Close
today i learned (til)

Benchmarking Machine Learning for Basketball Performance

December 10, 2025

A recent study published in Knowledge and Information Systems shows us an analysis of machine learning (ML) models for predicting the future performance of  basketball players. This study tested 14 different ML models against 18 basketball statistics and Key Performance Indicators (KPIs).

Methodology and Data Focus

The researchers applied their methodology to a pool of 90 NBA players. To qualify, players had to meet strict criteria based on high averages across crucial KPIs like GameScore (GMSC), FOUR FACTORS, TENDEX, Fantasy Points (FP), and Efficiency (EFF) and the players had to have had a good amount of participation time (15+ minutes) over three recent seasons  (2019–20 up to 2021–22).

The study focused on developing individual scenarios for each player and the players averages in their past games in order to predict upcoming games. The 14 ML models tested spanned several categories, including Linear-Based models, Tree-based models, Non-Parametric (K-nearest neighbors or KNN), and Online-Learning models. To keep it simple, the study used different versions of models that best suit the measuring category.

In order to rank the models in comparison to each other, the authors made an evaluation metric called the weighted average percentage error (WAPE), which is derived from the weighted Mean Absolute Percentage Error (MAPE) results for each statistic and model.

Key Findings: Tree-Based Models Dominate

The research clearly showed that the Tree-based models are the best performers across most of the tested performance indicators.

The best models, based on performance against unseen data, were ranked as follows by WAPE score:

  1. Extra Trees (ET): Achieved the best performance with a WAPE of 34.14%. ET was a better predictor for metrics such as AST RATIO, PIE, PM, and TOV.
  2. Random Forest (RF): Scored a WAPE of 34.23%.
  3. Decision Tree (DT): Scored a WAPE of 34.41%.

While Tree-based models were really good, certain Linear-Based models like LASSO (34.54% WAPE) and LARS (34.53% WAPE) also showed exceptional overall performance, demonstrating good accuracy for specific metrics.

Optimizing KPI Forecasting

A crucial finding of the study was the success of a comprehensive forecasting approach designed to improve KPI prediction results, specifically focusing on Fantasy Points (FP).

Instead of forecasting FP as a single combined metric, the researchers experimented with forecasting the six constituent individual metrics (Points, Rebounds, Assists, Steals, Blocks, and Turnovers) separately, and then constructing the FP KPI from these predictions.

This optimized approach led to a significant improvement. The overall forecasting accuracy for FP improved by 3.6% MAPE on unseen data when utilizing this method. The Gradient Boosting Machine (GBM) model, for instance, achieved a MAPE of 29.81% when using the reformatted approach, which outperformed the best result from the single-metric forecast approach (LARS at 33.39% MAPE).

Implications

The findings show us lots of new  insights for the sports analytics industry, including researchers, coaches, data scientists, and organizations that invest in creating data science departments for scouting insights. The research not only benchmarks current ML model performance but also introduces an innovative approach to improve KPI forecasting accuracy, showing us that advanced statistical techniques combined with practical methods can set a new benchmark for predicting professional basketball player performance. Future work suggested by the authors includes incorporating external factors such as sentiment analysis, betting odds, and motion capture data to further refine predictions.

What does this mean for the future of sports? 

We are heading towards a future where statistical analysis can help teams attract the best talent and retain the best evaluate performers. The benchmarks are only getting better and more accurate. This begs the question of whether this is fair or not? Would it be fair for a team to cut a players contract based on data from a machine learning model? As these models start becoming more and more involved in team environments, the ethical considerations of using these models are going to start taking effect as well. So, what do you think; do you think these models are doing more good than bad for sports?

 

Reference

Papageorgiou, G., Sarlis, V., & Tjortjis, C. (2024). Evaluating the effectiveness of machine learning models for performance forecasting in basketball: a comparative study. Knowledge and Information Systems, 66, 4333–4375. https://doi.org/10.1007/s10115-024-02092-9
0Likes
About Author

You May Also Like

today i learned (til)
Do You Live in a Kind Area?
today i learned (til)
Uncanny Valley
today i learned (til)
Are sustainability trends legit?
today i learned (til)
Introduction: The Problem with Wet Raincoats

Your journey starts here! Let’s talk!

ThemeRex © 2026. All rights reserved.
  Timothy Durham was convicted of rape when a test showed his DNA genotype matched the DNA recovered from the crime scene. Eye witnesses testified that Durham was in a different state at the time of the incident, but he wasn’t exonerated from his 3000 year sentence until follow up DNA analysis showed that the preliminary forensic analysis used in the trial was misinterpreted (Thomas et al.). DNA testing serves an invaluable part in our criminal justice system, but it comes with complications and bias that can lead to wrongful convictions and misinformed juries.  

