Friday, February 11, 2011

Speed Up the Middle

Things I'm Not Worried About, Part 3


It's been a low-key offseason for the Twins. That's fine with me - I didn't see any need to tamper too much with a team that won 94 games and finished in the top 5 in the league in scoring and pitching. There will be several new faces in the bullpen; that was inevitable given the FA market for relievers this year. The other major change comes in the middle infield. Those moves were elective.

Last offseason, the Twins traded for JJ Hardy and signed Orlando Hudson to form an elite DP combo. Both had been fairly recent representatives at the All-Star game, and each routinely posted better than average numbers at the plate and in the field. Adding them to the lineup turned what had been a weakness in 2009 into a strength.

After the season, Gardy publicly expressed a desire to improve the team's speed up the middle. There was never any indication that they would try to bring Hudson back - they felt it was time to give Alexi Casilla, a cheaper, faster option, another chance. But it was a surprise when the Twins won the bidding rights to Japanese League batting champ Tsuyoshi Nishioka. It appeared that his acquisition would send Casilla back to a bench role. But the Twins chose instead to deal Hardy to Baltimore for a pair of relief prospects.

That created quite an outcry from Twins fans. After all, Hardy, even in a season weakened by injuries, still managed to outhit the average MLB SS by a little bit, while rating very highly according to the defensive metrics. Nishioka has some good numbers in Japan, but what does that really mean when most NPB position players have been huge disappointments in the States? Casilla has had some flashes of quality play, but has played his way out of jobs in 2007 and 2009. Is he someone we can really count on in 2011? Doesn't entrusting the MI to these two amount to a serious weakening of the team, opening the door for the White Sox to take over the division?

To answer those questions, the last one in particular, I want to get realistic about what the Twins lost in Hudson and Hardy, and what they can expect to get from Nishioka and Casilla in 2011.

What We Lost - Offensively

Hudson has been strangely undervalued over the last couple of seasons, considering he's been playing above-average 2B for that whole time. There seem to be some concerns about his clubhouse presence and, more importantly, his on-field contributions over the course of the season. The Twins found out about that when they saw him hit .243/.311/.340 over the final 4 months of 2010. Here's his overall line was:

.268/.338/.372, 24 2B, 5 3B, 6 HR, 10/13 SB (126 G)

Compare that to the average MLB 2B (per 162 G):

.265/.330/.389, 31 2B, 3 3B, 14 HR, 12/17 SB

And that of all Twins' 2B in 2010:

.261/.329/.379, 32 2B, 7 3B, 10 HR, 12/17 SB

Pretty darn close to exactly average. Add in the fact that all three of his slash lines have diminished over the past couple of seasons, and that 2011 will be his age 33 season, and it's not hard to see why the Twins figured they could do just as well with someone else.

Hardy had a nice little bounce-back season from his horrid 2009 campaign. His BA and OBP were essentially back to the levels of his 2007 season, in which he made the All-Star team. However, he came up 20 HR short of that performance, so his overall line:

.268/.320/.394, 19 2B, 3 3B, 6 HR, 1/2 SB (101 G)

was, as I mentioned earlier, only slightly better than that of all MLB SS (per 162 G):

.262/.319/.374, 29 2B, 4 3B, 10 HR, 13/19 SB

The guys who filled in for Hardy during his frequent injury absences didn't slug as well, but matched his other slash numbers. All Twins SS in 2010 hit:

.268/.321/.371, 30 2B, 4 3B, 7 HR, 4/7 SB

That is also just about exactly average, though clearly lacking in the SB department. Hardy is young enough that he could still produce well-above average numbers in a healthy season. But he struggled so much at Target Field (a Punto-esque .252/.313/.340 with 1 HR) that the Twins had some reason to think that his hitting style wasn't a good fit for their new park. And since he had nothing to offer in terms of threatening speed on the bases, the Twins seemed to feel that everybody would be better off if he went elsewhere.

What We Lost - Defensively

Much of the value assessed to Hudson and Hardy comes from their high rating on defense. Hudson's +12 UZR/150 ranked 3rd among all MLB 2B, and Hardy's +12.8 UZR/150 was the best among players with at least 800 innings at the SS position. That sort of excellence will be nearly impossible to replace. Some caveats, though:

2010 wasn't just the best UZR rating for Hudson in a while - it was actually his first positive rating since 2005. However, according to Total Zone he was just about as good in 2006 and 2009 as he was in 2010: a little above average. And by the measure of Defensive Runs Saved, the only negative year of his career was 2008. DRS liked his 2007 just about as much as 2010.

Hardy's UZR has always been in the black, but 2010's rating was the best it's been for him since his first full season in 2007. Then again, TZ and DRS rated those 2 seasons as average and good, respectively. They were much more excited about Hardy's 2008 season, which UZR rated as merely good.

Depending on the metric of choice, a player's defensive value can really fluctuate. There's a lot of subjectivity involved in assessing which balls hit into a defender's area were playable by the "average" fielder. I wouldn't put any serious stock in any one of them - I think an average plus what you observed with your own eyes over the course of the season is a better method. By that measure, I'm not sure either one of those guys was truly elite on defense, but I'm comfortable in assessing each of them as solidly above average.

