San Francisco Giants pitcher Logan Webb works against the New York Yankees during the first inning of a baseball game in San Francisco, Wednesday, March 25, 2026. (AP Photo/Jeff Chiu)
Copyright 2026 The Associated Press. All rights reserved
There are a good number of baseball accounts on the social media platform X that are worth a follow, and @notgaetti is right up there among them. The operator of that account is a true baseball fan, and I would consider him a skeptic of advanced statistical metrics. His regular “NEPTA” (Not Enough People Talk About) posts tend to highlight players who are less appreciated by the modern, more statistically inclined baseball fan.
Now, as I can chalk up my career in baseball to my knowledge and development of advanced metrics (they got me in the door, and gave me the opportunity to learn to scout and round out my skill set), you might think I’d have an issue with @notgaetti. Not at all.
He’s right – staunch advocates of statistical analysis can be too firmly entrenched in their own world, just as scouting-only, traditionalist advanced stat-deniers can be too entrenched in theirs. The tools inform the numbers, and the numbers inform the tools – utilizing a mix of approaches is always best.
And @notgaetti is at his best – and his funniest – when he points out some of the outlandish WAR (Wins Above Replacement) comparisons one can make. Kevin Kiermaier accumulated 36.2 bWAR in his career, compared to Dante Bichette’s 5.7. Even when you’re only comparing offense, Kiermaier beats Bichette, despite the latter having major advantages in the traditional counting stat categories. Park factors, with Bichette’s Coors Field-powered numbers, are largely the reason why. I get it, and I think @notgaetti gets it to some degree, on some level as well. But it is worth discussion – and it is jolting at first glance.
You see, Coors Field doesn’t affect all hitters the same way – but the park factors are applied equally to all hitters, and that doesn’t make sense. Bichette WAS massively aided by Coors, hitting 177 of his 274 career dingers at home, posting a huge home/away SLG differential (.573 to .424).
But what about a ground ball/line drive hitter like DJ LeMahieu? He hit .299-.352-.408 in his seven years as a Rockie, and after he was docked heavily for the Coors Effect, that was good for only a 93 OPS+. Then he took his act to Yankee Stadium – a much more pitcher friendly park – and hit .278-.351-.412, almost his exact Rockies’ line, translating to a much better 110 OPS+. He was basically the same player for both clubs, with the advanced metrics only serving to confuse the issue.
Which brings us to what will go down as @notgaetti’s greatest hit/service. The Yankees’ Max Fried and Giants’ Logan Webb faced off in San Francisco on Opening Day. It was no contest, with the Yankees prevailing 7-0. Fried went 6 1/3 shutout innings, allowing two hits, with four strikeouts and one walk. Webb lasted 5 innings, allowing seven runs (six earned) on nine hits, with seven strikeouts and one walk.
Pretty cut and dry – Fried outpitched Webb, no matter the surrounding circumstances, right? Well, the WAR calculators tilted in favor of Webb, believe it or not, with the Giant righty posting Baseball Reference and Fangraphs FIP (Fielding Independent Pitching) marks of 1.54 and 1.61, compared to Fried’s corresponding values of 2.83 and 2.90. Ridiculous, for sure, and very appropriate that @notgaetti highlighted this in a post.
Where did the advanced metrics get it wrong? They properly valued Webb’s strikeout edge, but then basically valued all of the batted balls allowed by both pitchers equally. Therefore, Webb was a better pitcher on Opening Day, and it was good fortune that tilted the game in the other direction. Wrong.
Even Fangraphs’ xFIP metric, which attempts to take the types of batted balls allowed by pitchers into account, gave Webb a strong advantage, by 2.79 to 4.29. Yes, it’s one start for each pitcher, and no one should overreact to any method’s evaluation of it, but there is no way Webb had a better start than Fried. The scout in me knows this, and the analyst (and the fan) in me knows it too.
Well, there are analytical methods that get it right. My “Tru” ERA- statistic takes both batted ball type and authority into account. Both pitchers allowed 17 batted balls on Opening Day. Webb actually limited contact authority better than Fried (by 89.8 to 91.7 mph) overall. But Webb allowed a 105.7 mph fly ball, while none of Fried’s flies were over 91.8 mph. And most importantly, Webb allowed eight line drives, Fried only two.
Put it all into the hopper – the Ks, the BBs, the type and authority of contact allowed – and Fried had a 2.68 “Tru”-, way better than Webb’s 5.83 mark.
Yes, it’s still only one start – but it’s properly reflected as a good one for Fried, and a bad one for Webb. The numbers match the eyes. Both made their second starts last night, adding another data point over a long season.
There are a lot of ways to love baseball. Know and appreciate its history, its randomness, and the jaw-dropping abilities of its players, past and present. My wife, not a huge fan, loves the ambience of simply being at the ballpark. Advanced metrics provide new ways to measure performance, but are not the be all and end all. Take the opportunity to learn something about the game from someone who comes at it from a totally different perspective than you. This game gives daily rewards.
Source: https://www.forbes.com/sites/tonyblengino/2026/04/01/webb-fried-opener-shows-flaws-of-advanced-metrics-we-can-do-better/
