Who Will Close in Minnesota?
We’ve already gone over the trade of Matt Capps from a real-life angle, but most fantasy managers really only want to know one thing: who’s closing in Minnesota now? Incumbent interim closer Jon Rauch has not been great, but he hasn’t been terrible either.
The kicker about this one is how similar these guys are. Just look at their FIPs right now – Capps has a 3.46 FIP, Rauch a 3.42 – and you realize that maybe the Twins didn’t need a second dude just like the first in their bullpen. The comparison runs even deeper than that, as Capps (7.43 K/9) and Rauch (6.34 K/9) are not the traditional closers, but instead are two relievers that rely on control (Capps has a 1.76 BB/9, Rauch a 2.11 BB/9). But already we have a clue here that Capps is teensy bit better, because he is slightly better in both categories.
The nail in the sabremetric coffin, for Rauch, is groundball percentage. He’s a flyball guy, with 33.4% career flyballs and 37.3% this year. Capps, at least this year, is a comparative grass-guzzler, with a 48.3% groundball rate this year. On the other hand, we should be careful not to put too much stock into this number – his career number is much more modest (37.8%), and Capps is now (199 BF) not far past the total batters faced benchmark (150) for groundballs.
Ah, but Capps has 26 saves! Whoo!
The fact that the team traded for him, and his numbers are just slightly better across the board, probably means that Rauch is out of a job. But since the two are so close, Rauch owners should probably hold on to their now-devalued asset for just a little bit to see how things shake out. (That is, unless they can pick up Drew Storen in Washington, who is the favorite there if just because of his recent draft status and age, but that’s enough for another post.)
I’ll leave you with some Twins-centric anger from Aaron Gleeman, who excoriated the trade today on his website, because hey, who better to rant about a move than a die-hard fan of the main team involved:
In reality they traded Ramos for a setup-caliber reliever who accumulated saves on bad teams and is thus overrated and soon overpaid. Among the 93 pitchers who’ve logged 150-plus relief innings in the past three calendar years, Capps ranks 38th in xFIP, 49th in FIP, 50th in ERA, 61st in strikeout rate, and 85th in opponents’ average.
You’d think the Twins would have learned something about the created-not-born nature of the closer role and often spurious value of saves from Rauch’s relatively successful stint filling in for Nathan, but instead they just paid a premium for a guy whose perceived value and ability are much higher than his actual value and ability solely because of his role and save total. Capps is a good reliever, but the Twins paid for a great reliever and did so for all the wrong reasons.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
I’d have to say Capps since Gardy already said he was the closer.
Eno –
(this is really long, but I’d like to hear your thoughts; also, I, too, have enough in my head about the Storen situation for another post, so I’ll save you from that)
“Ah, but Capps has 26 saves! Whoo!”
With all the sophisticated data and depth of analytical knowledge proffered by this site, I find it rather disturbing that the Capps analysis presented here closed itself down from the outset by getting itself trapped into the conventional-wisdom based commentary that Fangraphs has so effectively combated. The quote above exemplifies this, since it falls back on a saberist’s own sort of CW in assuming that team management makes their decisions based only on such superficial statistics as the number of saves one reliever has over another when selecting who will close. The problem here is that there are clear reasons for taking Capps based on the very statistics Fangraphs provides us, if one were only to take care not to get distracted by our own assumptions.
To properly express just how much more capable Capps is over Rauch, I am going to put their respective numbers as a percentile of all 145 qualified relievers. So, for example, taking the article’s assertion that Capps’ 7.43 K/9 vs. Rauch’s 6.34 as the first bit of evidence that they are similarly atypical closer types does have a point (both have a below average rate relative to their reliever brethren), but that extra K/9 that Capps has makes him better than 40% of all RPs (or nearly average) while Rauch is in the 20th% (solidly in the “bad” category) … a significant difference made all the more so by the addition of that small but existent BB/9 rate, as Capps’ K/BB (4.33) places him above a full 90% of all relievers, while Rauch’s 3.00 barely puts him in the top-third. In other words, according to this last, and very important, metric Capps is excellent while Rauch is merely good.
That isn’t just a “teensy bit better.” Nor, by the way, are their xFIPs: 3.53 (75th percentile) vs. 4.48 (30th), and that discrepancy in xFIP also takes away the surety of the argument that they are equal because their FIPs are alike.
The “Plate Discipline” stats are also quite striking, and quite relevant to the article’s focus on the two being control pitchers, as such numbers can suggest whether one has the tools to survive as this type more than the other. Rauch’s O-Swing% is exactly average, while Capps’ is not only great … it’s better than 98% of all others. Their Z-Contact% exactly mirrors their K/9 discussed earlier: Capps is close to average, while Rauch is plain bad at avoiding wood when he throws strikes. And even when Rauch is excellent in something, Capps outdoes him: the former is in the 90th% with his F-Strike%, but Capps is better than all but three other relievers! Finally, Capps is average (and not just “close to” as twice before) in SwStr%, while Rauch is yet again in the 20th%. So, Capps is a pretty typical RP when it comes to getting hit in the zone and throwing swinging strikes, but is exceptional in his ability to get batters to chase and helps himself do so by throwing more strikes early than just about anyone else; Rauch is nearly as good with that first pitch aggression, just average getting hitters to chase, and really can’t get hitters to either swing and miss or stop hitting his strikes. I’d say that makes them different.
