Ayecarumba
Ayecarumba
  • Threads: 236
  • Posts: 6763
Joined: Nov 17, 2009
February 9th, 2012 at 12:09:50 PM permalink
According to an AP article on the ESPN website, a camel in New Jersey went 15-6 this year, including a correct pick of the Giants over the Pats in the Superbowl. With that win, the camel is now 89-51 lifetime.

Would a proper analysis of whether the camel is better than a coin flip need to include the spreads on each game picked? Or is it enought that the expected random results should be divided 50/50?

If the expected result is 50/50, is the camel's record significantly better than random?
Simplicity is the ultimate sophistication - Leonardo da Vinci
Wizard
Administrator
Wizard
  • Threads: 1493
  • Posts: 26498
Joined: Oct 14, 2009
February 9th, 2012 at 12:27:08 PM permalink
According to my math, the odds of luck this good or better are 1 in 931. That is using the binary distribution, not a Gaussian estimate.

As long as he was randomly picking then his odds wouldn't matter whether it was against the spread or straight up.

If I ran a zoo I would have every single animal make such picks. One of them would probably finish with a record this good and give the zoo some good free publicity.
"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)
  • Jump to: