Table 1. Prediction accuracy of the quantum chemistry and group contribution method modeling approaches.

	G1 (n = 8) Carboxylic Acid to Carbonyl	G2 (n = 59) Carbonyl to Hydroxycarbon	G3 (n = 23) Carbonyl to Amine	G4 (n = 15) Hydroxycarbon to Hydrocarbon
Quantum Chemistry	MAE = 45 mV Pearson r = 0.43 R2 = 0.19 No. params. = 2	MAE = 31 mV Pearson r = 0.59 R2 = 0.35 No. params. = 2	MAE = 17 mV Pearson r = 0.70 R2 = 0.49 No. params. = 2	MAE = 34 mV Pearson r = 0.45 R2 = 0.21 No. params. = 2
Group Contribution Method	MAE = 52 mV Pearson r = 0.54 R2 = 0.17 No. params. = 6	MAE = 34 mV Pearson r = 0.48 R2 = 0.21 No. params. = 13	MAE = 31 mV Pearson r = 0.22 R2 = -0.23 No. params. = 5	MAE = 66 mV Pearson r = 0.16 R2 = -3.39 No. params. = 6

The number of available experimental values for each reaction category is indicated in parentheses. MAE = Mean Absolute Error; R² = coefficient of determination. Note that for the G1 category, quantum chemistry has a lower MAE, but GCM has higher values of Pearson r. While the Pearson r can range from -1 to 1, R² can take on any negative value. A prediction method with the same accuracy as the mean predictor (a constant model that always predicts the mean value of the experimental data) has a value of R² = 0; negative values of R² indicate prediction accuracies that are worse than the mean predictor. GCM estimates of standard redox potentials were obtained from the implementation by Noor et al. [36,37] used by eQuilibrator (see Methods).