Background:

Forensic genetics aims to identify the origin of a biological sample. It can be traced back to Karl Landsteiner who discovered human ABO blood groups. With blood groups, suspects could be eliminated based on their blood type. For example, if a suspect had type A blood and there was type B blood left at the scene by the perpetrator, the investigators could rule out the suspect. Unfortunately, the reverse didn’t work. If the suspect and perpetrator both had the same blood type, it can’t be concluded that the suspect was guilty. Blood identifiers evolved into 8 distinct categories which improved the ability to eliminate suspects, but could only conclude that (if the system matched) it might be the perpetrator or this person might be related to the perpetrator (Jobling and Gill). The field evolved in the 1980s with the discovery of Minisatellites, lines of repetitive DNA that don’t encode proteins. Minisatellite patterns in DNA create what we know today as the DNA fingerprint, and it has a much higher accuracy rate than previous methods (Augustyn et al.). Around the same time, a breakthrough for victims of sexual assault was discovered. In the case of assault, bodily fluids of both the victim and the assaulter can get mixed. A process called differential lysis was created which could selectively increase the concentration of sperm in a fluid sample thereby allowing for accurate DNA testing (Jobling and Gill). Forensics evolved to use single-locus probe testing after several years because of its ease of interpretation. It looked at specific minisatellites that had high variability between subjects. This was the first DNA-based method to be used in a criminal investigation. Two murder homicides from several years apart were connected using DNA evidence. The evidence was then used to exonerate an innocent man, and the guilty party was found when he tried to evade mandatory DNA testing for all men in the small town. His guilt was then reinforced when the DNA profiles matched, and he gave a confession (Jobling and Gill). Finding DNA at a crime scene doesn’t guarantee that you’ll find a murderer. There are issues with non-pure samples and limited samples that make DNA work difficult, and accreditation of facilities and new techniques and take time (Jobling and Gill).  

Subjectivity of DNA Interpretation:

In a 2009 publication, the National Academy of Science discusses the possibility for subjective influence in lab based forensics. The interpretation bias happens in complex cases where, for example, there could be a mixture of DNA. A study published in 2011 looked specifically at this issue. A gang rape happened in Georgia, and the mixture in DNA from the case was used to prosecute one suspect after another suspect took a plea bargain to testify against him. With DNA mixtures, there are three possible conclusions: suspect excluded, suspect cannot be excluded, and inconclusive. If the suspect is excluded, it means that there’s little to no possibility that his DNA is part of the sample. If the suspect cannot be excluded, it means that there’s a possibility that his DNA is part of the sample. In the trial, the DNA examiners who knew the context of the case concluded that the suspect being prosecuted could not be excluded. The 2011 study gave the DNA evidence from this case and the DNA profiles of the suspects to 17 forensic examiners. These examiners had no context of the case, and they all worked independently. Of the 17, only 1 came to the same conclusion (suspect cannot be excluded) as the examiners who knew the context of the case. Twelve of the 17 concluded that the suspect who was prosecuted and convicted had no relation to the DNA found in the sample. Not only does this show that subjectivity occurs due to the context of the case, but the 17 examiners came up with different findings based on their past experiences meaning that not all DNA analysis is objective. Some organizations like the Scientific Working Group on DNA Analysis Methods say that quantitative statistical evidence should be presented with every categorical conclusion, but the US doesn’t require this (Dror and Hampikian).

False Positives:

There are two categories that cause wrongful identification in a DNA report: coincidental matches and false positives. Coincidental matches are DNA matches between two unrelated people. The probability of these matches is about 1 to 500,000 for white individuals and about 1 to 350,000 for black individuals. This small probability of coincidental matches makes people feel confident in DNA evidence. False positives, on the other hand, are matches between DNA and test DNA that result from issues in the forensics lab. There have been multiple cases of forensics labs accidentally switching victim and suspected rapist DNA which lead to convictions that were later overturned when the lab realized the error. There are also cases of misreading or misunderstanding a sample. In a case from 1993, a man was convicted of rape because someone interpreted a mixed sample of victim and perpetrator DNA as an individual sample. The conviction was again overturned. To prevent false convictions, retesting DNA is allowed, but it’s not an efficient solution. Many people are pushing for statistical information on the probability of false positives, so that the jurors can make accurate judgements based on DNA evidence (Thomas et al.).  

Conclusion:

DNA analysis is a useful tool for the criminal justice system, but subjectivity can be an issue in forensic testing. To combat wrongful convictions, there are organizations like The Innocence Project who use DNA testing to free the innocent, but reformed forensic testing standards should be put in place to prevent subjectivity and to ensure juries are correctly informed during trials. https://www.innocenceproject.org/    

Citation:

Augustyn , Adam, et al. “DNA Finerprinting.” Encyclopedia Britannica , 26 Apr. 2018, www.britannica.com/science/DNA-fingerprinting#ref153186. Dror, Itiel E., and Greg Hampikian. “Subjectivity and Bias in Forensic DNA Mixture Interpretation.” Science & Justice, vol. 51, no. 4, 2011, pp. 204–208., doi:10.1016/j.scijus.2011.08.004. Jobling, Mark A., and Peter Gill. “Encoded Evidence: DNA in Forensic Analysis.” Nature Reviews: Genetics, vol. 5, no. 10, Oct. 2004, pp. 739–751., doi:10.1038/nrg1455. Thompson, William C, et al. “How the Probability of a False Positive Affects the Value of DNA Evidence.” Journal of Forensic Sciences, vol. 48, no. 1, Jan. 2003, pp. 1–8., doi:10.1520/jfs2001171.