Hi kids. Let’s talk about predictive analytics and the 2018 Royals!
First things first. Hide all the sharp objects. If you have brown liquor in your house, lock it up and give your spouse/dog/neighbor/priest your key. Plan to go on a brisk stroll and enjoy a soothing shower after this is over with.
Craig and David, who are smarter than me or you or practically everyone either of us probably knows, already tackled the pitching and hitting PECOTA projections, already laid those out in all their bleakly unhappy glory. Ninety-six losses. Tied with a Marlins team that traded everything short of the Versace Mansion in their quest to alienate their entire fanbase and tarnish Derek Jeter’s legacy rebuild to their championship glory days. The dog days of summer will begin in April, if you believe the numbers.
I elected to wait a few days before tackling projections or anything to do with them for a few reasons.
- I am very dumb and it takes me awhile to understand things.
- I keep thinking Eric Hosmer or Mike Moustakas will be coming home, and if that’s the case the PECOTA numbers will be a little off.
- Because I think Ned Yost will always find a way to…okay the bosses at Baseball Prospectus aren’t reading anymore. Sometimes, numbers are wrong. In this case, one hopes they’re wrong in the Royals favor.
The WARP numbers speak for themselves, loudly, in a way that makes one choke back vomit. Jorge Soler, who could barely scratch consistent playing time last year, is tied for the highest projected regular with Whit Merrifield, who couldn’t break camp with the big club last season. For some people (me, namely), this is vindication that Soler’s burial last season was unnecessary and a hindrance to his development. It’s also an indictment of what the 2018 team is going to be, and regardless of how bad one thinks things can (or will) get … c’mon. Soler as the best Royal by WARP (and second-best, behind Salvador Perez, via VORP) is ridiculous. By projection, the best OBP is supposed to belong to Soler at .330 … followed by Alex Gordon at .326.
Salvy, Whit and Jorge Bonifacio are back … yet Gordon, who was very garbage last year and displayed no signs of improvement, should be relied on to get on base? Even if you think he’s got a bounceback in him, why should you feel compelled to trust that just because a computer said so?
And if you don’t like the lineup (and you shouldn’t, because computer or no, it’s a pile of flaming poo), the pitching staff projections will make you wish to chug chloroform. None of the projected starting rotation are projected to produce more than 11 quality starts, with none of the five projected to hit a 50 percent quality start rate.
Wait. That seems low. Is that low?
Turns out, it is low. Lower than any projection for any club outside of Baltimore, which has a rotation that is bad and has been bad for several years, and at least gets to make up the (projected) difference of its pitching shortcomings by having Manny Machado. The Royals are paying $40 million for this privilege; the Orioles are … um … not paying that. In fact, the Orioles salary figures for the projected rotation wouldn’t quite pay Ian Kennedy for the year. But this tale gets more worrisome.
Sixty-three total quality starts is 2017 White Sox territory. And that could be charitable — the league-average last season was 71 for the year. Sixty-three is basically one fewer a month; you can weather that with a good bullpen, which the Royals may have!
Seventy-one also was the lowest average quality start number across the league since 1995 (which was shortened thanks to the strike; I went back to 1990, but in terms of modern baseball, anything that happened before 1990 may as well have been as prehistoric as The Dick Van Dyke Show). Twenty-two years—through the steroid era and the offensive explosion and everything else, teams could typically rely on a starter to get through six innings with three earned runs or fewer on the board at least 72 times a season.
| Season | Quality Starts (League-Average) | Percentage |
| 2017 | 71 | 44% |
| 2016 | 75 | 47% |
| 2015 | 81 | 50% |
| 2014 | 87 | 54% |
| 2013 | 85 | 53% |
| 2012 | 83 | 51% |
| 2011 | 87 | 53% |
| 2010 | 86 | 53% |
| 2009 | 78 | 48% |
| 2008 | 78 | 48% |
| 2007 | 77 | 48% |
| 2006 | 76 | 47% |
| 2005 | 82 | 50% |
| 2004 | 76 | 47% |
| 2003 | 79 | 49% |
| 2002 | 79 | 49% |
| 2001 | 78 | 48% |
| 2000 | 75 | 46% |
| 1999 | 75 | 46% |
| 1998 | 80 | 49% |
| 1997 | 80 | 49% |
| 1996 | 74 | 46% |
| 1995* | 67 | 46% |
| 1994* | 56 | 49% |
| 1993 | 82 | 50% |
| 1992 | 87 | 54% |
| 1991 | 85 | 53% |
| 1990 | 85 | 52% |
I get that baseball has been trending toward expanded bullpens and getting the best work out of starters, not the most work. A six-percent drop in quality starts across the league over two years is a lot.
Now take into consideration what we know (or at least suspect) about the ball and MLB’s “Nothing to see here!” response to concerns about it and it’s easy to see runs staying up this season. Add in that starters work less innings than ever before anyway and it’s not hard to pick the thread of this narrative back up at “Kelvin Herrera is working for the fifth day in a row” and “Now in for his 88th appearance, Brandon Maurer.”
There’s science and logic behind this. The computers that spit this out don’t hate the Royals. A machine doesn’t want you, personally, to be miserable for six months while watching seven-through-nine in the lineup go 2-for-29 during a three-game series. No one’s MacBook Pro is vindictive enough to wish that much Onelki Garcia into anyone’s life. That’s not fun; baseball is supposed to be fun, even if the Royals are unlikely to be much fun in 2018.
(FUN FACT: If you recite this entire article out loud and record it, then play it backward, it’s actually an impassioned plea for Eric Hosmer to fire Scott Boras, accept the Royals deal and at least guarantee one competent ballplayer in blue this season. This is my version of The White Album. Moose is dead.)