One player's discrepancy between ratings and performance ... why?

[Danya]

I have a player in my league I've been playing for four years named Jung-Hwan Park. I drafted him with my Vancouver team fifth overall in my first draft. I always think up pretty extensive backstories/narratives for my fictional players and Jung-Hwan is basically my "favorite." I imagine that he was born in South Korea, immigrated with his family to suburban D.C. (Maryland) at a very young age, became a multi-sport star in high school, went to Notre Dame and is now on his way to being an Asian-American icon and Mr. Vancouver Thunderbird.

Anyway. Jung-Hwan's ratings show that he "should" be a consistent All-Star level performer:



As you can see, his year last year was good, but not that good. And with this year almost 2/3 done, his performance has been worse:



And his month-by-month splits this season have been the most confusing -- he started out on fire and has just gotten worse and worse:



Does anyone have any idea why he isn't doing better? Is this all within the margins of statistical random chance? I'm starting to feel like it isn't, but I might just not have the proper perspective. Any insight would be greatly appreciated!

[Overkill]

How are the players around him? Using an example from my own game, I have edited MANY players, and Giancarlo Stanton is capabale of a .278/.360/.584 season with 43 HR according to the editor. However, his numbers in my last two sims have been closer to his real-life performance of .237/.350/.435. I attribute this not to just him as a player, but the lack of ability of the players around him. There are also park factors and league totals modifiers, but having a strong supporting cast would probably benefit him here.

I am not saying this is your problem, but I do not think you should worry too much. His numbers are not too bad at this point and may even out by the end of the season.

[Danya]

Supporting cast is good ... I've got Geoffrey Baldwin (real-life scrub in Royals org who somehow metamorphisized into a 1.000+ OPS monster last year and this), Jemile Weeks and Brett Lawrie who've been doing pretty well (though also worse than last year/their ratings would indicate), a Japanese import named Toki Matsuo who's been OPSing around .800. Park is pretty much neutral. You're right that performance could even out by the end of the season though, he'd have to go on a pretty big tear but I guess his April has shown that he can do that.

[injury log]

Variance explains a lot in baseball. Notice also his potentials are lower than his ability, so he's going to get worse.

[Lukas Berger]

So far as I understand, the statistical prediction in the editor is made in a vacuum. It doesn't take into account the makeup of your specific league.

So it may not be accurate in your specific league. If you have a ton of really highly rated players in the league, then some players will significantly under perform their projections. And vice-versa. The ratings only determine how good the player is in relation to the other players in the league, and thus what amount of the league's statistical output an individual player can get. The ratings don't set an absolute level of performance.

Given that he's made many rating modifications, I'd guess mostly increases, that's most likely the issue that Overkill is seeing causing Stanton to "underperform" in his league. It may well be the case for you too.

Beyond that, it really does seem as if the performance/ratings discrepancy isn't all that large. It appears to be well within the bounds of potential statistical variance.

It's also noteworthy that your guy has a massive platoon split. If you're playing him a great deal against lefties his numbers will look worse than you'd otherwise expect. He looks like he's actually an average hitter at best when facing lefties. He might be a good candidate for a strict L/R platoon.

[JMDurron]

lukasberger covers the "overall talent being high could cause players to underperform their neutral context editor stats" angle, but another thought is that if your league has a large population of LHP, Jung-Hwan Park does not seem to be capable of handling them very well. If that projection is based on some kind of 80/20 RHP/LHP split, but Park is facing a 60/40 split in your league, then that would hurt his production quite a bit with those ratings.

[beorn]

Posting his splits versus right and left handed pitching would be interesting and useful.

Another interesting thing to look at. Go to the Positional Strength Report. How does he rate compared to other players at his position around the league? This will give you a quick sense of the talent level in your league. If it is unusually high you can expect players to generally underperform versus their editor predictions. (Although admittedly that is an indirect measure. The real question is the level of pitching around the league.)

All this aside, I am pretty sure there are other hidden factors that affect player performance, which is why it is risky to take on a player with good ratings but a poor performance record. Much more often than not, they will continue to underperform their ratings. But that is realistic -- some players appear highly skilled but never play up to it in real life, too.