I'm perfectly fine with using both advanced and traditional statistics to evaluate player performance. The main point that I'm trying to make is that Berrios produced excellent results in 2024 and as such was essentially worth the money he was paid as a result. I'm not a fan of cherrypicking FIP and the associated FWAR on their own as it completely ignores what actually happened on the field in favor of a theoretical result that would occur if there a baseball league that existed without any fielders on the field of play. I like to look at more than just FIP/FWAR and have a peak at xERA, Baseball Reference WAR values, RAR etc.
For instance by RWAR, Berrios was worth 2.4 wins in 2023 and 2.2 wins in 2024. By RA9 WAR Berrios was worth 3.8 wins in 2023 and 3.7 wins in 2024. Yet despite essentially producing identical results by these two measures Berrios was only worth 1.0 wins in 2024 vs 2.8 wins in 2024. I just tend to think that it's sort of ridiculous to completely ignore what happened on the field of play in terms of actual results.
I tend to think the same thing with position player performance. I've raised this example in the past but I think it would be exactly the same thing to only look at xWOBA for a player when evaluating their season, and completely ignoring the on the field results. That's what it's doing with pitchers in essence, and it's just incredibly inconsistent that they utilize completely different methodologies between pitchers and position players as neither of them has any control over what happens to batted balls on the field, and as such each of them are prone to wild swings in batted ball results as a result. This would be like if Fangraphs used xWOBA instead of wRC+ for offensive contributions. Vlad produced a .378 xWOBA in 2023, so if you ignore the disappointing 118 wRC+ and used the expected statistic Vlad would likely rate as closer to a 3-3.5 win player that season instead of the 1.3 value he was credited with.