Posted by sjpoker on 6/27/2017 8:25:00 PM (view original):
If the point is that WAR can compare performance across Eras, here I think it fails.
You aren't going to like this.
I get your argument. How could you look at the year Scherzer had--tons of innings, 143 ERA+, sub-1 WHIP, tons of K's, Cy Young, etc.--and say it was just as valuable as Brad Radke's 2000 season? It doesn't make sense.
But WAR doesn't look at those stats. It looks at the pitcher's runs allowed and considers what a league average pitcher would do against the same opponents, in front of the same defense, in the same ballpark.
In 2000 Radke allowed 4.73 runs per 9. An average pitcher in the same situation allows 6.40. Over 226 innings, he was 41 runs better than average.
In 2016 Scherzer allowed 3.04 runs per 9. An average pitcher in the same situation allows 4.51. Over 228 innings, he was 39 runs better than average.
Scale both of those to replacement instead of average and you get the same WAR.