Jankowski projection

[Tom_T]

Travis Jankowski has been consistent right out of the gate, and his monthly splits are good. Not sure why he merits a .190 BA projection the rest of the way. xBA does not support that.

[JonE]

Well, if you subtract his YTD results from his season-long projection, that's what you get. His overall projection has him at .256, which is a nice bump for a career .232 hitter, but he's already hit 50 points higher than that, so the projection anticipates negative regression. That's reasonable when you see his H% is 37%. On the plus side, as you say, his xBA is currently .272, so maybe he won't get all the way down to .256. Still, if you bump his season-long projection to .272, that only translates into 5 extra hits, which would make his ROS BA just .225.

It's an interesting case study, of course. How much weight should we give to sudden improvement in a part-time 32-year-old player with a total of 261 plate appearances in the last four seasons? Some of his peripherals recall his early level of performance, but because he's been so inconsistent and had so few appearances, he's especially hard to project accurately.

Because he's a part-timer, he's only ownable in an AL-only league, or in some mixed leagues because of the SB. I wouldn't trade anything of substance for Jankowski, but if I owned him I'd hang on and hope for the best.

[Tom_T]

This reminds me of a debate that took place on the Forum many years ago. Question was: if a player's preseason projection was for him to hit .280, and he hit .240 after two months, does the remaining projection compensate so that he ends up at .280? And the consensus was, no. Regardless of what he hit in April and May, if he was projected to hit .280, then he's projected to hit .280 the rest of the way, and maybe he ends up at .270. A couple of statistical gurus on the site at the time were extremely adamant on that point.

[RAY@HQ]

This reminds me of a debate that took place on the Forum many years ago. Question was: if a player's preseason projection was for him to hit .280, and he hit .240 after two months, does the remaining projection compensate so that he ends up at .280? And the consensus was, no. Regardless of what he hit in April and May, if he was projected to hit .280, then he's projected to hit .280 the rest of the way, and maybe he ends up at .270. A couple of statistical gurus on the site at the time were extremely adamant on that point.

That's not how our projections work, actually. The pre-season projection is weighted against the YTD performance, with YTD gaining more weight as the sample size grows.

[The Shadow]

That's not how our projections work, actually. The pre-season projection is weighted against the YTD performance, with YTD gaining more weight as the sample size grows.

{SARCASM}So, we can expect that by September 15, Carpenter will be projected to hit .240 if he keeps on hitting .166 for the next 6 weeks?{/SARCASM}