Well, everyone, it’s that time again. Specifically, I’m referring to the time where we review a few of the individual player projections that have been released for the 2017 Atlanta Braves season. This has all happened before (and will probably happen again), so I will try to dispense somewhat quickly with the introductory portions (in a Q&A format) and get right into the player projections themselves.
(Last year, I wrote that this exercise is essentially tantamount to pushing numbers at another set of numbers until a third set of numbers appears, and I stand by that assertion. Alternatively, this.)
What projections is this exercise looking at?
Three: Steamer, ZiPS, and a very rudimentary system I’ve come up with myself, which I jokingly call IWAG. Steamer and ZiPS are both featured on Fangraphs. As best I can tell, Steamer is maintained by Jared Cross, Dash Davidson, and Peter Rosenbloom. ZiPS, of course, is curated by the inestimable Dan Szymborski. The data from Steamer and ZiPS were collected off of Fangraphs’ individual player pages. The only adjustments I made were pro-rating the values to a different number of PAs, to put each projection on equal footing.
What stats is this exercise looking at (for hitters)?
Three: wRC+, Def/600 (Def/450 for catchers), and WAR/600 (WAR/450 for catchers). You can find a quick rundown of these in many places, including last year’s post or the Fangraphs glossary.
Essentially, wRC+ is a great catch-all offensive stat that credits hitters for good results at the plate (e.g., homers, walks, etc.) while ignoring anyone on base in front of them, the game situation, and so on. It’s park-adjusted, and the best thing about it is that it’s scaled to 100, meaning that a 150 wRC+ indicates a hitter produces 50 percent more runs than a league-average hitter, while a 75 wRC+ indicates a hitter produces 25 percent fewer runs, etc.
Def is Fangraphs’ method of combining how well a player fields his position with the inherent value granted a player for playing a given position. (This latter chunk is referred to as the positional adjustment.) The positional adjustment for catchers currently remains +12.5 runs over about 600 PAs, but since we are pro-rating catchers to 450 PAs, it’s about +9.4 runs instead. So, if you see a Def/450 in this post below 9.4 (and you will), that suggests that the catcher(s) in question is projected to be a below-average fielder. Pitch framing for catchers is not (yet?) included in Def.
WAR is the biggie, and attempts to quantify a player’s overall production, across hitting, fielding, and baserunning. For hitters, WAR is neutral with regard to context (whether runners are on base, or whether it’s a close game), so it’s really just an assessment of the specific outcomes the player was involved in, whether that’s hitting a double versus making an out, making a play versus failing to field the ball, or taking an extra base versus going station-to-station.
Why not counting stats? Aren’t you excluding most of what the projection systems actually do?
Yes, but I don’t find the counting stats that interesting. Is it possible to project RBI for every hitter? Absolutely, and Steamer/ZiPS do it. But RBI aren’t fully-influenced by the player’s own skillset, and don’t give you a great sense of the player’s overall production or skill level by themselves. I’ve chosen three manageable endpoints that give you the big things you probably want to know about a player: how well he hits, overall; how well he fields, adjusted for the difficulty of position; and, his total projected value for the upcoming season, over a full season’s worth of play.
Blah blah I know all this already. What’s changed since last time?
For hitters, not much. The only major thing I’ve been tinkering with (and still haven’t fully resolved) is a good way to align (or smush together) the additional hitting information available for players with recent minor league stints with the relative paucity of fielding information for these players, especially if the players spent entire seasons in the minors. (As you can tell, this is a big concern for the Braves, who are going to be leaning heavily on young players in the immediate future.)
Where this disconnect occurs, the Def and WAR values are going to be highlighted in gray, and will look somewhat weird, because these will be the cases where the listed wRC+ and the listed Def don’t suggest the listed WAR. The short explanation for this is that the Def and WAR projections I use (and again, this is only for IWAG, Steamer and ZiPS are much more sophisticated and have already solved this problem, if they’ve ever even encountered it) lean very heavily on the player’s major league results, while the wRC+ projections incorporate more minor league data, leading to a disconnect. If I had to choose, I’d go with the WAR values listed for these players over adjusting them to account for different (usually better) offense, but I think the minors-influenced wRC+ numbers are interesting in and of themselves, too.
The Braves used only three catchers in 2016 (A.J. Pierzynski, Tyler Flowers, Anthony Recker), and currently have three catchers on the 40-man roster (Tyler Flowers, Anthony Recker, Kurt Suzuki). Flowers and Suzuki will be the starting tandem, with Suzuki potentially relegated into a knuckleball-corraler role, and Recker appears to be the insurance policy biding his time at Gwinnett. As of the time of this writing, Recker has already been optioned down to the minors, so there’s not much of a chance that this set-up changes before Opening Day, though the Braves could always surprise everyone with another last-minute deal that shakes up the catching situation.
*Projections for other players can be provided on demand, but it’s doubtful that anyone wants to see projections for another catcher in the system that’s even remotely likely to make the big league Braves club in 2017.
The projections for these three amigos look like this.
Let’s quickly review, player-by-player.
Tyler Flowers
Tyler Flowers is very intriguing, but unfortunately for him, the intrigue doesn’t tend to show up in the bottom-line numbers. He rocks a career 88 wRC+, and his inability to do even a passable job of controlling the running game caught up to him in earnest last year. Pitch framing, which is one of his great skills, doesn’t (yet?) show up in the Def or WAR calculations.
Flowers’ saving grace is that he hits the ball quite hard (11th last season per the Statcast leaderboard, sandwiched between David Ortiz and Joc Pederson). Unfortunately, his launch angle isn’t great (too many groundballs), and he often experiences swing-and-miss issues.
