This article explores the ranking of treatments in frequentist network meta-analysis, a method used to compare multiple treatments for the same condition. The authors propose a frequentist analogue to the Surface Under the Cumulative RAnking curve (SUCRA), called the P-score, which does not rely on resampling methods. The P-score is based on the point estimates and standard errors of the frequentist network meta-analysis estimates under the assumption of normality. It measures the mean extent of certainty that a treatment is better than other treatments. The authors demonstrate that the numerical values of SUCRA and P-scores are nearly identical through case studies on diabetes and depression data. They conclude that while both SUCRA and P-scores induce a ranking that largely follows the point estimates, they do not offer a significant advantage over looking at credible or confidence intervals. The article emphasizes the importance of considering both the numerical values and the precision of estimates when interpreting treatment rankings.This article explores the ranking of treatments in frequentist network meta-analysis, a method used to compare multiple treatments for the same condition. The authors propose a frequentist analogue to the Surface Under the Cumulative RAnking curve (SUCRA), called the P-score, which does not rely on resampling methods. The P-score is based on the point estimates and standard errors of the frequentist network meta-analysis estimates under the assumption of normality. It measures the mean extent of certainty that a treatment is better than other treatments. The authors demonstrate that the numerical values of SUCRA and P-scores are nearly identical through case studies on diabetes and depression data. They conclude that while both SUCRA and P-scores induce a ranking that largely follows the point estimates, they do not offer a significant advantage over looking at credible or confidence intervals. The article emphasizes the importance of considering both the numerical values and the precision of estimates when interpreting treatment rankings.