The Theory of Jeopardy! Categories

This is where all of the games are discussed.

Moderators: alietr, trainman, econgator, dhkendall

Post Reply
User avatar
ZachTheRiah
In Search of the Unknown
Posts: 232
Joined: Sun Jun 08, 2014 12:43 am

The Theory of Jeopardy! Categories

Post by ZachTheRiah »

In a recent article, the head writer for Jeopardy! stated there are four major types of categories: POP CULTURE, ACADEMIC, WORD PLAY, and LIFESTYLES ("everything else"). I've been trying to figure out distribution of these four types across Jeopardy! boards. The obvious ones are the WORD PLAY categories: "THE 'L' WITH IT", "'GOOD' PEOPLE", and "'DIS' HERE CATEGORY" are some recent examples.

The other three are tricky to classify. Some are obvious, like "WRITERS BY MIDDLE NAMES" being ACADEMIC, and "TV ODDS & ENDS" being POP CULTURE. But I am not sure about others, like "LIGHT THE CANDLES" or "3 CHEERS FOR THE BLUE, GREEN & YELLOW".

What do you all think?
Last edited by ZachTheRiah on Sat May 09, 2015 6:22 pm, edited 1 time in total.

User avatar
Robert K S
Jeopardy! Champion
Posts: 2575
Joined: Thu Jun 30, 2011 1:26 pm
Location: Cleveland, Ohio
Contact:

Re: The Theory of Categories

Post by Robert K S »

In his book published 1993 Harry Eisenberg told how original Trebek-era head writer Jules Minton first "explained that each show needed a balance among the categories ... [which] fall into four broad, general areas: Academic subjects such as Science, Literature, History and Geography; Lifestyle such as Fashion, Food and Drink, Religion, Automobiles and Business and Industry; Pop Culture included Movies, TV, Pop Music and Sports; finally, Word Play consisting of categories like 13-Letter Words, "T" Time, Famous Quotes and Number, Please. The idea was to have an interesting and varied mixture of categories. Something for everybody." Inside "Jeopardy!", p. 8. He also talks about the changing composition of the contestant tests in terms of these four broad areas on pp. 32-33. Balance was achieved in games by pasting approved clues to colored strips: "blue for academic, green for lifestyle, etc." Id., p. 36. Too much of the same color in a single board signaled bad composition. [Later in the book, on pp. 213 and 214, he mentions five colors--"blue, green, pink, yellow, orange"--prompting the question, what's that extra color for? Writers liked doing pink and yellow categories the best, it seems, and so those colors had to be fairly allotted among writers.]

Now, the game has changed a lot since the early days, when a relatively limited number of categories appeared on the show. These days recurring categories are the outlier rather than the norm; the writers are now encouraged to be creative and most shows have 9-12 categories that have never appeared on the show before, at least not by the exact names they are given. So it may be futile to try to classify certain categories into the same broad areas. In the examples you present, 3 CHEERS FOR THE BLUE, GREEN & YELLOW seems clearly academic while LIGHT THE CANDLES seems to fall mostly under wordplay.

The significance of area categorization for the would-be player is probably mostly in efficient Daily Double hunting, as I suspect that Daily Doubles are more likely to fall in academic categories, and in the Double Jeopardy! Round, the two DDs are more likely to fall in categories of differing broad areas. This hypothesis could be tested if we could go through and flag all the categories presented on the show, but that would be quite big project, and would involve some speculation/fudging given that many of the categories these days are hybridized.

User avatar
goforthetie
(username no longer operative)
Posts: 1337
Joined: Tue Oct 18, 2011 5:01 pm

Re: The Theory of Categories

Post by goforthetie »

"GOOD" PEOPLE isn't a wordplay category. It's really a people (which turned out to be mostly pop culture) category. Would you consider "B" IN SCIENCE to be a wordplay category?

