some numbers about Chicago Open

[grapesmoker]

Since I took home the scoresheets from CO, I thought I'd spend some time entering them into a Google spreadsheet to do some statistical analysis of the questions. There's one sheet for every round, and two columns, tossups and bonuses, for each room. If someone converted a tossup, I put a 10 or 15 there, depending on whether it was a power or not. I didn't bother taking negs into account because I was primarily interested in seeing the conversion rates. Likewise, for the bonuses, the value in the column represents whatever the team answering the bonus got on it. There's no attempt here to disaggregate data by teams; I just didn't care that much about it. What I was mainly interested in was the conversion rate for tossups and bonuses overall, and on a per-tossup question basis, to see which questions were harder. Since not all tossups are answered, not all bonuses are heard, hence the blank spaces. For bonuses, the conversion percentage is the fraction of available points that the field earned overall on the bonus. Obviously if fewer teams heard the bonus, fewer overall points would be available.

The last sheet currently contains some plots. They're both histogram plots of the conversion rates for the whole tournament. You can see that if you squint, the conversion rate for tossups looks like a sort of power law with a long tail. About half the tossups had a 100% conversion rate, then another fifth or so had a conversion rate between 87.5% and 100%, and down from there. The bonuses look somewhat Gaussian, but I also didn't take into account bonuses that were only heard by one or two teams. Probably should correct for that but I'm lazy. Anyway, I don't have any particular conclusions to draw from these numbers, but I thought people might be interested in them. Tomorrow I'll try to go through and note which questions were the hardest.

[theMoMA]

I put up some conversion numbers last year as well, but I can't seem to find the post where I discussed them. In any event, in case anyone is interested in comparing basic tossup/bonus stats, here is the spreadsheet: https://docs.google.com/spreadsheets/d/ ... sp=sharing

[grapesmoker]

So I crunched some more numbers and present to you the hardest tossups in each round. My criteria for "hardest" was "answered in no more than 50% of rooms" although obviously that still leaves a pretty decent gap between 50% conversion and 0%. I'll do the bonuses another time. Number in parentheses is the number of rooms that converted the tossup.

Round 1: Hiranyakashipu (4)
Round 2: Robert Rosenthal (0), Bezier curves (3)
Round 3: molar conductivity (1), Caracas (3), dissipation (4), phonetics (3), holomorphic (4)
Round 4: knowing how (3), bandit (2), difference-in-differences (0), Dona Rosita(0)
Round 5: Lane-Emden (4)
Round 6: Sigismund III (4)
Round 7: Baldur's Dream (2), Richard Duke of York (3), galactic luminosity (2), Ruth Marcus (3), Gergovia (3)
Round 8: Arginusae (3), Purity and Danger (4)
Round 9: Seferis (4), Tobias (4), density of states (3)
Round 10: Etain (3), Tezuka (4), Kolakowski (2), BCR-ABL (2), North Dakota Non-Partisan League (0), Zero lower bound (3), Alessandri (3)
Round 11: atomization (2), Wilkes (2), mental representations (2), Acharnians (4)
Round 12: Abu Sayyaf (3), cancer stem cells (4), Ramakrishna (2)
Round 13: Cesar Vallejo (4)
Round 14: The Parnassus (4), Cornell boxes (3)
Round 15: Henry I (2), realism-formalism (1), C++ templates (3), Elihu (1), Brahms piano concerto #2 (4)
Round 16: Noyori (4), Neyman-Pearson (2), Compton's cafeteria riot (3), G6PD (4)
Round 17: Vautrin (3), Fouquet (4)

[Good Goblin Housekeeping]

clearly quizbowl needs to remove science as it's TOO HARD

[naan/steak-holding toll]

Caracas was only converted in three rooms?!? Were there a lot of negs?

[Tejas]

Caracas was only converted in three rooms?!? Were there a lot of negs?

The giveaway wasn't that easy, only mentioned that it was where Bolivar grew up and nothing about Venezuela.

[grapesmoker]

Caracas was only converted in three rooms?!? Were there a lot of negs?

What Tejas said. Also, negs wouldn't matter because I'm only looking at whether it was answered eventually.

[Rococo A Go Go]

North Dakota Partisan League (0)

Was this actually a tossup on the Nonpartisan League?

[grapesmoker]

North Dakota Partisan League (0)

Was this actually a tossup on the Nonpartisan League?

Yeah, whoops, typo.

[Nicklausse/Muse]

[unread question mention deleted]