Closer of the Year: Part 1

Francisco Rodriguez had a hell of a season, blowing by Bobby Thigpen's single season saves record and being chosen by The Sporting News as the top reliever in the American League. Ever since he came up as a 20 year old September call-up in 2002 and struck out 13 and walked only 2 in 5.7 spectacular innings, then became a post-season hero, he's been considered among the elite relievers in baseball. Six full seasons later, his success culminated with that big number 62 (no, not that #62) just in time for free agency. But was he the best in baseball?

Not even close. Sorry for killing the suspense so quickly, but chances are, if you track reliever stats beyond saves, there never was any suspense, unless maybe you were wondering is he really asking that? Well, I was, but only to set up the article. The saves record is certainly nice, but aside from that, K-Rod probably pitched the worst of his career this season.

Among some, that opinion is unpopular. Having more saves in a season than anyone ever has can't be a fluke, some have argued (I think it was John Kruk who used pretty much these exact words, apparently unaware that his former teammate, Bobby Thigpen, of the 57 saves at age 26 and 53 career saves after age 26 variety, held that distincion for 18 years). It's hard to see anything but pure chance in K-Rod's 69 save opportunities last year, though, which was more than all but 4 other entire teams had. Jose Valverde, with 51 SvO, was the only pitcher in baseball within 20 of that total. No one since 2004 had even had 55 in a season. K-Rod himself had never had more than 51. The only other pitcher in ML history with a season in the 60s is Thigpen. In fact, the last 4 pitchers to hold the season saves record (Dan Quisenberry, Dave Righetti, Thigpen, and K-Rod) also set the record for most save opportunities in a season in the year they broke the saves record. The last pitcher to break the record without getting more save opportunities than anyone before him ever had was John Hiller, who saved 38 games in 42 opportunities in 1973. Sounds like a pretty fluky thing, truth be told. Just like it wouldn't make much sense to conclude that K-Rod sucked because J.J. Putz was the only closer in the A.L. who blew more saves in 2008, judging his season by his save totals gets you nowhere.

Instead, we see that K-Rod had his worst K/9 and FIP since his rookie season and his worst WHIP, K:BB ratio, xFIP (FIP with the HR/FB normalized to league average), BA against, OBP against, OPS against, strikeout total, and hits allowed of his career. His K/9 peaked in his second season at 13.18 and then settled between 12.03 and 12.16 over the next three seasons. Then, in 2008, it dropped nearly 2 full strikeouts to 10.14. Meanwhile, his control has never gotten any better. His BB/9 of 4.48 in 2008 and 4.54 in 2007 are the two worst figures he's posted in his career. His velocity has also been following an alarming trend the past few years.

K-Rod Velocity by Pitch, 2006-08


FB
SL
CH
2006
94.8
84.2
84.9
2007
93.6
80.4
83.2
2008
91.9
79.6
84.4

Not only was his velocity way down last year, his fastball-to-change-up differential dropped from about 10 mph to 7.5. Velocity coming off the fastball and not the change is a bad thing. K-Rod's agent has suggested that the loss of velocity was intentional, and that K-Rod could still throw as hard but chose not to to improve his control. The problem with that story is that K-Rod's control still sucks. He wasn't throwing any more strikes this year than ever before, and he was walking more batters than he has over his career. He wasn't fooling as many hitters as he had been in previous years, as indicated by a drop in swings outside the zone from about 30% to 25%. So if his agent is right and K-Rod is taking velocity off on purpose, he should probably tell his client to knock it off, because he's not throwing more strikes, he's just making his balls easier to identify and lay off of.

There's another significant warning sign in K-Rod's season. For the first time in his career, he averaged less than an inning per appearance. In fact, he didn't pitch more than 1 inning all season until he went 1.1 innings against the Sox in the ALDS. Compare that to Papelbon, who had 11 saves in which he went more than 1 inning last year, or Mo Rivera, who had 9 such saves. Before this year, more than 1 out of every 8 saves K-Rod converted were multi-inning saves. More than 1 in 4 of his appearances had lasted longer than an inning. Even in 2007, he went past 1 inning 8 times in 64 games. This may have been a conscious effort on Scioscia's part to simply save K-Rod for as many short save opps as possible, or to keep him as fresh as possible for the playoffs, but considering:

a)it was the same manager who's always been managing K-Rod, and;
b)K-Rod didn't pitch more than 1 inning even early in the season (which he had doen every year until 2008) when his arm was fresh and Oakland was still threatening for the division,

paired with his dropping velocity, that's not really a good sign for K-Rod, and the effects were definitely showing in his 2008 numbers.

The following graph shows how K-Rod's career has progressed. The figure I am using for runs per game is similar to xFIP, but it normalizes both K-Rod's BABIP and HR/FB to his career averages and uses Voros McCracken's 4 hitting rates from his work on defense-independent pitching stats and Jim Albert's formula for converting them to runs per game. This is to minimize the effects of random variation and luck from his career norms in each single season. The points on the graph are expressed as the ratio of each season's R/G to his career mark, so the horizontal line labeled "1" is his career average. Points above that line are worse than his career average, points below are better.


As you can see, he had his big breakout in his sophomore year. Predictably, he regressed from his big 2004 the following year, and he showed steady declines over the next few seasons. Then, in 2008, he plummeted to his career worst. So it seems pretty unlikely that with K-Rod getting worse across the board, he would have all of a sudden become the best reliever in baseball. But who knows. Let's check it out.