User avatar
triviawayne
Hoping I don’t drown in this contestant pool
Posts: 2117
Joined: Tue Feb 17, 2015 9:50 pm

Re: The Theory of Categories

Post by triviawayne »

I think if you were to do this huge undertaking, you'd need a 5th category for the unclassifiable ones just called "General Knowledge".

There will be some that can fit two of the other categories and others that are just too weird to put anywhere.

Things such as the Three Cheers category would require you to read the questions to gain understanding of where to put it.
Total game show career losings = $171,522

User avatar
ZachTheRiah
In Search of the Unknown
Posts: 232
Joined: Sun Jun 08, 2014 12:43 am

Re: The Theory of Categories

Post by ZachTheRiah »

Hmmm. Interesting thoughts. Most of my questions come from trying to figure out their coding scheme. When they call something "POP CULTURE", I was trying to figure out what they meant by that, but it doesn't seem to be as cut and dried as they made it out to be. I thought that a lot of the categories were blended, or none of the above too. I guess you all just confirmed my thoughts. :)
goforthetie wrote:"GOOD" PEOPLE isn't a wordplay category. It's really a people (which turned out to be mostly pop culture) category. Would you consider "B" IN SCIENCE to be a wordplay category?
Well, by that argument, "B" IN SCIENCE would be an ACADEMIC category, since it is a science category. That make things like "WORDPLAY" the only true WORDPLAY categories, which I'm not sure is what the writers intended. I would also include things like "AMERICAN HERITAGE DICTIONARY PERFERRED PLURALS" also in WORDPLAY

User avatar
goforthetie
(username no longer operative)
Posts: 1337
Joined: Tue Oct 18, 2011 5:01 pm

Re: The Theory of Categories

Post by goforthetie »

ZachTheRiah wrote:
goforthetie wrote:"GOOD" PEOPLE isn't a wordplay category. It's really a people (which turned out to be mostly pop culture) category. Would you consider "B" IN SCIENCE to be a wordplay category?
Well, by that argument, "B" IN SCIENCE would be an ACADEMIC category, since it is a science category. That make things like "WORDPLAY" the only true WORDPLAY categories, which I'm not sure is what the writers intended. I would also include things like "AMERICAN HERITAGE DICTIONARY PERFERRED PLURALS" also in WORDPLAY
Of course "B" IN SCIENCE should be an academic category; that's my point. It requires science knowledge much more than facility with linguistics. I'm not saying that every category with quotation marks should be classified as non-wordplay; obviously some are. My point is just that not every such category is automatically wordplay.

"Wordplay" should perhaps be broadened to "Language and Vocabulary". Categories like "PULLING OUT ALL THE STOPS" should be included.

User avatar
DadofTwins
Watches Jeopardy! Way Too Much
Posts: 1206
Joined: Fri Jul 01, 2011 10:11 am
Location: Fortress of SHC-itude
Contact:

Re: The Theory of Categories

Post by DadofTwins »

I don't know how you would classify these more broadly, or how the data might look different in 2015 from 2005, but when I was creating the SHC I observed that for every week's worth of boards I could break things down by genres.

Straight academic material covered about 25 categories per week (give or take). Literature, History, Geography, Science, and Art/Classical Music were roughly equally represented back then, though art doesn't seem to come up as often as it used to.

Pop culture generally showed up 10-12 times per week. Movies & TV combined for half of that, with sports, music, and other pop culture rounding out the balance.

Straight words/wordplay/vocabulary categories showed up about once a day. This is your before & after, rhyme time, crossword clues, words of X length, anagrams and letter finds, etc.

About 10 categories per week were "Lifestyle" -- Potent potables, religion, and the "ampersand" categories (Colleges & Universities, Travel & Tourism, Food & Drink, etc.)

