Friday, November 20, 2015

My Entry in the New Yorker Cartoon Caption Contest #498

Here is my entry in the New Yorker Cartoon Caption Contest #498 for November 16, 2015. The double-edged drawing is by Jack Ziegler.

"The deal's off! I'm going to rule England!"



Would you like to see how I arrived at this? Here's the full caption progression:
"Miss Baxter, I asked for a pen!"
"Miss Baxter, do we have anything mightier than the sword?"
"Guess what? I'm going to rule England!"
"I have to cancel. I'm going to rule England!"
"The deal's off. I'm going to rule England!"


November 21, 2015 Update:  This caption contest was the first to use a crowd-sourced survey to select the finalists. This is the subject of some discussion in the comments section. The online survey presented 25 captions. You were asked to rate each as funny, somewhat funny, or unfunny. Names of the contest entrants were not associated with the captions. The survey did not forbid multiple entries and different captions came up each time. Here are some examples of what it looked like. As you can see, they do not appear to be prescreened.





























November 23, 2015 Update:  The Finalists




December 7, 2015 Update: Winning Caption



Note: Last week Zachary Kanin bellied up to the bar with a locomotive. My caption just didn't have any steam. Chug along with the results of Contest #497.

Jack Ziegler is one of the funniest cartoonists in the world, so of course you can see more of his work on this blog.

01691

14 comments:

  1. Here's mine: "Stop poking into my business!"

    Did you hear about the survey they're doing at the Cartoon Lounge blog to pick the finalists this week?

    ReplyDelete
  2. I noticed you could take the survey more than once so I found myself taking it a few times to see if mine would come up! I wonder if they put all entries into the survey or just some of them. There were some real clunkers in there (one I saw used profanity) so it doesn't look they screened them before putting them in.

    ReplyDelete
    Replies
    1. I too wonder whether there was any screening of the captions. If not, this is a very democratic way of selecting the finalists, except for the quibble that no one should be able to vote in more than one survey.

      Delete
    2. If they're actually including all entries, that might be fair. But I'm not sure how that would work with thousands of entries and each person only voting on 25. I'd really like to see how that's a fair way to do it. Also, my example shows how it can be biased. I can't be alone in the desire to seek my own entry. There's also the temptation to vote down others in order give your entry a better chance. To avoid these problems, the judges should pick a short list of about 50-75 (like they do normally) and have the readers pick the finalists instead of the editors. And each voter gets to see the entire short list all at once so they can see if their entry is in the list before voting.

      Delete
    3. With 5,000 entries, it would take 200 groups of 25 to get through them all once, not terribly daunting if a large enough crowd participates. I could foresee an algorithm where captions consistently voted unfunny would be eliminated from the pool while the voting continued, although I'm not sure this is strictly fair. With a smaller pool of, say, 100 choice captions, I can also anticipate the potential for manipulating the outcome. A well-connected individual with a short-listed caption could recruit others through social media, for example, to vote only for that one particular caption. This is potentially an issue already in voting on the finalists.

      Delete
    4. For something to get "consistently voted unfunny" you would need a lot more than 200 respondents. I did a little research and there is an interview with Bob Mankoff about the new documentary that just came out. See here: http://www.aintitcool.com/node/73783. I noticed toward the end he talks about this experiment and he says "we're going to crowdsource about 500 captions that people can vote on". So it appears they randomly selected a sample of 500, which would make this exercise more feasible. Although it also makes it completely unfair!! 4500 entries get no chance!!

      Good point about people rigging the vote by recruiting others. There are just a ton of problems with making this a public vote.

      Delete
  3. I agree that using 500 "randomly selected" captions for the survey would be totally unfair, but Bob Mankoff doesn't say precisely how the 500 are to be chosen. If there's some sort of rudimentary screening process in which only the best 500 could be used—including, it goes without saying, yours and mine—then this public voting concept makes a little more sense. But the 25 survey examples I added to this post today does not inspire confidence that only the best were selected—although a few of them are quite good!

    ReplyDelete
    Replies
    1. Yes some of the ones you posted are plain awful, plus I saw one that used profanity, so that to me makes it pretty clear they were randomly selected. Some of the fairly good ones you posted I saw as well when I went through it. So I think that's part of the algorithm...the high-scoring ones stay in contention and probably get shown more frequently.

      Delete
    2. Well, that leaves us once again presumably with a base of only 500 captions selected at random. If the purpose is to pick the three funniest captions submitted, this is clearly no way to run the contest. Nevertheless there should be no problem finding three excellent finalists even in a 10% sample.

      Delete
    3. Another problem with the way they did things was how in the video Bob picked his favorite among a handful of entries, thus introducing a bias. I wouldn't be surprised to see the only one he rated funny (I told you I'll meet you at the lake) be a finalist.

      Delete
    4. Quite right. But then the caption contest was never fair, was it?

      Delete