Skip to content

sidharthrajaram/mvp-predict

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 
ย 

Repository files navigation

Alt text

2019 - 2020 Predictions

UPDATE: YES, the official 2020 NBA MVP is Giannis Antetokounmpo.

Predictions made prior to the start of the season (as of June 27, 2019)

Player Name Rank
Giannis Antetokounmpo 1.393
Rudy Gobert 1.331
James Harden 1.172
Nikola Jokic 0.881
Anthony Davis 0.828
Karl-Anthony Towns 0.629
Kevin Durant 0.339

Notes

None yet.

2018 - 2019 Predictions

UPDATE: YES, the official 2019 NBA MVP is Giannis Antetokounmpo.

Predictions made prior to the start of the season

Player Name Rank
Giannis Antetokounmpo 0.899(1)
LeBron James 0.871(2)
James Harden 0.835(3)
Anthony Davis 0.830(4)
Karl Anthony Towns 0.73(5)
Kawhi Leonard 0.55(6)
Nikola Jokic 0.33(7)

Notes

The methodology used for 2019 was different than the method used for 2018. The 2018 MVP was predicted using a progressive method that updated every week. The 2019 MVP was predicted entirely before the season began (I'll be using this method moving forward).
Upon running the 2019 model on the 2018 field, these were the results:
James Harden Anthony Davis LeBron James Karl Anthony Towns (?)
1.559(1) 1.036(2) 0.799(3) 0.608(4)

data/mvpForecast.csv contains forecasted advanced stats for ESPN's top 20 NBA players of the 2018-2019 season.

forecast.py is responsible for generating the aforementioned future stats.

predict.py is responsible for ranking the MVP candidates based on the model.

mvp_statline.py is an independent script that predicts the per-game and advanced stats of this season's MVP.

mvp_statline's predictions for this season's MVP

ppg rpg apg per vorp +/- ts%
28.8 7.1 7.3 29.0 8.67 10.2 0.61

2017 - 2018 Predictions

Final frontrunner: James Harden

ppg per vorp +/- ts%
30.4 29.8 8.3 10.9 0.619

Predictions as of April 12, 2018

James Harden LeBron James Anthony Davis Russell Westbrook
29%(1) 21%(2) 14%(3) 7%(4)

In October of 2017, mvp-predict generated this as the projected MVP statline:

ppg per vorp +/- ts%
27.63 29.01 8.79 10.24 0.606

Releases

No releases published

Packages

No packages published

Languages