Comments (211)
sorted by:
You're viewing a single comment thread. View all comments, or full comment thread.
DiscoverAFire 9 points ago +9 / -0

There's excessive amounts of data showing the disparity between polls and real-world results. If they were actually trying to predict results they would use the (trivially easy to implement) correction factors from previous polls.

They're not trying to predict the results. They're trying to pump up their team, demoralize the other team, and provide cover for fraud.

Yawnz13 3 points ago +3 / -0

Option 2 it is!

ManyDirt 2 points ago +2 / -0

In actuarial science for retirement plans we do this. For plans that deviate from standard mortality tables we do "experience studies" to correct for the deviation. Easy to do if you care about getting it right.