Friday, September 2, 2016

My Entry in the New Yorker Cartoon Caption Contest #534

Here  is my entry in the New Yorker Cartoon Caption Contest #534 for August 29, 2016. The drawing is by P. C. Vey.

"Okay, let's review some marketing basics."

September 5, 2016 Update:  The Finalists

September 6, 2016 Update:  The Rankings
The crowdsourced rankings are available here.
The three finalists were selected from the top 21 entries (of 4295). Here are the top 40:

My own caption only made it to the top 76% of entries, a dismal ranking:

...And the bottom 40:

Was my caption this week really that bad? Shouldn't I have been in at least the top 50%? I can only conclude that Vladimir Putin has been tampering with the Caption Contest ranking results. 

September 12, 2016 Update:  I like that third finalist's caption, but once again I neglected to vote.

September 19, 2016 Update:  Winning Caption

Note:  Last week, cartoonist Joe Dator served up a Caption Contest cartoon down in the subway. My caption didn't rise up to the occasion. You'd better mind the gap for Contest #533.

P. C. Vey is the Caption Contest's go-to guy for office humor.



  1. My caption ranked 2227. "Oh, good, you're free."

    So this week showed a glaring problem with using the crowdsourcing. If they don't trust the crowd's number 1 pick, how can they trust that the crowd's other highest picks are the true best captions? It seems that Bob's dislike of this week's #1 rank shows the whole system is flawed and they shouldn't be depending on it so much.

  2. They shouldn't, but they will. It's far easier for the editors to have hundreds of random people spend a few thousand hours crowdsourcing the rankings than to read and evaluate each and every caption themselves. It isn't going to go away unless it is gamed to the point where it becomes meaningless. Unfortunately, the Survey Monkey (I think) system as it exists is pretty easy to subvert. I joked about Vladimir Putin tipping the scales, but there's no reason for him to interfere. It is far more likely that a contestant could design a ranking bot. It could be run hundreds of times for each contest. Whenever it recognizes the botmaster's caption it ranks it as funny, captions with similar keywords to the botmaster's could be ranked as somewhat funny, and all others would be unfunny. That's one way an awful caption could be ranked number one. Come to think of it, maybe it's already happening...