When I started my 2012 runs prevented calculations — calculating each NL starter’s expected, park-adjusted runs allowed against his actual runs allowed, I was shocked to learn that Bronson Arroyo had such a strong season for the Reds. After serving as a near-replacement pitcher in 2011, probably one of the worst in the National League, Arroyo will undoubtedly rank among the 20 best starters in the 2012 National League. Shaving 20 home runs and 10 walks from 2011, while striking out 20 more batters, Arroyo spun a modest 129 K/35 BB/26 HR performance into 17 runs prevented — he allowed approximately 17 fewer runs than his expected, park-adjusted runs average. While few would flock to 15.4% strike out rate, Arroyo limited the damage with his extremely low walk rate and passable home run rate (for Great American Ballpark, anyway).
On Tuesday, I analyzed the Giants’ and Reds’ pitching swap, as the Reds became a pitching dominant club, while the Giants were actually below average from the mound. Today, I want to highlight one of the keys to that swap, that of the collective performances of Tim Lincecum and Arroyo. These types of changes in expected performance and actual outcome are significant for a ballclub, because they run straight to the core of roster value. A club might ask, with a young, excellent pitcher such as Lincecum, whether they should offer that pitcher a contract extension; a club might ask with an aging, on-again, off-again veteran whether they should stick with that pitcher. While fans clamor for aces to lead their ballclubs to Championships, the reality of pitching fluctuations is frequently much messier.
What Makes an Ace?
Heading into 2012, Lincecum was one of the National League’s textbook aces. After pitching two exceptional seasons in 2008 and 2009 (preventing more than 40 runs in both seasons), Lincecum remained in the group of top pitchers in 2010 and 2011, preventing more than 15 runs against his league and park during both of those seasons. Of course, beneath the surface, even though Lincecum’s overall Fielding Independent Pitching performance looked good, his strike outs declined (like his velocity) as his walks increased and home runs fluctuated. Suddenly for 2012, his home run rate increased by 65% (using HR/PA), his walk rate continued its climb from 2009, his strike outs didn’t increase, and his average fastball velocity was the lowest in his career (I suppose that’s less a cause of his poor performance than an easy-to-write, newsworthy line. Like, “Bronson Arroyo Wins Championships.”)
The ace is missing. I bring this up because it’s interesting to present perceptions of whether a pitcher is an ace against actual performance. (If you read most sportswriters, just about anybody is an ace; most people write the word as though every team has an ace; “The Padres’ ace squares off against the Marlins ace today…”). I tend to favor the category of aces that is tiny — truly elite pitchers that maintain their performance over several seasons. Someone like me could hang my hat on Lincecum; if ever there was an example of an ace over the last four seasons, Lincecum was it.
212.3 IP, 84 R (16 runs prevented), 114 ERA+
202 IP, 86 R (17 runs prevented), 113 ERA+
One of the interesting things about judging pitchers is that a guy like Arroyo — veteran, unassuming stuff, innings eater, etc. — can come around and produce ace-worthy campaigns in one season. Arroyo’s 2012 is approximately as valuable (to Great American Ballpark) as Lincecum’s 2010 campaign was to the Giants and their park. In fact, Arroyo’s 2012 is every bit as good as a certain ace-in-waiting that the Brewers traded out west:
212.3 IP, 84 R (18 runs prevented), 114 ERA+
202 IP, 86 R (17 runs prevented), 113 ERA+
Yet, Arroyo’s not an ace; I wouldn’t call Greinke an ace either, although his Fielding Independent stats beg to differ; now, what do we do with Lincecum? Does one bad season knock him out of the “ace” category? I mean, aces don’t always improve, and sometimes they even pitch ho-hum seasons — look at Lincecum’s intradivision rival, Clayton Kershaw; an under-the-radar, “who knew?” 35 runs better than Dodgers Stadium. Would we knock Kershaw out of the ace category if he doesn’t pitch as well next year? Or pitches below average?
I am trying not to be facetious about this. It’s important to figure out what makes a great pitcher great for an extended number of years. Looking at Lincecum’s K/BB/HR constellation over the last few years, one would certainly expect him to pitch closer-to-average seasons, but not fall off the map; one would certainly expect him to pitch better than Bronson Arroyo. The same goes for the biggest pitching free agent of the off-season, Zack Greinke; no one outside of Cincinnati or Boston would take Arroyo over Greinke, and yet Arroyo’s been every bit as good as Greinke in 2010 and 2012. That 2011 stinker by Arroyo, that sets him apart; we know aces don’t pitch replacement level seasons (what does that mean for Lincecum?). Of course, Greinke was average in 2011 — which looks a hell of a lot better than 2011 Arroyo and 2012 Lincecum. Is ace-hood achieved by negative classification? That is, you’re an ace if you don’t pitch below average seasons, but as soon as you do, you’re out?