So despite losing Hudson and Hardy, the Twins' offense will not lose a beat between 2010-2011 so long as their replacements provide average production. Run prevention will take a step back unless the new guys prove to be plus defenders. Can Casilla and Nishioka deliver that?

What We're Getting - Offensively

Casilla has been maddeningly inconsistent in his short MLB career. He flopped when handed the 2B job after the Twins traded Luis Castillo at the deadline in 2007. He started 2008 in the minors, then took advantage of a depleted MI in May of that year and had a dandy season. Based on that, the 2B job was again his to lose in 2009, and lose it he did, with a wretched season both at the plate and in the field. Relegated to a bench role in 2010 due to being out of options, he put together a solid campaign in limited duty last season.

Add it all up, and you have a .249/.306/.327 hitter through nearly 1100 PA. Some might say that's enough to define his ability, but I'm not so sure. I don't like to include player's first tastes of the big leagues when trying to determine what they're capable of as veteran's in their prime. It tends to understate their abilities. (I also wouldn't want to use Jim Thome's prime years to determine what should be expected of him at age 40.) I think 3-year averages are a pretty fair measure of where a hitter is at this stage of his career. I like 5-year averages even better for guys who have been in the league 6+ years.

In Casilla's case, we'll have to make due with the smaller sample. Throwing out his rotten first 210 PA from 2006-2007 leaves him with a line of .256/.316/.344 over 863 PA. That's close enough to the average SS in BA and OBP, though plenty lacking in power. The lack of power is somewhat offset by fantastically efficient base stealing: he's 24/27 since 2008. Just because he doesn't hit a lot of 2Bs doesn't mean he can't get himself into scoring position by other means.

If that's what Casilla reduces down to, that's not such a bad thing from a #9 hitter. But we can't be sure he actually would produce numbers like that. He's never come anywhere close to them. He's either been a lot better or far worse:

2010 MLB SS: .262/.319/.374, 29 2B, 4 3B, 10 HR, 13/19 SB
Casilla 2007: .222/.256/.259, 5 2B, 1 3B, 0 HR, 11/12 SB (204 PA)
Casilla 2008: .281/.333/.374, 15 2B, 0 3B, 7 HR, 7/9 SB (437 PA)
Casilla 2009: .202/.280/.259, 7 2B, 3 3B, 0 HR, 11/11 SB (256 PA)
Casilla 2010: .276/.331/.395, 7 2B, 4 3B, 1 HR, 6/7 SB (170 PA)

He's yo-yoed between being one of the worst hitters in baseball to being slightly better than average. Truly feast or famine. He's never had a full season's worth of PA, so if that changes we might see more moderate results. For now, I see a young player who has proven in 2 of the last 3 seasons that he's capable of holding his own as a major league MI. Entering his age 26 season, I expect him to deliver that kind of performance again.

Nishioka is a tough guy to track down. I could only find reasonably complete numbers for his NPB career for 2007-2009. I also came across some very incomplete career numbers which happened to contain inaccurate BABIPs. However, by piecing the two together and doing a little forensic algebra (and you wondered when you'd need that in the real world!), I've come up with a pretty accurate picture of his progression as a hitter.

2003: 11 PA, 9 AB, 3 H, 2 BB, 5 TB Age 18
2004: .255/.304/.396, 212 AB, 54 H, 6 HR, 84 TB, 8/8 SB, Age 19
2005: .268/.320/.394, 447 AB, 120 H, 4 HR, 176 TB, 41/50 SB Age 20

He got a cup of coffee with the big club in 2003 just after he turned 19. He played about 1/3 of the season the following year, then more or less stuck for good in 2005 as a 20 year old. In those first 3 up-and-down seasons, in what amounted to just over one full-time season's worth of PA, Nishioka showed promise. His combined line for those 1st 700 or so PA was something like .265/.317/.397. That's good pop for a MI and some prolific base-stealing, but a mediocre BA and low BB rate. Still, for a very young player at what seems to be a more or less AAA level of play, he was holding his own. With his feet sufficiently wet, Nishioka took a step forward in his age 21 season:

2006: .282/.358/.390, 426 AB, 120 H, 4 HR, 166 TB, 33/50 SB Age 21
2007: .300/.366/.393, 559 PA, 148 H, 50 BB, 73 K, 3 HR, 194 TB, 27/40 SB Age 22

Here we see much improved BA and OBP, but a big drop-off in IsoP and SB efficiency. This, to me, represents a change in approach. Whether because a coach got through to him or he simply got acclimated to the league, Nishioka became much more selective at the plate, drawing more walks and getting better pitches to hit. He cut down on his swing, becoming more of a singles hitter when he did put the ball in play. This new, contact-oriented approach, combined with his good (though clearly diminished as he filled out) speed, established a new BABIP baseline well north of .300.

2008: .300/.357/.463, 522 PA, 142 H, 36 BB, 68 K, 13 HR, 219 TB, 18/29 SB Age 23

Another step forward. Nishioka added power to his game without increasing his K%. His BB% fell off a bit, but thanks to fewer SF and more HBP, his OBP only took a modest hit.

