Last time out, we looked at representation of contestants on Jeopardy! by Nielsen Designated Market Area. In this second part, we work with a different set of boundaries. This one is more familiar, older (largely set since the start of the 20th century), and far more difficult to change. It’s the fifty states and the District of Columbia themselves.
Everything about the representation ratio, and the caveats and conditions on where contestants are assigned to, carries over from the previous post. This time, the population figures are those estimated by the U.S. Census Bureau, effective July 1, 2023.
How do things stack up this time?
You might look at that and think, “Who’s got the 14.46? Where is it?” It might be difficult to find it without zooming in — but if you recall who came out on top in part one, it won’t be surprising. It’s the District of Columbia. DC proper has 0.20% of the population, but supplied 2.93% of all Jeopardy! contestants in the last decade.1
The state with the highest representation ratio is Vermont (2.55), followed by Massachusetts (2.16), Maryland (1.90), and New York (1.79). The show’s home state of California comes in at 1.47, and that’s even with the boost it got in Season 37 due to COVID restrictions.
Seventeen states and DC have representation ratios greater than 1.00.2 They are:
the six New England states (Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, Connecticut)
the District of Columbia and six Mid-Atlantic states (New York, New Jersey, Pennsylvania, Delaware, Maryland, Virginia)
the three Pacific Coast states in the “lower 48” (California, Oregon, Washington)
the states with the largest cities in the Central and Mountain Time Zones (Illinois and Colorado, home to Chicago and Denver, respectively)
At the other end of the spectrum, recall the quote from FreddieMarkury that I shared at the end of part one: “Florida and Texas tend to be underrepresented.” Ain’t that the truth, and it’s a truth these data unequivocally tell. Twelve states have rep ratios less than 0.50; of those twelve, by far the two most populous are Texas (0.45) and Florida (0.48). The Lone Star and Sunshine States certainly get their due on Jeopardy!, with 282 contestants between them — but that’s 7.31%, against their total share of the population of 15.86%. In addition to the aforementioned DC and Vermont, three other states with 2023 populations of fewer than one million have higher ratios than Florida and Texas do.3
Does that map have a familiar look to it?
Because it did to me. I mean this in terms of the two diverging colors. In this sense, the three exceptions would be Wisconsin, New Mexico, and Hawaii.
If you flipped those three states from some shade of brown to some tint of green, it might bear an even more striking resemblance to the 2020 Electoral College map.
Given what I’ve come to know about the Jeopardy! contestant set over the last ten years, the map coming out this way didn’t surprise me. But I further augmented my data set, and one particular rearrangement brought the point home in stark relief. I present that in the image from my Excel file below. This is the fifty states and DC sorted in order of rep ratio, highest to lowest. Alongside them is how each voted in the last three U.S. Presidential elections.
If you set a dividing line at 0.833 (5/6), you have 19 states and DC above that line, and the remaining 31 states below it. Above the line, the aggregate electoral outcomes: 58 D, 2 R, and those two were Pennsylvania and Michigan in 2016 — won by Donald Trump by a combined 54,996 votes out of nearly eleven million cast in the two states that year. Below the line, taken together: 77 R, 16 D, and nine of those latter sixteen were Democratic sweeps in two sparsely populated Western states (Nevada and New Mexico) and a third nowhere near the mainland (Hawaii).
On June 7, the airdate of the final syndicated episode of Wheel of Fortune hosted by Pat Sajak, the left-wing commentator Emma Vigeland observed: “Jeopardy is Democratic Party coded and Wheel of Fortune is Republican Party coded”.4 At least as to the first half of Vigeland’s statement, Exhibit A to that effect, presented above.
Exhibit B — origin states of contestants over the last three seasons, compared against their 2020 Presidential election results.
Season 40: 74.0% D (111/150), 26.0% R
Season 39: 76.2% D (275/361), 23.8% R
Season 38: 82.0% D (364/444), 18.0% R
Seasons 38-40 combined: 78.5% D (750/955), 21.5% R
2023 50+DC population: 56.7% D (188.9M/334.9M), 43.2% R
Back around to what I posed at the start
Namely, this: “what, if any, implications does, or should, [the distribution of Jeopardy! contestants relative to population] have for the show?”
First of all, the words “relative to population” in my brackets there are highly important. Coming to this point, I recalled something I said several years ago on JBoard. Having gone and found it, more than the specific turn of phrase I was thinking of is relevant — here’s the full quote:
I don't think it's quite right to state that a particular place or group of places is over- or under-represented. This is a game show, not the House of Representatives - the producers need only embrace geographical diversity to the extent that, in their opinion, they believe it will drive ratings.
(October 11, 2017; source)
This stands in full. There is no right to a place on Jeopardy!. The factors of the geographic distribution of contestants elucidated here and in part one – favoring larger media markets, and left-leaning states – are not necessarily problems that the show has to solve. If they believe that addressing those things will broaden the appeal of the show, then by all means the producers and contestant coordinators should pursue that avenue. In fact, there’s evidence that Jeopardy! is thinking along those lines; see this Tweet, this Thread, and this Instagram post.5 Any efforts the show might undertake in this regard could be limited, however. The pool of available contestants is, of course, necessarily limited by the pool of applicants. Jeopardy! cannot frog-march recalcitrant South Dakotans, West Virginians, or Mississippians to computers and compel them to sit for the Anytime Test.
There has been a lot of turnover in the Jeopardy! contestant department since Harry Friedman retired. Three long-serving members of that department are no longer with the show (Maggie Speak retired, Glenn Kagan was dismissed under dubious circumstances, and Corina Nusu departed to pursue other opportunities). That’s decades of experience that can no longer be brought to bear on who appears on the Alex Trebek Stage. But the current departmental staff, while not always getting it right (because no human being can), by and large hit the mark with their choices (in my completely subjective opinion). I trust them to continue to do so.
Besides: if where the people on Jeopardy! come from, or how they present themselves, or what they do, or what they believe, is or becomes a problem for you, the simplest and most elegant of solutions is available.6 I’m sure you can deduce what that is.
Population: 678,972 / 334,914,985; contestants: 113 / 3,855.
At two decimal places, Minnesota’s is exactly 1.00; taken out one further, it’s 0.999.
Wyoming (0.60), Alaska (0.59), and North Dakota (0.55).
In a quote of her Tweet, I noted: “The funny thing is, of course, that during this relatively brief period of Jennings hosting #Jeopardy and Sajak hosting #WheelOfFortune, that coding matches perfectly onto those hosts”
The four states highlighted: Georgia (rep ratio 0.82), Alabama (0.49), Texas (0.45), and Wisconsin (0.53).
This also goes for objections to clue material of any sort.