bradley
PhD Student
Posts: 1842
Age: 74
Reg: 01-15-16
|
03-17-18 12:19 PM - Post#253608
In response to sparman
There was a fair amount of topic on this debate last year and I thought about it but thank goodness I decided not to go thru a relatively easy exercise to prove the point.
Take an adequate sample size, let's say the last 10 years. Run a probability estimate for the top 4 seeds (IL Tournament) -- let's say 40-30-20-10% for the IvyMadness. Slot the IL teams, seeds 2-4, into a best estimate NCAA seed and include the IL team that represented the league for that year with their actual NCAA seed that year. Hit the button and run the actual overall NCAA win/lost % by seed.
For example, last year was Princeton #12 seed, Harvard estimate #14 seed, Yale #15 and Penn 16. This year, obviously everyone would probably be a 16 seed although Harvard arguably could have been a #15 seed.
It can be done and it is really not all that difficult. There is clearly some advantage to send the regular season winner -- not always but there is clearly an advantage.
If it does not matter to someone as who represents the league that is ok.
|