Hacker News Reader

Comment by hdivider | Hacker News Reader

Nice calculation, but it doesn't contradict the 'rational' argument at all. Slightly disappointing: looking at the title and first paragraph I expected some weird statistical wizardry on why the calculation of expected values is fundamentally flawed somehow. (Which would be surprising, given that it's an important mathematical tool in quantum mechanics (even though the formalism there is different).)

Also, why pay for the privilege of imagining a better outcome? Imagination is >free< and virtually unlimited, surely.

View on HN

He does contradict the "rational" argument that you should only factor in money when doing your expected value calculations.

> Imagination is >free< and virtually unlimited, surely.

I guess he gets more entertainment out of knowing that what he is imagining has a chance of becoming a reality.

I just imagine that my dreams can come true! Maybe I bought a ticket yesterday and just forgot! Or my mom secretly bought a ticket for me!

Exactly. I'm entertaining myself every time I release some software and imagine that I'll sell millions of copies :)

Reminds me of a dilbert comic:

http://dilbert.com/strips/comic/2011-02-12/

That's not >free< You probably spent many hours writing that software.

No, the dreams are free, as I make software for a living anyway.

The expected value isn't flawed, but it's often not an useful guide to what action you should take. Imagine a lottery with an accumulated jackpot of $11 million, and there's one million $10 tickets (this isn't all that rare). Should you take all your savings and buy 1000 tickets? You have a positive expectation value, and 99.9% chance of ending up penniless. Is one that buys the tickets really a much more rational person than one who does it when the jackpot is $8 million? One can make just positive expectation bets and be certain to lose it all. The Kelly Criterion is a nice little formula for calculating the optimal bet size from the risk and the expectation.

"The expected value isn't flawed, but it's often not an useful guide to what action you should take."

Yeah, I agree with that - if good decisions could be made that easily we'd all be phenomenally successful. There are mountains of complications (both qualitative and quantitative) that must be heaped on top of expected values if they're going to be any use at all in deciding what action to take. Perhaps this was the author's point, but perhaps that's sort of obvious anyway.

This is because you're not after money but after welfare (whatever that is).

Reply on news.ycombinator.com