Table 5 Summary of predictive performance with the test set, the root mean square error (RMSE), the concordance correlation coefficient (CCC) and its 95% confidence interval, the Pearson correlation coefficient (r), the accuracy measure (\(\text{ C}_{b}\)), the number of observations in the test set (\(N_{\text {Test}}\)) and the number of observations in the training set (\(N_{\text {Training}}\)).
From: A hierarchical approach for evaluating athlete performance with an application in elite basketball
Test set | RMSE | CCC | r | \(\text{ C}_{b}\) | \(N_{\text {Test}}\) | \(N_{\text {Training}}\) | ||
|---|---|---|---|---|---|---|---|---|
\({\hat{\rho }}_{CCC}\) | Lower | Upper | ||||||
Season 2015–2016 | \(6.810 \times 10^{-4}\) | 0.9612 | 0.9603 | 0.9622 | 0.9615 | 0.9998 | 28,978 | 81,369 |
Season 2016–2017 | \(8.492 \times 10^{-4}\) | 0.9558 | 0.9548 | 0.9568 | 0.9631 | 0.9925 | 29,099 | 81,248 |
Season 2017–2018 | \(8.254 \times 10^{-4}\) | 0.9562 | 0.9553 | 0.9572 | 0.9627 | 0.9932 | 26,165 | 84,182 |
Season 2018–2019 | \(8.892 \times 10^{-4}\) | 0.9526 | 0.9515 | 0.9537 | 0.9598 | 0.9925 | 26,105 | 84,242 |
\(10\times 4=40\) Games\(^{*}\) | \(6.831 \times 10^{-4}\) | 0.9639 | 0.9569 | 0.9710 | 0.9647 | 0.9992 | 877 | 103,289 |
\(50\times 4=200\) Games | \(6.862 \times 10^{-4}\) | 0.9637 | 0.9597 | 0.9677 | 0.9643 | 0.9993 | 4,218 | 99,948 |
\(150\times 4=600\) Games | \(6.882 \times 10^{-4}\) | 0.9637 | 0.9617 | 0.9758 | 0.9644 | 0.9993 | 11,925 | 92,201 |
