To evaluate the staff, I’ll be looking at a few statistics we might be aware of. ERA will only be mentioned once, but the team’s FIP and tRA (see here for a description of tRA; it’s very good) will be mentioned often. I’ll be using both tRA and FIP to compare the effectiveness of both, as tRA encompasses batted ball data that seems to be decently under control of a pitcher (i.e. pitchers do have some control over the types of balls in play they elicit, just not the location of such batted balls) while FIP heavily weights only peripheral data (complete defense-independent data such as strikeouts, walks, and home runs).
Team ERA: 4.38
Team FIP: 4.08
Starter FIP: 4.21
Starter tRA: 4.83
Starter tRA*: 4.72
Relief FIP: 3.85
Relief tRA: 4.27
Relief tRA*: 4.41
Those are a lot of numbers to digest of course. Each of them is weighted as a measure runs allowed per nine innings or 27 outs, so it gives us an idea of what the pitching staff is giving in each regulation game. tRA* denotes park adjusted tRA based on park factors for batted balls in play. Let’s first delve into some of these differences.
As FIP is a measure of balls not in play, it takes away the potential for any sort of non-HR hit. tRA uses batted ball data and gives linear weight values for each batted ball type, which helps to take into account some of the downsides of FIP, such as the consistent undervaluing of extreme groundball pitchers. By taking into account the run value of batted ball types, pitcher-controlled effects such as location, often blamed for hard hit balls and valued for “soft grounders” and other poorly hit balls in play, can be better determined. Not every pitcher is going to locate to get strikeouts, so while FIP is in general a solid defense-independent measure, it may not provide the entire picture of how a pitcher has performed.
That being said, both FIP and tRA determine that the starters have been outshined by the relief staff. This would come to surprise for Marlins fans, as the Marlins have seen excellent starting pitching from not one, but now two top-of-the-rotation guys in Josh Johnson and Ricky Nolasco, the team’s “pocket aces,” while the Marlins’ closer Matt Lindstrom sported a 6+ ERA before heading to the DL and the team has been scrambling for relief pitching for some time now. However, to underestimate the performance of the bullpen would be criminal. The Marlins bullpen has accumulated 16.7 runs above replacement via FIP and 37 runs above replacement via tRA. While the numbers definitely differ in value, showing that there is some difference in perhaps the translation of a replacement level pitcher between Fangraphs’ FIP and Statcorner’s tRA, both these values do show that the bullpen has been worth something in the ballpark of two wins above replacement so far this season. This is a testament to the performance of both Kiko Calero and Dan Meyer, the two most consistent men in the pen for the Fish. In both systems those two rank the highest in runs above average/replacement for the relief staff
It is important to note that while there has been clamoring for Leo Nunez to take the closer job over Lindstrom even after Lindstrom returns from injury, neither player has been much better than the other. Given the low amount of innings logged by either player, FIP has Nunez holding a 0.4 run advantage over Lindstrom with numbers that are far more likely to regress in the negative fashion than Lindstrom. If even at Lindstrom’s worst and at Nunez’s best, Nunez has barely been better than Lindstrom on the mound. The panic of seeing a closer struggle is gutwrenching for everyone, but when we check the numbers we can see that the setup man, everyone’s favorite reliever and soulbrother to the backup quarterback in football, has not been good either and is not much better. This isn’t to say that a replacement shouldn’t be searched for, but rather that given the current personnel, an internal change just isn’t necessary.
As for the starters, the numbers indicate more of a struggle than one expected. tRA suggests the Marlins staff has been a bit below average, while FIP has them at 10th in the league in that catergory and a bit above average. This is more indicative of the difference between FIP’s focus of strikeouts and walks and the more accurate description of pitching give tRA. The importance of batted ball data can be seen in evaluating starters like Ricky Nolasco and Andrew Miller, who have struggled more in tRA’s minds due to the increased line drive ratio in their batted ball profiles. I’ve been touting that Ricky has been unlucky for much of the season, but he has been hit harder this year than in previous years when he’s been hit.
The biggest struggle for the starters however has been the fifth starter slot. The Marlins top four starters (Josh Johnson, Ricky Nolasco, Chris Volstad, and Andrew Miller) have combined for a 4.20 tRA and a 3.75 FIP this season, far lower than the team’s corresponding season totals. As the Marlins move forward this year, they may have to consider skipping more fifth-starter days if the likes of Sean West and Anibal Sanchez cannot perform.
