4,294,967,295-1

How many laserdisc would that be?

6.7762636e-21 can someone convert that to odds? thats the number i got, assuming every game is a coinflip thats .5 to the power of 67(number of games)

DePaul math professor Jay Bergen calculates that picking all #1 seeds in the first round then random selections after yields a 1 in 128 billion chance at a perfect bracket. As mentioned earlier, there is a maximum of 10 million entries in this contest.

If each of these entries submits a unique bracket including all #1 seeds advancing in the first round, and these teams all do advance, the odds of there being a winner are about 1 in 12,800. Ofc there will be many duplicate entries, as you mentioned in OP. For current purposes, let's assume those don't effect the EV of the contest players as a group. Under this back-of-the-envelope calculation, $1 billion divide 12,800 suggests a before-profit insurance price of $78,125.

I was thinking of pretty much doing this, other than maybe having a few strong teams going a little deeper.

Bunch of party animals in this thread. Taking buzzkill to a whole new level. 10 mins and 2 shots of Jager for a dream. Relax

In other words, people are getting excited and starting threads in multiple subforums over $0.25 in equity, weeeeee

I don't really agree with it, because for most people the money's value has pretty steep diminishing returns. I think it would be better to do whatever would give you the best chance of winning since it wouldn't really be any more "life changing" to chop with more people unless it's all going up your nose.

Yep

I don't believe your point is valid. It is unbelievable hard to get a perfect bracket, so why waste your energy and EV on those picks? There are 9,233,322,036,854725,808 possible brackets. Do you really think you have to pick a team less likely to win a game, because you thereby decrease the chance that you'd be splitting the pot? Even if you think you have an edge in basketball-betting your well-educated pick will be likely to win 1:128,000,000,000.

Maybe I'm understanding you wrong, but your EV in this thing is so absurdly low that picking slight-underdogs-with-minor-fanbases is not increasing it in any sensible way.

I've always said that if I won the lottery I would take the lump sum, but if I were to win this I think I would take the $25million/year for 40 years (assuming that is the payout). Thoughts?

Or everybody decides they've got to pick a few upsets and nobody actually picks all favorites. And then the one guy who does gets the whole billion.

Seriously though, if you pick all better seeds down to the final four (at which point all #1 seeds are left) along with 800,000 people and those entries are evenly split through the rest of the combos, that leaves 100,000 winners taking home ten grand each. And that's with 8% of all brackets picking all favorites.

Does it actually take a few million to insure a billion dollar payout (1/333) with 1 in {many billion} odds of hitting? We might all be in the wrong business.

Has coletranedog commented on this thing yet? lol

For those who'd like to see the math.

http://www.sfgate.com/technology/bus...-s-5166595.php

Got your fish math right there.

Check the ESPN bracket challenge standings. The leaders never have anything close to the same bracket.

Found this related article while trying to figure out what Sklansky was talking about. http://www.bloomberg.com/news/2014-0...rivatives.html

Basically, fair value of insurance is estimated at $250k (not sure if right) but if you're going to take on $1bln risk you want to make a decent profit. Somewhere in the millions at least.

Kinda seems crappy to make a contest that is so unlikely to have a winner and the only one making money is Buffet who has more money than he knows what to do with.

Although it's an interesting idea it's not even getting much press for Quicken Loans, the company making the contest. Instead Buffet's name dominates the headlines. Marketing fail.

When people at Quicken Loans read this post they can feel free to pay pal me a couple million in consulting fees. Your welcome.

TIL that Warren Buffett would make more money if he quit actively managing Berkshire Hathaway and instead hired a pack of traveling salesmen who did nothing but convince the entire Fortune 500 to offer huge sweepstakes payouts with horrible odds.

Definitely

Lump sum is $500 million, if I remember correctly. After taxes in the U.S., round it off to ~$300 million? If you can invest the whole amount at 5% over 40 years, that's $2.1 billion. Of course, with $300 million, you could become your own venture capital firm or something.

With $25 mil/yr, I think I just end up spending $20 million a year on frivolous crap, and feel good about "making" $5 million a year. At least, that would be me. With the lump sum, I feel like I'd be motivated to do something big with it.

1. Encode all possible combinations as a Quantum Superposition.
2. When he refuses to pay, get a physics professor as an expert witness and sue him for 1 billion plus damages.
3. Settle out of court for 1 million.

Juk

Yeah it's like a super obvious point that anyone who knows basic math and has done office football pools knows and has used plenty. Maybe they haven't done division but they know they want games with high x and low y.




And this too

I think that a graph of x as it approaches zero will look similar to y as it approaches zero (barring some pockets of populations that pick underdogs for sentimental reasons). This would seem that x/y calculations would more or less mimic the value of picking simple favorites/dogs (ie, ignoring y).

espn said they havent had 1 perfect bracket in 16 years of running the bracket challenge. I dont know how many total there were in that time.