I go back to Tom Tango's quote about expected metrics quite frequently.
An interesting read if anyone is so inclined.
Properly Diving Into Expected Stats | Community Blog (fangraphs.com)
“Tango then stressed that the expected metrics were only ever intended to be descriptive, that they were not designed to be predictive, and that if they had been intended to be predictive, they could have been designed differently or other metrics could be used.”
For what it is intended to do, which is try to separate a player from extenuating circumstances and tell the story of a player differently, expected stats have accomplished their job. Asking it to do more than that (predict future success) would be asking too much based on the purpose of the metric when formulated, which is evident in the data.
The most popular metrics on the website, however, are their expected stats: expected batting average (xBA), expected weighted on-base average (xwOBA), expected on-base percentage (xOBP), expected slugging percentage (xSLG), and expected isolated power (xISO). Essentially, these statistics are what you’d expect based on the name; they indicate what a player’s “true talent level is” based on the quality of their contact, frequency of contact, and, depending on the batted ball, sprint speed.
This would appear to be a gold mine on the surface. With the ability to know what numbers a player deserves to have, we should be able to separate their talent level from outside circumstances, and thus better predict future performance. Yet that actually isn’t the case.