TDSM 4.7

From The Data Science Design Manual Wikia
Jump to: navigation, search

NBA MVP

NBA-MVP.png Reference: [1]

As shown in the chart above, Westbrook won the 2017 NBA MVP. Here is how the NBA MVP works:

There are 101 voters. Each voter has a 1st vote, 2nd vote, 3rd vote, 4th vote, and 5th vote. So There are 101 permutations, where each permutation has five positions. Assign weights to each position: 10 points to 1st vote, 7 points to 2nd vote, 5 points to 3rd vote, 3 points to 4th vote, and 1 points to 5th vote.

NBA uses a merging rankings method (chapter 4.4.2), more specifically Borda's method, to generate the MVP. For example, Westbrook has 69 1st votes, 19 2nd votes, and 13 3rd votes. We add up the weights: 69*10 + 19*7 + 13*5 = 690 + 133 + 65 = 888. Westbrook has the highest score, so he is 2017 MVP.

NFL MVP

NFL-MVP.png

Reference:[2]

Matt Ryan won the 2016 NFL MVP. He got 25 votes, which is half of the total 50 votes. Matt has the most votes, so he is the MVP.

The NFL MVP (by Associated Press) voting is simple. There are 50 voters, and each of them picks his/her MVP. So there are 50 permutations, and each permutation has only one position. There is no weight for a position, or you can say the weight is unit one. The winner is the player who received the most votes.

MLB MVP

NLB-AL-MVP.png

MLB-NL-MVP.png

Reference: [3]

MLB has two MVP each year: American League (AL) MVP and National League (NL) MVP. Their voting methods are the same, shown as follows.

There are 30 voters. Each voter has a 1st vote, 2nd vote, 3rd vote, 4th vote, ..., and 10th vote. So There are 30 permutations, where each permutation has 10 positions. Assign weights to each position: 14 points to 1st vote, 9 points to 2nd vote, 8 points to 3rd vote, 7 points to 4th vote, 6 points to 5th vote, 6 points to 5th vote, 5 points to 6th vote, 4 points to 7th vote, 3 points to 8th vote, 2 points to 9th vote, and 1 points to 10th place.

MLB uses a merging rankings method (chapter 4.4.2), more specifically Borda's method, to generate the MVP. For example, Jose has 27 1st votes, 3 2nd votes. We add up the weights:27*14 + 3*9 = 405. Jose has the highest score, so he is the MVP.



Discussion

The MVP method of NBA, NFL, and MLB are similar. They are all instances of Borda's count method.

They all have many permutations. The difference is the size of each permutation. NBA's size is 5. NFL's size is 1. MLB's size is 10.

Personally, I think the permutation size of NFL is too small and the permutation size of MLB is too large. I prefer the NBA whose permutation size is moderate. Therefore I think the NBA's MVP is the most sensible.