K-Rod was in the bottom half of Major League closers in K:BB ratio in 2008. Eighteen teams had closers with better WHIPs (and one of those teams, the Dodgers, had 2 closers with better WHIPs thanks to Saito's injury). He wasn't even elite in strikeouts in 2008: he ranked 8th among regular closers in K/9. He was really only among the best in the league in saves. Looking deeper at his saves, we can see his total is more deceptive even than his ridiculously high save opportunity total.

Bill James classifies saves into 3 categories: easy, regular, and tough. Easy saves ask the pitcher to get 3 outs or fewer without the tying or winning run at the plate when he enters the game. Tough saves occur when the pitcher enters with the tying or winning run on base. Regular saves are all other saves. Of K-Rod's 62 saves, 39 fell into the Easy Save classification. That was by far the most in baseball. It works out to 63% of his saves. Compare that to the percentage of saves that were classified as easy for other top closers: Lidge 61% (25/41), Soria 60% (25/42), Nathan 54% (21/39), Papelbon 54% (22/41), Rivera 49% (19/39). Lidge and Soria were both in the 60s, but each took advantage of the high percentage of easy saves and converted a higher save percentage than did K-Rod. Only one of K-Rod's saves was a "tough" save. Rivera and Papelbon paced the Majors in tough saves with 5 and 4 respectively. K-Rod's lone tough save, in case you were wondering, came on June 29 against the Dodgers. He came in with 2 outs in the 9th and runners on 1st and second with a 1-run lead, threw a wild pitch, walked his first batter, and then got a groundout to second to end the game and record the save.

K-Rod only recorded 8 saves when he entered with men on base. In those 8 games, he allowed 5 of the 16 runners to score, including one save where he entered with the bases loaded and 1 out in a 5-run game, walked in the first run, allowed the second to score on an RBI groundout, allowed the third to score on a base hit, and then got another groundout for the save. He had a 9th save opportunity in which he entered with men on base; he allowed both runners to score in that outing, bringing is total IS/IR in save opps to 7/18 (39%). That's well below average and puts him in the bottom third of relievers in IS/IR. Three more of K-Rod's blown saves were classified as easy save opportunities, bringing his total easy save opps to 42, more than the total save opps for all but 5 other pitchers in the Majors.

There really was nothing special about K-Rod's season. Even with all the saves (and the accompanying bias in LI, which he led baseball in by a wide margin), he still finished outside the top 5 relievers in WPA. His WPA/LI was less than 1.00 (4 relievers posted WPA/LIs over 2.00). He was simply a benefactor of a perfect storm of circumstances. Lots of save chances, with a disproportionately high number of them being relatively easy, masked the major declines K-Rod showed last year and the alarming trends in his numbers. He really was nowhere close to the best closer in baseball last year, especially with other candidates standing out so strongly. But we'll leave those for future installments. For now, just know that when K-Rod joined the exclusive company of Thigpen, Righetti, Quisenberry, and Hiller, he fit right in.

3 comments:

Schruender said...

I thought when you asked was he the best that you were going to say somebody else. Considering he set the all-time saves record it is impossible to disagree. If he simply led the league in saves it would have been more debatable.

Along the lines of how much concern to have with K-Rod I think there are a lot of things going against him. I don't think that the length of his appearances last year is one of them. The Angels had a really deep bullpen and while you cite Papelbon needing to pitch more than one inning 11 times also note how often the Red Sox could rely on someone else to get outs. Okajima was very shaky. Also Papelbon pitched on back to back days much less.

Overall though this was a great first time read of your stuff - keep up the good work!

Kincaid said...

Thanks for the compliment.

The reason I'm concerned with K-Rod not going more than an inning all season isn't just how he compares to other top closers, but that it is really unusual for a closer/bullpen ace to never go out for a second inning. It was a bit lazy of me to just cite Papelbon and Rivera there, because you're right, it's not quite apples to apples, but I was mostly trying to establish some sort of baseline to compare to. No one, not even a left-handed specialist, has ever thrown that many games without going more than an inning in at least 1 of them. Most closers who never go out for a second inning are older pitchers toward the end of their careers (i.e. Borowski at 36, Hoffman in his late 30s). K-Rod is only the second pitcher in his 20's, closer or otherwise, to ever throw at least 30 innings in a season without going more than 1 inning at least once (Mike MacDougal is the other, in 2003). Deep bullpen or not, there are times over a 162 game season that you will want to send your best pitcher back out there for a 4th out if you can. The Angels bullpen was largely the same as it was last year, and K-Rod was going more than an inning then. It's not like he was only sent out for a 4th out a couple of times. They never did it, which sticks out both in K-Rod's career and in historical usage of closers, especially young closers.

By itself, I wouldn't take much notice. As you point out, there are reasons for him to pitch more than an inning fewer times than normal, particularly the number of times he went back-to-back days because of the sheer number of save opps that presented themselves. But with the pattern of not going more than an inning presenting itself even early in the year before those other factors were established, and considered with all the other signs (sharp drops in velocity and strikeouts, worsened control, fooling fewer hitter, drops in numbers across the board), it definitely raises some concerns for me. Maybe I'm reading too much into his usage pattern, but I do think enough other signs are there that things that wouldn't normally raise alarms on their own start to jump out.

hostile postulate said...

and, for the record, although he didn't name the best closer yet (part two teaser), he did say that it was definitely not k-rod last year.

Post a Comment