Best Performer: Josh Johnson
It is of no surprise that JJ took the nod as best performer. Johnson lost his 2007 season to Tommy John surgery and came back roaring in mid-2008, posting 1.7 wins above replacement in just 87 innings of work. This year, he picked up right where he left off, to the tune of a 2.99 FIP and 3.51 tRA, good for 3.2 and 2.9 WAR based on the two systems respectively. JJ has walked more men this season than in years past, but he’s made up for it with a strong strikeout rate and a miniscure home run rate, one that has a small amount of regression back up to his career norms. Of interest is JJ’s high ground ball rate this season, up over 50% to 52.5% this year. This trend can only Johnson continue his dominance, as traditionally it is somewhat safer in terms of run value to force ground balls over fly balls. If he can keep up this performance, he’ll be someone to carry this team for years to come.
Worst Performer: Matt Lindstrom
It’s easy to jump on Lindstrom as he struggled to adjust to the role of closer. What people don’t realize is that he actually has pitched about as well as years past. The last two seasons Lindstrom has struggled with his control and seen his K/BB ratio plummet. He will always have the rocket fastball, and his secondary stuff is still good, as shown by his high strikeout rate. His biggest issue this season has been mostly uncontrollable: BABIP. Lindstrom is forcing more groundballs than in years past, up toa 54% GB%. This has had some part in his massive .372 BABIP, much higher than his career totals in the past. This has led to a higher number of singles which, compounded with his higher walk totals, have caused runs to score in bunches when he has gotten into trouble. His strand rate of 76% indicates that he’s gotten into trouble more often than you’d like and gotten away with it, but the few large innings during which he’s struggled, such as the seven-run outing against Philadelphia and the blown save against Baltimore, have come with bunched singles and high run totals, mostly contributing to his sky-high ERA. All of that can be expected to regress, but with the volatility in small sample size for relievers, especially closers, it would be difficult to predict.
One final interesting note on Lindstrom. It would seem as if hitters have learned to lay off his pitches more often now. Perhaps hitters have learned that Lindstrom can walk himself into trouble and have decided to not attempt to hit his heater. Hitters are watching both pitches in and out of the zone at a higher rate than last year, 25% O-Swing% down from 27.6% last year, 63.9% Z-Swing% down from 72.9% last year. The largest difference is in the zone swinging rate, which has led to greater strikeouts and more called strikes (18.6% called strike % up from 13.6% last year). But because Lindstrom has been out of the zone more often this year, throwing only 48.6% of his pitches in the zone, he’s also increased his walk totals a great deal. If Lindstrom can begin finding the zone more effectively, hitters with more patient approaches are going to watch more fastballs in the zone go for K’s rather than walks.
Key Second-Half Improvement: Ricky Nolasco
Nolasco’s demotion in the middle of the first-half may have served as a wakeup call for Ricky, who aside from a weak start against Arizona has been absolutely filthy since his return. In reality, it might have been simple regression that has brought Ricky back to his 2008 dominant form. With Nolasco, you get what you see, a high strikeout pitcher with excellent control and a penchant for bad innings and long fly balls at times. His strikeouts are up this year to almost one per inning, while his walks have only taken a slight raise from 1.78 to 2.08 per nine innings between last year and this year respectively. His home run rates as evidenced by HR/FB% have stabilized around an expected value, meaning we can expect to see Ricky in his correct peripheral form.
What had been an issue for Ricky prior to his demotion was his batted ball luck. He has been hit harder a bit this year, to the tune of a 22.3% LD%, but such a change from his norm should not have caused a subsequent rise in BABIP, currently sitting at .360 and having been much higher before his return to form. After the call up, the balls started falling in a bit less, though they still came in bunches at times. Nolasco’s strand rate, however, is beginning to climb to a respectable level, now up to 58%, and should continue to rise as he plays his way through the so-called “slump.” If he continues to pitch this way, we’ll have two aces set up for what hopefully turns out to be a final table with the playoffs in sight in September. Hopefully the aces can carry us to a first-ever division title.