ISSUE 2: This is actually short and sweet: The benchmark you cite is nothing but the point at which we should accept something as statistically reliable, plain and simple. So why say that Capps’ GB% isn’t reliable because he just barely has enough PAs for his new rate to stabilize according to that benchmark? If it stabilizes at a certain number, then yes, actually, we should by definition put stock in it, despite the fact that his career average is different. He met the requirement to have an R of .70, end of story, right?
Finally, about your Aaron Gleeman citation: I agree that it is better for a fan to be the one to rant, but I also feel that one should do so where emotion far eclipsing logic has a place … as in the blog he wrote it in, but not here. I’m not saying I’m against emotion (love Carson), just when it gets in the way of what is also being presented as reason.
Of course, I just might not get what he’s saying, in which case maybe the following perspective could be corrected in a reply:
Why does Gleeman think it’s relevant that the saves he got were for a bad team in terms of whether he is overrated and soon to be overpaid or not? Really, what the hell does getting saves on a bad as opposed to a good team have anything to do with anything? And as for his monetary value, he is only costing the Twins $800,000 for the remaining third of the season and, if we assume Gleeman is right that he’ll get more than 5 million in his final-arbitration 2011 year season, so lets say 6 million, you have 6.8 m. for 1 1/3 yrs., or an average of 5 million/yr. between this year and next. If you average 2007, ’08, and an assumed 4 mil. ’10 – and take out an ’09 in which a .370 BABIP and a HR/FB a full 50% higher than his career avg. makes it meaningless as a predictive factor – you get, yup, 5 million. And if I’m off with my assumptions, well, the point is that it won’t be sky-is-falling overpay (which reading the whole article he wrote makes it seem to me).
To be clear, the only reason this shocked me into delving into all of this was because I have become so accustomed to the insight you guys bring every day (so I’m only talking about this one piece), and also maybe because I, myself, got carried away in the joy of doing the kind of analysis this site has taught me how to do and just kept finding more and more in this minor debate about Capps. This is not some attack, I really hope you don’t take it that way…
This is an interesting post, but there’s a fundamental piece here that I think might make this come into focus. I don’t think Aaron and I were talking about overpaying in terms of money – we’re talking about overpaying in terms of prospects.
You’re right that we can probably consider his groundball percentage reliable, but it is a recent occurrence without a significant change in his pitching mix, so it’s hard to say that he is really a groundball wizard suddenly. I was using that benchmark really more to illustrate that even one year of a reliever’s work isn’t a huge sample.
So we are left with two relievers, one of which is slightly better than the other. They’ll pitch 25 innings each probably. So Capps will strike out three or four more batters, give up a home or two less. Is that worth real prospects? What are those couple of scoreless innings worth?
I think they overpaid for a marginal improvement. I apologize for the snark. It comes out from time to time, but I try to keep it to a minimum. But when you feel a team overpaid with prospects for a marginal improvement to the team, you are left guessing as to why they did so.
Or: I can’t speak for Aaron, but I would say the overpayment is more about the prospects, not the money.
Thanks much for the reply (I notice you’re more engaged with commenters than some, which is awesome). A lot of what you said made good sense. I didn’t give much thought to Ramos because A) I don’t really know much about how to value prospects, or trades for that matter!, and B) I was only responding to the part where he said they traded for a guy who is “soon overpaid”, which would just be about money, then. That said, you two are most likely in agreement, as his whole article focused much more on the type of value you meant. At any rate, I can now see the outrage, having finally read some comments about the trade, especially the fact that it is irrelevant whether the Twins will ever find a place for him or if he’s better or worse than expected, because they surely sold low by trading him now…
Anyway, all I wanted to do overall was just push back a little at the idea that it was obvious the two pitchers were nearly the same, as well as the notion that, as the quote said, that this was done solely on a reliance on the save statistic.
Oh, and I definitely hear you on your question about where this GB% came from this year… in fact, I looked into that but forgot to add that PitchFx shows Capps throwing a new 2 seamer almost a quarter of the time. I know about pitch classification changing over the years, but that pitch was on other pitcher’s ’09 charts so there’s good reason to believe that, well, that there is a good reason for this change.
Which would be nice, since I wasted far too much time unpacking plate discipline numbers and now maybe, just maybe, there could be something to that beyond pure randomness.
Well, anyway, thanks for the good response…
Excellent insight from both Eno and William, thanks guys. Eno, from your response, I’d say you’re a class act.
I didn’t think so at first, but as much as I hate to admit it (I own Rausch in my main fantasy league) Capps is a legit upgrade due to the facts that William stated. From the Twins perspective, I think they could have done better for Ramos but at the end of the day that’s neither here nor there as far as I’m concerned. I’m just PO’ed as I now have to look into closer help for the stretch run. I’m looking at you Mr. Hanrahan…
Thanks for giving me hope there was something there, 3Fingers … and to make you feel better, an admission: after three appearances, I traded Capps away because I was a genius who could trade someone who “already had two saves by April 10th” while secretly knowing it couldn’t last because he’d already thrown five walks in just 3 innings… um, yeah.