I’m trying to capture what forms our expectations of pitchers. If we compare Lincecum, Greinke, and Arroyo in their seasons prior to 2010-2012, their levels of performance are hardly comparable. Even if we give Arroyo the benefit of the doubt and pick things up in Cincinnati, it’s still not close:
Greinke: ‘06-’09: -0.09 FIPratio; 560 IP, 536 K/146 BB/45 HR; 3.31 runs average
Arroyo ‘06-’09: 1.22 FIPatio; 871.7 IP, 630 K/260 BB/119 HR; 4.38 runs average
Lincecum: ‘07-’09: -0.45 FIPratio; 598.7 IP, 676 K/217 BB/33 HR; 3.17 runs average
Arroyo was then what he is now — a low strike out guy that gets by because he typically limits mistakes and he pitches boatloads of innings. This is the replacement value riddle for general managers; Arroyo will stick around because even when he slips up (like 2011), he’ll pitch enough innings to keep that General Manager from making two or three acquisitions to replace him. See, that’s how pitching replacements work; it’s hardly ever one-for-one; over the last few seasons in the National League, you’ll usually find several replacement transactions for each spot vacated by a regular starter.
If we take Fielding Independent Pitching stats at their value, we’d conclude that in any given season, a pitcher might out-perform or under-perform their K/BB/HR ratios. However, we’d note that over time, we would expect pitchers with stronger K/BB/HR ratios to establish better performances than poor FIP pitchers. This is what fascinates me about the Giants/Reds pitching switcheroo, as well as the top free agent pitcher for 2013: despite posting Fielding Independent Ratios not nearly as strong as Lincecum or Greinke, despite relying significantly on batted balls in play, even despite worse K/BB/HR ratios in 2010-2012 than 2006-2009, Arroyo stayed the course and kept his 4.38 runs average for the Reds over the last three seasons. Both Lincecum and Greinke posted much worse runs averages (and in Lincecum’s case, a much worse FIP) during 2010-2012 than their 2006-2009 campaigns suggested.
For all their strike outs, for their strong K/BB/HR ratios, even for their stuff, even with Arroyo’s awful 2011, even with Arroyo removed from his prime years, Arroyo remains within shouting distance of both Lincecum and Greinke over the last three seasons:
Greinke ‘10-12 ($34.25M): 0.02 FIPratio; 604 IP, 582 K/154 BB/55 HR; 4.17 runs average (4.37 park; 13 R better than park)
Arroyo ‘10-’12 ($24.5M): 1.64 FIPratio; 616.7 IP, 358 K/139 BB/101 HR; 4.38 runs average (4.43 park; 4 R better than park)
Lincecum ‘10-’12 ($39.0M): 0.33 FIPratio; 615.3 IP, 641 K/252 BB/56 HR; 3.93 runs average (3.93 park)
Arroyo’s 4.38 runs average was approximately four runs better than Great American ballpark over the last three years, and Greinke’s 4.17 runs average ultimately places him 14 runs better than Arroyo — not insignificant. What’s mind-blowing about these performances is that if we look at these ptichers’ WAR, Greinke and Lincecum are light-years ahead of Arroyo; WAR says Greinke is 11-12 wins better than Arroyo, while Lincecum is approximately 7-8 wins better than Arroyo. Yet for all that, Arroyo is better against his park than Lincecum over the last three years, and he trails Greinke’s overall value by less than 20 runs.
The world of MLB pitching is much closer to these types of messy equations, I gather, than the clear, logical confines of Fielding Independent Pitching ratios. For all the magic of great stuff and high strike out numbers, a guy like Bronson Arroyo can come along and work his batted balls in play into a strong performance every now and then.
Even though Brewers GM Doug Melvin said that the Brewers were unlikely to pursue Greinke (or high priced free agents) according to the official Brewers site, fans will talk about the Brewers landing Greinke’s services to solidify the club’s rotation for the coming years. We all know Greinke has great stuff, we all know he’s young, we all know he’s going to strike out a bunch of guys and limit the walks and home runs. But, when we look at his overall performance, we should ask, “What will he accomplish in front of the Brewers’ defense? Or at Miller Park?” If you think that Greinke is worth Matt Cain money due to his age, FIP, and stuff, step outside of the box for a moment and look at a veteran like Arroyo. What would you pay for Bronson Arroyo’s 2010-2012 performance? Obviously, Arroyo’s stuff, FIP, and age yield a much lower contract than that of Greinke (and of course, Lincecum). But does that contract value match their respective performances?