Skip to content

Latest commit

 

History

History
58 lines (39 loc) · 3.5 KB

README.md

File metadata and controls

58 lines (39 loc) · 3.5 KB
Raw

March Madness Prediction

Using a model built from historical, season level stats to help me fill out my bracket for the 2019 tournament.

About the Code

  • The first 3 scripts import the relevant datasets (see below).

  • The fourth script joins the season aggregate stats with the tournament performance data. It then builds a random forest model with the ranger package, using data from 2010 - 2018 to predict outcomes in 2021.

  • The fifth script joins the ranked ranger predictions with the rankings from kenpom.com, and averages the two sets of scores to arrive at a final ranking.

About the Data

The data has three components:

  • Season level stats, for NCAA teams from 1995 to 2019. This is scraped from sports-reference.com. Sadly there is a bias here, sports reference includes data from the post-season each year as well. So I limited the features in my model to only included percentage columns, not any with absolute quantities.

  • Tournament-game data from 1995 to 2019. This is from mmclaughlin87 github repository for the years 1995 to 2018. For 2017 and 2018, I dug further into the original source in this pdf. This data is the source of my dependent variable, which is the sum of the total winnings margin for each team for each year.

  • The 2021 kenpom.com ratings.

My Session Info

R version 3.5.2 (2018-12-20)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS High Sierra 10.13.4

Matrix products: default
BLAS: /System/Library/Frameworks/Accelerate.framework/Versions/A/Frameworks/vecLib.framework/Versions/A/libBLAS.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] tictoc_1.0      ranger_0.11.1   lme4_1.1-17     Matrix_1.2-15   bindrcpp_0.2.2  lubridate_1.7.4 rvest_0.3.2     xml2_1.2.0      forcats_0.3.0  
[10] stringr_1.3.1   dplyr_0.7.8     purrr_0.2.5     readr_1.1.1     tidyr_0.8.2     tibble_2.0.1    ggplot2_3.1.0   tidyverse_1.2.1

loaded via a namespace (and not attached):
 [1] progress_1.2.0     tidyselect_0.2.5   reshape2_1.4.3     splines_3.5.2      haven_1.1.1        lattice_0.20-38    colorspace_1.4-0  
 [8] yaml_2.2.0         utf8_1.1.4         rlang_0.3.1.9000   nloptr_1.0.4       pillar_1.3.1       glue_1.3.0         withr_2.1.2       
[15] selectr_0.4-1      RColorBrewer_1.1-2 modelr_0.1.1       readxl_1.1.0       bindr_0.1.1        plyr_1.8.4         munsell_0.5.0     
[22] gtable_0.2.0       cellranger_1.1.0   labeling_0.3       GGally_1.4.0       curl_3.2           fansi_0.4.0        broom_0.5.0       
[29] Rcpp_1.0.0         scales_1.0.0       backports_1.1.3    jsonlite_1.6       digest_0.6.18      hms_0.4.2          stringi_1.2.4     
[36] grid_3.5.2         cli_1.0.1          tools_3.5.2        magrittr_1.5       lazyeval_0.2.1     crayon_1.3.4       pkgconfig_2.0.2   
[43] MASS_7.3-51.1      prettyunits_1.0.2  minqa_1.2.4        assertthat_0.2.0   reshape_0.8.7      httr_1.3.1         rstudioapi_0.7    
[50] R6_2.3.0           nlme_3.1-137       compiler_3.5.2