Prediction of basketball game results using machine learning algorithms: NBA & TBL

Download

Analysis and Prediction of Basketball Game Results - Caner Kahraman Thesis Report November 2022.pdf

Date

2022-11-30

Author

Kahraman, Caner

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

496
views

299
downloads

There are several factors affecting the basketball game results including team strength, home court advantage, resting, and momentum. Using advanced metrics and data analysis, it has become much easier for teams to measure the impact of these factors on the game results over the past few years, especially in the NBA (National Basketball Association). Only a few studies are performed related to the prediction of the basketball game results, which consider advanced team statistics and player specific factors together. This study analyzes the variables that affect basketball game results, including overall team strength via Four Factor metrics, home-court advantage, schedule, back-to-back games, momentum, and player-based variables such as maximum points per game for both NBA and TBL (Turkish Basketball League). Afterward, using the analysis findings of these factors, machine learning models are used to predict the game results in NBA & TBL for three seasons in 2016- 2017, 2017-2018, and 2018-2019 regular seasons. In this study, ELO Rating Model, Logistic Regression, Support Vector Classifier, Decision Tree, Random Forest, Naïve Bayes, KNN, LGBM, XGBoost, and Neural Network models are used. Analysis and results show that using superstar players to advance in the league is a valid option in NBA while it is not in TBL. Moreover, TBL is more predictable (up to 77.5%) than NBA (up to 67.5%) since there are power imbalances among teams in TBL and scheduling imbalances in NBA. Also using advanced variables has a better impact on the accuracies in NBA.

Subject Keywords

Sports analytics, Machine learning in basketball, Game result prediction, NBA, TBL

URI

https://hdl.handle.net/11511/101219

Collections

Graduate School of Natural and Applied Sciences, Thesis

Suggestions

OpenMETU
Core

Modeling mixed clustered outcome data with missing variables Kalaylıoğlu Akyıldız, Zeynep Işıl(2014-12-31) Projenin amaci asagida siralanan karakteristiklere sahip ve biyolojiden genetige sosyolojiden psikolojiye kadar pek cok alanda karsimiza cikan veri turlerini analiz etmek icin bir model ve yontem gelistirmek. Sozkonusu verilerin ozellikleri 1) Denekler ortak bir takım durumlara birlikte mağruz kaldığı için bu deneklerden elde edilen veriler ilişkili (böylece bu ilişkili denekler birer küme (cluster) oluşturuyorlar), 2) Her deneğe dair birden fazla cevap verisi varr ve bu korelasyonlu cevap değişkenlerinin ...
Fatigue does not affect the kinematics of free throw shooting in basketball Uygur, Mehmet; Korkusuz, Feza; Department of Physical Education and Sports (2009) Kinematic analysis of basketball shooting is evolving, however the effects of fatigue on free throw shooting have not been studied. Therefore the effects of fatigue on the kinematics of free throw shooting among elite male basketball players was assessed. Ten healthy male collegiate basketball players participated in the study. Resting and fatigue heart rates of the participants were measured. After a 15 minute warm-up period, markers were placed on seven locations on the shooting arm’s side upper and lower...
Prediction of the Ball Location on the 2D Plane in Football Using Optical Tracking Data Amirli, Anar; Alemdar, Hande (2022-01-01) Tracking the ball location is essential for automated game analysis in complex ball-centered team sports such as football. However, it has always been a challenge for image processing-based techniques because the players and other factors often occlude the view of the ball. This study proposes an automated machine learning-based method for predicting the ball location from players' behavior on the pitch. The model has been built by processing spatial information of players acquired from optical tracking dat...
Preliterate preschoolers oral explanationsand visual narratives in response to picturebooks Ulusoy, Mustafa; Altun, Dilek (null; 2016-05-01) Hikâye dinleme ve okuma, çocukların hem ev içi hem de okul öncesi eğitim kurumlarında gelişimlerinin eğlenerek desteklendiği önemli etkinliklerden biridir. Çocuklara sunulacak zengin içerikte, kaliteli çocuk kitapları çocukların gelişimlerini desteklemek adına büyük önem taşımaktadır. Bu nedenle, okul öncesi öğretmenlerinin piyasada yayınlanan kitaplar arasından çocukların gelişim düzeylerine uygun kitapları seçmeleri önem taşımaktadır. Öğretmen adaylarının okuma tutumları ve alışkanlıklarına bağlı olarak k...
Toplu Taşıma Hareketliliğinin Bölgesel Özellikleri: İzmir Kent Bölgesi Örneği Şenbil, Metin ; Yetişkul Şenbil, Emine; Özuysal, Mustafa (2021-05-27) Hareketlilik, erişebilirliğin ortaya koyduğu potansiyelin gerçek hayattaki karşılığıdır. Bir toplu taşıma ağı söz konusu olduğunda hareketlilik, yolcuların özelliklerinden istasyon ve durak çevresinin arazi kullanıma çeşitli özellikleri içerir. Bu çalışmada toplu taşıma hareketliğinin coğrafyası ve bunun ortaya çıkardığı bölgesel unsurlar incelenmiştir. Bu amaçla İzmir’de toplu taşıma kullanımına ilişkin bir haftalık elektronik kart verisi kullanılmıştır. Elektronik kart verisinde, kartın ilk okutulduğu ist...

Citation Formats

C. Kahraman, “Prediction of basketball game results using machine learning algorithms: NBA & TBL,” M.S. - Master of Science, Middle East Technical University, 2022.