Recently I have been handing out some club promotional items to my friends. Nothing fancy.
Some of them are have better printing quality than others. Nothing major; a vast majority were 10/10 quality, several were 9/10, and a couple were 8/10.
Rather than try to decide to distribute them according to a personal rating of who my bestest, best friends are, and risk offending people twice (once because they got a poorer item, and again when they realised it was because I didn’t love them), I decided to distribute the items completely randomly.
Now, it isn’t enough to assign them randomly; people would be suspicious I was playing favourites. So I proceeded to hold out two items, one in each hand, with the printing hidden, and ask “Left or Right?”, to demonstrate that I wasn’t forcing a choice. In actual fact, I deliberately blinded myself to the choices – I didn’t know which, if either, was better, but they had no way of being sure of that.
The responses I got were odd.
In a sample of over a dozen, I don’t think a single person actually chose “Left”. That’s very surprising, until I reveal I don’t think a single person chose “Right” either.
Instead, a few said “I don’t care”, which is rational, but put me in a tiny bind. If I selected a bad choice on them, I could then be accused of doing it deliberately. The real reason for offering the choice was to protect me, not them. If they didn’t choose, I lost that protection.
The rest dithered over which to choose, until I became frustrated and thrust one at them and moved to the next person. “Left? Right? Oooh, so hard!” Actually, it isn’t hard. There was absolutely no indication that would help them. The expected value of each hand is identical; the only irrational choice would be to waste effort trying to rationally choose an answer with no data. Instead, they stalled like human instantiations of Buridan’s donkey, unable to choose. Bah!
None of them clicked that the offering of this choice was a piece of theatre anyway. If they were going to stop to think it through, they shouldn’t have been considering left versus right; they should have challenged me to prove that I wasn’t offering them a choice between two of the “dud” items, thus forcing one on them.
Even if I proved that I was holding one good and one bad item, that still wouldn’t show that I wasn’t cheating. A majority of the items were good, so I could be offering my least favourite friends one good and one bad, while offering my favourite friends a choice between two good ones. That wouldn’t guarantee any one individual would get a bad one, but just increase the risk to 50%, where the favoured few enjoyed a much lower rate.
I repeat, I wasn’t doing any of this; I deliberately hid the printing and shuffled them so even I didn’t know which were which.
Nor have I sat down and ranked my friends objectively. “You’re my best friend, and you’re my second best friend” is a manipulative game best left to the schoolyard.
… but if I did, just pointing to a hand and saying “THAT ONE!” might just rocket you to the top.
Comment by configurator on July 6, 2011
You need to pick your friends more carefully. Most of the people I know aren’t as donkeyish.
Comment by Aristotle Pagaltzis on July 6, 2011
Unless his friends are all equally good choices, then he needs to pick them randomly.
Comment by Sunny Kalsi on July 6, 2011
Maybe they thought it was a trick question?!?
You: “Make an arbitrary choice”.
Me: “…. wwwwhyyy?”
Comment by Andrew on July 6, 2011
The ones who know you best are the ones who take the time to think about why you would be offering them an apparently random choice.
Comment by Geoff on July 11, 2011
“they stalled like human instantiations of Buridan’s donkey”
Wonderful!
To the left, a dozen chickens plus $600. To the right, a 1D.
Comment by Aristotle Pagaltzis on July 19, 2011
Is the author of the arXiv blog reading you? (Are you the author of arXiv blog?)