The rest of the categories (8-10 per week, depending on what else they had) were what I came to call "Interdisciplinary." These are categories where the answers all have something in common (people with the same first name, words that start or end with the same combination of letters or are synonyms of the same thing), but the genre in which the clue's "touch point" occurs varies, and often doesn't repeat. For instance, a category on "Angels" might include references to Angel Falls, Touched by an Angel, the L A Angels of Anaheim, Angela Merkel, and Angelman Syndrome. Sometimes the reference to the theme can be in the clue.

I found early on that Interdisciplinary categories were always the hardest to write, because you had to have both an interesting idea and the writing chops to pull it off. I'd brainstorm for six months and maybe come up with twenty I could use. When I started using guest writers, I found out that Paucle was one of the best I'd ever seen at this.

countyguy
Vacuously Undefeated on J!
Posts: 561
Joined: Sun Apr 27, 2014 4:12 am

Re: The Theory of Categories

Post by countyguy »

triviawayne wrote:I think if you were to do this huge undertaking, you'd need a 5th category for the unclassifiable ones just called "General Knowledge".
Robert K S wrote:"Balance was achieved in games by pasting approved clues to colored strips: "blue for academic, green for lifestyle, etc." Id., p. 36. Too much of the same color in a single board signaled bad composition.
So maybe, in the categories that are too broad to be placed into one classification, the clues are split into multiple classifications? Then, 3 CHEERS FOR THE BLUE, GREEN, & YELLOW would have the $1200 and $2000 in "Academic", the $1600 in "Pop Culture", and the $800 could go either way because it has an academic clue and a pop culture clue. It would be the writers' call, the same way that in LL, Commissioner Integrity can choose to put a question that ties multiple categories together in any category. Take the following question as an example:

The capital city of Chile, the home stadium of Newcastle United F.C., the brewery that produces Guinness beer, and the current home of Side Show on Broadway are all named after an individual best known as what in English?
Spoiler
St. James

There are a number of ways to arrive at the answer. In order, the four clues would be easily categorized as GEOGRAPHY, GAMES/SPORT, FOOD/DRINK, and THEATRE. The last three clues make the answer obvious if you know a certain bit of trivia, though the first clue has the individual's name in Spanish (and translating "tiago" as "James" isn't exactly intuitive), and the first clue is referring to a far better-known entity than the other three, so there is also an element of LANGUAGE. Commissioner Integrity chose GEOGRAPHY as the category, which makes a lot of sense, though it's not the only category that could fit for that question.

User avatar
ZachTheRiah
In Search of the Unknown
Posts: 232
Joined: Sun Jun 08, 2014 12:43 am

Re: The Theory of Categories

Post by ZachTheRiah »

DadofTwins wrote: The rest of the categories (8-10 per week, depending on what else they had) were what I came to call "Interdisciplinary." These are categories where the answers all have something in common...
These are, indeed, the most interesting, and the ones I think are not mentioned on the coding scheme (as mentioned by triviawayne). I really think this is what the "fifth color" was for in Robert's quote above. I like the idea of calling these "interdisciplinary".

I've been trying to see if there is a quiz-bowl-esque "breakdown" of how often these occur. I don't think the writers consciously write 15% word play, 25% pop culture, etc. a week, but I wonder if they do it unconsciously.

User avatar
ZachTheRiah
In Search of the Unknown
Posts: 232
Joined: Sun Jun 08, 2014 12:43 am

Re: The Theory of Jeopardy! Categories

Post by ZachTheRiah »

Here is some preliminary data, based on seven games from the last two weeks and one from April 23, 2014:

Academic: 27/96 = 28.125%
Pop Culture: 16/96 = 16.667%
Word Play: 14/96 = 14.583%
Lifestyles: 17/96 = 17.708%
Interdisciplinary: 22/96 = 22.916%

All percentages are truncated to three decimal places. I think the interdisciplinary category type is more important than I thought. Obviously the sample isn't "simple random", and there is a margin of error (for those who are stats nit-pickers), but I think these results are close to reality. The fact that the interdisciplinary categories are more frequent is good I think, as it keeps the show interesting. :)

Post Reply