With all these things together, it’s not surprising that Steamer and ZiPS basically have him looking pretty similar to his career line. IWAG, by contrast, is more heavily influenced by his 2016, where he made some gains in walk rate while cutting down on his strikeouts, especially his swinging strikes. He’s also put up a 94 wRC+ over the last three years, which is around when he became a primary catcher or at least half-time player, so IWAG is banking on that being a little more representative of his offensive production, at least for 2017.
His fielding has fewer reasons for optimism. No matter how you slice it, he’s been a below-average defensive catcher since 2013, and the extent to which he’s backslid has gotten worse with time (-1.1, -1.8, -3.9, -7.0, 2013 through 2016, in terms of fielding runs below average). Unless he figures out a way to address his issues with the running game, this probably won’t change, and teams will just run on him more, which will penalize him further. Steamer tends to heavily regress most defensive projections, so it’s not picking up on this as much, but ZiPS and IWAG are also conservatively figuring his defense won’t be as bad as last year’s travesty, where he allowed 60 stolen bases and threw out three (!!!) baserunners.
Put some below-average offense and some below-average defense together, and despite the hefty catcher positional adjustment, you’ve got a below-average player. Just how below average is up in the air.
Kurt Suzuki
It’s tempting to describe Kurt Suzuki as a bizarro Tyler Flowers who ultimately ends up no better or worse than him. Suzuki doesn’t strike out much (but also doesn’t walk much), his batted ball profile is pretty generic and features a lot of pulling the ball (whereas Flowers hits it up the middle a lot, given how hard he stings it when he makes contact) and he’s a better pitch blocker, but a worse framer. In something that doesn’t support the bizarro hypothesis, he’s also fairly bad at controlling the running game: if you excluded Flowers’ abysmal 2016 SB/CS numbers, you’d be hard-pressed to tell Suzuki and Flowers apart on that front.
The bottom-line similarities are pretty substantial. Suzuki’s career wRC+ is 86. He’s also rated as a below-average defensive catcher for the last four seasons. In many ways, this makes him a poor complement to Flowers: they don’t so much cover each other’s bases as they do falter in many of the same ways. The logic was potentially that Suzuki’s pitch-blocking would come in handy for handling R.A. Dickey’s knuckleball, but aside from this, they’re both right-handed catchers without rosy offensive outlooks that struggle to prevent stolen bases.
With all that said, Suzuki’s hitting talents are less interesting than those of Flowers, and as a result, there’s somewhat less room for the projections to maneuver and differentiate themselves. Suzuki is probably going to put up a wRC+ in the high 70s/low 80s, though looking at his past seasons that’s really a central estimate, since his wRC+s for the past three years have been 106, 65, and 86. He’s probably not going to suddenly get very good defensive marks. He’s basically a lower-tier backup catcher at this point, but the Braves are paying him $1.5 million to subject himself to the punishment of catching a knuckleball every five days, so expectations probably weren’t very high to begin with.
Anthony Recker
Anthony Recker may not catch a single pitch as an Atlanta Brave this year. But, that doesn’t mean he’s not worth a few sentences. Recker is a career 77 wRC+ hitter over about 600 major league PAs, but he’s played parts of six seasons in the majors. He’s also played parts of seven seasons at AAA, and has generally hit well there, so he’s basically your AAAA backup catcher.
The projections for him are kind of interesting in the, “Who knows what happens when there’s not much data to go on” way. Steamer appears to have him as a pretty decent backup, and not just a replacement level four-A player, by combining his poor hitting with some potentially decent defense. (The potentially decent defense is a central assumption, I’d guess, from the fact that he doesn’t have much major league playing time.) ZiPS agrees with Steamer’s hitting conclusion, but heavily penalizes his defense, dropping him to that four-A tier. And IWAG, well...
IWAG is perhaps foolhardily giving him credit for hitting the ball really hard in his 112 PAs last year, combined with a strong AAA track record. Maybe giving a 33-year-old catcher a projection boost for beating up on guys in their early-to-mid twenties in AAA is a mistake that needs to be rectified, but Recker also walks a fair bit and reined in his strikeouts last year. The 90s wRC+ is probably very optimistic, but taking poor defensive projections into account, and you have basically a backup catcher, which is again, what he is.
The Braves may have some exciting, above-league average catchers some day, but it probably won’t be 2017.
Next up: Freddie Freeman and the Gang, aka, Infielders
Bonus: Pitch Framing
While not included in fWAR or Def, pitch framing is still useful. Here are the 2014, 2015, and 2016 pitch framing numbers for the three prospective Braves catchers.
Courtesy of Statcorner:
- Tyler Flowers, calls per game: -0.34, +1.79, +1.30
- Tyler Flowers, Framing Runs Above Average: -5.6, +22.5, +13.3
- Conclusion: Tyler Flowers might be about 1 win more valuable based on his framing. That would potentially make him an average-y catcher!
- Kurt Suzuki, calls per game: -1.43, -0.07, -0.40
- Kurt Suzuki, Framing Runs Above Average: -19.8, -1.1, -5.0
- Conclusion: Kurt Suzuki might be slightly under 1 win less valuable based on his framing. If so, that makes him replacement level-y. Eh.
- Anthony Recker, calls per game: -1.95, -1.16, -1.81
- Anthony Recker, Framing Runs Above Average: -11.8, -3.1, -6.4
- Conclusion: Anthony Recker might be slightly under 1 win less valuable based on his framing. If so, that makes him replacement level-y or worse. Eh.
Loading comments...