a cover wager on FJ?

Has anyone mined the archive to answer this type of empirical question, and published the results?

Another question, what is the percentage breakdown of FJ wagers by second place, in cases where B>(2/3)A and B>2C ?

The Shore's Conjecture statistics come close to answering the second condition, though Shore's Conjecture requires the extra condition that A - B <= B - 2C. --Bob

[cosmos]

In what percentage of regular games does a contestant get N or more correct responses, for different values of N ranging from N=1 to the highest ever value of N. This could be for winners only, or all contestants, or both.

In other words, if I see a game where a contestant gets 20 correct responses, does that put him or her in the top 5%, or top 20%, or some other value, in terms of that particular measure?

(I know there are potential flaws with ranking contestants that way, but I just want to see the numbers and worry about the significance of the numbers later.)