With the regular season coming to a close, our perceptions of various players have certainly changed since the beginning of the season. How have the projections' perceptions changed? That's what I wanted to know. The two best public projection systems, Steamer and Zips, release their projections before the year begins, but they continue to update those projections throughout the year as players accumulate new stats.
I certainly don't think the projections are the final say. If I did, I wouldn't take the time to dive into a player's various stats a couple times a week; I could just look at a single number. However, the projections do provide an unbiased point of view, based on the historical performance of other players similar to the player in question. It's a lot easier than delving into the stats of each individual player and jamming it into one post. They're also a good starting point for discussion.
With that in mind, I grabbed all the preseason projections for all MLB players appearing on fangraphs' opening day Depth Charts, as well as the projections for every MLB player that is assigned playing time on their updated rest of season Depth Charts. I used some code to match the two groups up by player.
First, we'll look at position players. We'll just look at wOBA for two reasons: (1) people have complained in the past about rest-of-season defense projections being wonky and (2) the plate appearances remaining are really small, and since projected WAR is rounded to the nearest tenth, extrapolating that out to WAR/600 PA isn't very reliable without the unrounded number. I bet you can guess who leads the way in terms of improving their projection:
A .370 wOBA on the year has pushed Aledmys Diaz's projected wOBA from solidly below average to right at average. I wrote on the prospect of Carpenter exceeding projections before the season began, and he has done so, raising his wOBA by the second largest margin.
It's not surprising that Yadier Molina, Greg Garcia, and Jedd Gyorko have all also raised their stock. Jeremy Hazelbaker has raised his stock as well, but still sits at a pretty low .293 wOBA projection. I was surprised though, to see such small increases for Stephen Piscotty and Brandon Moss. However, Moss was projected for a bounce-back from his poor 2015 performance anyway, so he really only did a little better than expected.
For all the discussion, demotions, and promotions, Kolten Wong and Randal Grichuk's projections are virtually unchanged. Being that both players were struggling in the contact quality area, that's not too surprising. Results on balls in play and homers have a higher variance attached to them, so the projections weren't convinced that anything real was going on when those two were struggling.
Matt Holliday, Brayan Pena, and Jhonny Peralta bring up the rear, all losing 4 points of wOBA. With average wOBA rising this year in a more offensive-heavy environment, they probably would have otherwise lost more points. However, one shouldn't give up on either of those players. Matt is still projected as an above average hitter, and Jhonny is still a slightly below average hitter who can play all the non-battery infield positions. Of course with Jhonny, the underlying contact quality stats are down, and we all know how thumb injuries can sap power upon return. It's easy to take the under on Peralta's projection now, but when he's fully healed in 2017, he could still be useful.
Let's get back to Diaz though. Here's his projections and 2016 production again, but broken down into the four core hitting stats:
Diaz has been much better than advertised in each category, and he's projected better than he was preseason in all four. It's easy to imagine Diaz being better than his current projection though, because we've seen him better all-around this year. Even so, let's look at how Diaz compares against the 469 other MLB players projected at the beginning of the season and now. Here's the Top 10 among all position players:
Fifth best out of 469, not to shabby. Diaz has raised his stock more than almost any other position player in the game this year, which is pretty impressive, even if that means he still just sits at average.
Let's move on to the rotation:
Luke Weaver had yet to throw a pitch in Double-A by the time the 2015 season ended, so his ascendance is pretty incredible. Alex Reyes had less than 35 innings at Double-A, and of course had a much higher pedigree as the team's unanimous top prospect before the year began. Wacha, curiously, has had a strong increase, but to be honest I think it might just be from adjusting his projection from a starter to a reliever, which creates quite an improvement in expectations. Leake is still projected exactly the same, and it gets rough from there.
Carlos had a great season last year and I expected even more this year, but things have went backwards. He's still projected well going forward though. Wainwright's change in projection is probably more concerning, given his age. Jaime Garcia takes the biggest fall, despite the fact that just a few months ago he was one of the best pitchers in baseball over the last year.
Garcia we've talked about a lot recently, and to be honest, with him probably being gone at the end of the year, I'm just not as concerned about him. Let's look at Waino's projections and results a bit closer:
In good news, Waino's already below average strikeout rates haven't slipped further. But he's walking more guys and he's not expected to get back where he was expected before the year. The homer rate is up, but it's subdued a bit by the fact that it's up league-wide (1.02 to 1.18). No, Waino's not garbage now by any means, he's just not A.D.A.M. anymore, not exactly a controversial position for the projections to take.
Let's get back to positive news though. For instance, you might be wondering where Weaver's 82 point drop ranks against the rest of the league. Well, being that he hadn't thrown a pitch at Double-A until this season, Weaver wasn't projected for any innings pitched in the preseason projections, so he actually wasn't included. I looked up his Steamer projection (Zips didn't project Weaver before the season) and added it in manually. Same goes for Reyes. If they had they been included, they would have rated third and sixth respectively. However, there's probably other pitchers missed by my method that could have improved expectations similarly, so who knows where they'd rank if we included everyone.
Still though, great news for the Cardinals, Weaver, and Reyes. Despite some bumps along the way, both have been about as good as you could reasonably expect when the season began. Lastly, let's look at the pen:
Bullpens can frequently feature a lot of turbulence, and the 2016 Cardinals are no different. The three biggest gainers this year have also become three of the most used. Unfortunately, Kevin Siegrist is in a free fall, and has still been used excessively. He's lowered his stock even more than demoted closer Trevor Rosenthal. Seth Maness missed this list because he's not projected for any innings going forward, but his projected FIP has rose 20 points. It's crazy to think there's some chance that Rosenthal, Siegrist, and Maness all have a chance to not be part of the 2017 Cardinals.
Remember though, the projections are, at best, a great starting point of discussion. Every player is unique, and has aspects of their game not evaluated by the projections. Besides that, predicting baseball is hard, and that's why they're called projections and not predictions. Who do you think the projections are low on going forward? What about high?