Abstract
Analytics and professional sports have become linked over the past several years, but little attention has been paid to the growing field of esports within the sports analytics community. We seek to apply an Adjusted Plus Minus (APM) model, an accepted analytic approach used in traditional sports like hockey and basketball, to one particular esports game: Defense of the Ancients 2 (Dota 2). As with traditional sports, we show how APM metrics developed with Bayesian hierarchical regression can be used to quantify individual player contributions to their teams and, ultimately, use this player-level information to predict game outcomes. In particular, we first provide evidence that gold can be used as a continuous proxy for wins to evaluate a team’s performance, and then use a Bayesian APM model to estimate how players contribute to their team’s gold differential. We demonstrate that this APM model outperforms models based on common team-level statistics (often referred to as “box score statistics”). Beyond the specifics of our modeling approach, this paper serves as an example of the potential utility of applying analytical methodologies from traditional sports analytics to esports.
Appendix A R code to reproduce results
The following code allows readers to download Dota 2 data in order to reproduce or extend results from this manuscript. Obtaining the data in the correct format requires multiple steps. The first step is to obtain the game ID from https://www.opendota.com. The website offers several search options, we collected the data by searching for all professional games in the given time periods and retained the game ID. An example of this can be found at our github page at https://github.com/nick3703/Code-For-A-Bayesian-Adjusted-Plus-Minus-Analysis-for-the-Esport-Dota2 as GameIDs.csv.
A.1 Obtaining the data
The game IDs were then used with the R library RDota2 (Boutaris, 2016) and tidyverse (Wickham, 2017) in the following code. Note that this code requires an API key that can be obtained via https://steamcommunity.com/dev.
A.2 Preparing data for exploratory analysis of Section 4
What follows is code to obtain data in format to conduct player statistics used in Section 4.2. This code can be found on our github repo and is called CreatePlayerData.R.
A.3 Creating team summaries of data
The following code creates team summaries used in Section 4.1 to create team data. This code can be found on our Github repo entitled ‘CreateTeamData.R’
A.4 Preparing data for APM models
The following code uses the ‘player.data.by.game.rds’ file constructed above to prepare the data for use in the APM model. The code can be found on our Github repo entitled CreateAPMData.R.
Acknowledgments
The authors would like to thank the reviewers and editors for several useful comments that helped improve this work.
Author contribution: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: None declared.
Conflict of interest statement: The authors declare no conflicts of interest regarding this article.
References
Boutaris, T. 2016. RDota2: An R Steam API Client for Valve’s Dota2. URL https://CRAN.R-project.org/package=RDota2, rpackage version 0.1.6.Search in Google Scholar
Carpenter, B., A. Gelman, M. D. Hoffman, D. Lee, B. Goodrich, M. Betancourt, M. Brubaker, J. Guo, P. Li, and A. Riddell. 2017. “Stan: A Probabilistic Programming Language.” Journal of Statistical Software 76. https://doi.org/10.18637/jss.v076.i01.Search in Google Scholar
Deshpande, S. K., and S. T. Jensen. 2016. “Estimating an NBA Players Impact on His Teams Chances of Winning.” Journal of Quantitative Analysis in Sports 12: 51–72.10.1515/jqas-2015-0027Search in Google Scholar
DOTA2.com. 2019. The International Championships. https://www.dota2.com/international/overview/ (accessed January 09, 2019).Search in Google Scholar
dota2.gamepedia.com. 2019. DOTA 2 Wiki – Gold. https://dota2.gamepedia.com/Gold (accessed July 10, 2019).Search in Google Scholar
Drachen, A., M. Yancey, J. Maguire, D. Chu, I. Y. Wang, T. Mahlmann, M. Schubert, and D. Klabajan. 2014. “Skill-Based Differences in Spatio-Temporal Team Behaviour in Defence of the Ancients 2 (Dota 2),” in 2014 IEEE Games Media Entertainment, IEEE, 1–8.10.1109/GEM.2014.7048109Search in Google Scholar
Eggert, C., M. Herrlich, J. Smeddinck, and R. Malaka. 2015. “Classification of Player Roles in the Team-Based Multi-Player Game Dota 2.” in International Conference on Entertainment Computing, Springer, 112–25.10.1007/978-3-319-24589-8_9Search in Google Scholar
Fearnhead, P., and B. M. Taylor. 2011. “On Estimating the Ability of NBA Players.” Journal of Quantitative Analysis in Sports 7.10.2202/1559-0410.1298Search in Google Scholar
Fisher, E. 2017. Colleges Offer Analytics Track. https://www.sportsbusinessdaily.com/Journal/Issues/2017/06/05/In-Depth/Data-side.aspx (accessed September 11, 2019).Search in Google Scholar
Ganguly, S., and N. Frank. 2018. “The Problem with Win Probability.” In Proc. of the 12th MIT Sloan Sports Analytics Conf.Search in Google Scholar
Gramacy, R. B., S. T. Jensen, and M. Taddy. 2013. “Estimating Player Contribution in Hockey with Regularized Logistic Regression.” Journal of Quantitative Analysis in Sports 8: 97–111. URL https://www.degruyter.com/view/j/jqas.2013.9.issue-1/jqas-2012-0001/jqas-2012-0001.xml.10.1515/jqas-2012-0001Search in Google Scholar
Ilardi, S. 2007. Adjusted Plus-Minus: An Idea Whose Time Has Come. https://www.82games.com/ilardi1.htm.10.1145/1226690.1226691Search in Google Scholar
Ilardi, S., and A. Barzilai. 2008. Adjusted Plus-Minus Ratings: New and Improved for 2007-2008. https://www.82games.com/ilardi2.htm.Search in Google Scholar
Kaytoue, M., A. Silva, L. Cerf, W. MeiraJr, and C. Raïssi. 2012. “Watch Me Playing, I am a Professional: A First Study on Video Game Live Streaming.” In Proceedings of the 21st International Conference on World Wide Web, ACM 1181–8.10.1145/2187980.2188259Search in Google Scholar
KC. 2017. Dota 2: Here’s Why Radiant has a Lot of Advantage than the Dire. https://www.itechpost.com/articles/77468/20170125/dota-2-heres-why-radiant-has-a-lot-of-advantage-than-the-dire.htm (accessed December 13, 2018).Search in Google Scholar
Kubatko, J., D. Oliver, K. Pelton, and D. T. Rosenbaum. 2007. “A Starting Point for Analyzing Basketball Statistics.” Journal of Quantitative Analysis in Sports 3.10.2202/1559-0410.1070Search in Google Scholar
Macdonald, B. 2011. “A Regression-Based Adjusted Plus-Minus Statistic for NHL Players.” Journal of Quantitative Analysis in Sports 7.10.2202/1559-0410.1284Search in Google Scholar
Macdonald, B. 2012. “Adjusted Plus-Minus for NHL Players Using Ridge Regression with Goals, Shots, Fenwick, and Corsi.” Journal of Quantitative Analysis in Sports 8.10.1515/1559-0410.1447Search in Google Scholar
Marshall, A. 2019. 2018 US Open Prize Money to Reach $53 Million. https://www.usopen.org/en_US/news/articles/2018-07-17/2018-07-16_us_open_prize_money_to_reach_record_53_million.html.Search in Google Scholar
Matano, F., L. F. Richardson, T. Pospisil, C. Eubanks, and J. Qin. 2018. Augmenting Adjusted Plus-Minus in Soccer with FIFA Ratings. arXiv preprint arXiv:1810.08032.Search in Google Scholar
Okamoto, D. M. 2011. “Stratified Odds Ratios for Evaluating NBA Players Based on their Plus/Minus Statistics.” Journal of Quantitative Analysis in Sports 7.10.2202/1559-0410.1320Search in Google Scholar
PGA.com. 2019. 2018 PGA Championship: Purse and Winner’s Share. https://www.pga.com/events/pgachampionship/2018-pga-championship-purse-and-winners-share(accessed January 09, 2019).Search in Google Scholar
Ramler, I., C. Lee, and M. Schuckers. 2019. “Identifying Symbiotic Relationships between Champions in League of Legends,” in JSM Proceedings. Sports Statistics Section, 2362–9.Search in Google Scholar
Rosenbaum, D. 2004. Measuring How NBA Players Help Their Teams Win. https://www.82games.com/comm30.htm.Search in Google Scholar
Schuckers, M., and J. Curro. 2013. Total Hockey Rating (THoR): A Comprehensive Statistical Rating of National Hockey League Forwards and Defensemen Based Upon All On-Ice Events. URL https://statsportsconsulting.com/main/wp-content/uploads/Schuckers_Curro_MIT_Sloan_THoR.pdf.Search in Google Scholar
Semenov, A., P. Romov, S. Korolev, D. Yashkov, and K. Neklyudov. 2016. “Performance of Machine Learning Algorithms in Predicting Game Outcome from Drafts in Dota 2.” In International Conference on Analysis of Images, Social Networks and Texts, Springer, 26–37.10.1007/978-3-319-52920-2_3Search in Google Scholar
Sill, J. 2010. “Improved NBA Adjusted+/-Using Regularization and Out-of-Sample Testing.” In Proceedings of the 2010 MIT Sloan Sports Analytics Conference.Search in Google Scholar
Spann, M., and B. Skiera. 2009. “Sports Forecasting: A Comparison of the Forecast Accuracy of Prediction Markets, Betting Odds and Tipsters.” Journal of Forecasting 28: 55–72. https://doi.org/10.1002/for.1091.Search in Google Scholar
Stan Development Team. 2018. RStan: The R interface to Stan. URL https://mc-stan.org/, r package version 2.17.3.Search in Google Scholar
steamcommunity.com. 2019. STEAM Community Webpage. https://steamcommunity.com/app/570(accessed January 07, 2019).Search in Google Scholar
Thomas, A. C., S. L. Ventura, S. T. Jensen, and S. Ma. 2013. “Competing Process Hazard Function Models for Player Ratings in Ice Hockey.” The Annals of Applied Statistics 7: 1497–524. https://doi.org/10.1214/13-AOAS646.Search in Google Scholar
Tibshirani, R. 1996. “Regression Shrinkage and Selection via the Lasso.” Journal of the Royal Statistical Society. Series B (Methodological), 267–88.10.1111/j.2517-6161.1996.tb02080.xSearch in Google Scholar
Walker, E., and J. B. Birch. 1988. “Influence Measures in Ridge Regression.” Technometrics 30: 221–7. https://doi.org/10.1080/00401706.1988.10488370.Search in Google Scholar
Wickham, H. 2017. Tidyverse: Easily Install and Load the ‘Tidyverse’, r package version 1.2.1.Search in Google Scholar
© 2020 Walter de Gruyter GmbH, Berlin/Boston