2009: .260/.361/.427, 531 PA, 118 H, 67 BB, 76 K, 14 HR, 194 TB, 26/36 SB Age 24

Here's a massive step forward in BB% while maintaining the IsoP and putting up the best SB% in years. Yet the season is a bit of a disappointment thanks to the lowest BA since his rookie year. His K% did tick up a bit, but the real culprit is a .284 BABIP, a huge drop after 3 straight seasons well over .300. Given a typical (for him) BABIP of about .330, his line would have been something like .297/.392/.465. His overall game improved once again - he was just unlucky that year.

2010: .346/.423/.482, 676 PA, 206 H, 80 BB, 71 K, 11 HR, 287 TB, 22/33 SB Age 25

The Wheel of Fortune spun the other way for Nishioka last year. Everything fell in, leaving him with an absurdly high .389 BABIP. His BB% and IsoP slipped back a bit from the career highs he established in 2009, but he was also able to reduce his strikeout rate back to 14.5%. Given a more ordinary .330 BABIP here, his line turns into .295/.379/.431 - revealing 2009 as actually the more impressive season peripherals-wise. Still, that's plenty good production from a MI table-setter.

Nick Nelson downplayed Nishioka's potential in this glass-half-empty post in December. It's certainly fair to mention Kazuo Matsui as a cautionary tale, and just because Nishioka got 200+ hits in a season doesn't mean he's going to be the next Ichiro. However, I found a lot of Nick's comments to be unreasonably negative. To say that, other than 2010, Nishioka's career was "unspectacular" is an insult to the people who sent him to 4 All-Star games and the National team prior to that season. He's clearly exceptionally good for his position and his league. Then Nick suggests that "no one would get excited" about the .287/.361/.427 line Nishioka put up from 2007-2009 if it came from a AAA player. Really? From a SS? In his age 22-24 seasons? Compare the numbers I listed above with what Trevor Plouffe did at the same age for the Red Wings:

.256/.292/.420 Age 22
.260/.313/.407 Age 23
.244/.300/.430 Age 24

If he'd put up Nishioka's numbers, with plus defense for his league, do you think he'd still be behind Matt Tolbert on the depth chart? I doubt he'd have spent 2010 in the minors, or that the Twins ever would have traded for JJ Hardy.

Nick dismisses Nishoka's 2010 as an "outlier" because of the unsustainable BABIP. Okay, but why should that good luck year be dismissed and not the bad luck from 2009? Wasn't that just as much a BABIP outlier from what he'd established the previous 3 seasons? Why disparage him as a .284 career hitter before 2010, when that figure includes not only the unfortunate 2009 but also 700 or so PA while feeling his way as a teenager at his nation's highest level? Why not throw out the ancient history from when he was a kid and just look at everything he's done over the last 4 seasons (the numbers I happen to feel extremely confident are correct)? Then his line is:

.300/.380/.443, 2291 PA, 15.1 K%, 9.9 BB%

Again, if your AAA MI is giving you that, unless you've already got Jeter and Cano, that guy is coming to the Show.

Nishioka has made adjustments throughout his NPB career. He's made himself into a premier player over there, and has earned his shot in the big leagues. It's pretty clear that Japanese power doesn't translate well, but that's not a huge part of his game. He's not an efficient base stealer, but I'm sure he'll get plenty of chances nonetheless. I expect him to be able to at least match what the Twins got from their 2Bs in 2010. In other words, I'll take the over on a .708 OPS and 12 SB.

What We're Getting - Defensively

There's no way to know this, of course. So far Casilla has ranked very poorly defensively across all the metrics. But that's at 2B, and he'll most likely be playing SS in 2011. You might say that someone who has a bad zone rating there won't do any better at the more demanding position. Fair enough, but consider this: there's more than one way to be a plus defender. You can have above-average range, enabling you to get to more balls than the typical guy and converting more chances into outs. Or you can have average range but be very sure-handed, committing fewer than usual errors and, therefore, converting more chances into outs. Or you can do both of those things, as Hudson and Hardy did last year.

Breaking down Casilla's UZR this way, we see that the bulk of his negative assessment as a fielder comes from the error side of things. We've all seen that from him from time to time - a lapse in concentration or sloppy fundamentals. (Interestingly, over 90% of his negative value as a 2B came from 2007 and 2009, when he was also hitting like crap. He took his ABs with him into the field, I guess.) He's only had a negative range component in one year, though. Despite his poor ratings as a 2B, there doesn't appear to be a systemic reason to expect below-average range from him at SS.

He's logged only 233.1 big league innings at that position so far, so the sample size is small enough to really distort attempted projections. But in that brief time, Casilla has been rated somewhere between average and very good at SS. He has the tools, it's his natural position, he's maturing as a player. We should expect him to play SS at least as well as anybody, and he's capable of playing it better than most.

As for Nishioka, the track record for Gold Glove Japanese MI isn't great. However, using Matsui as a comp once again, we see that while he struggled mightily as a SS, he actually rated out well as a 2B, slightly above average for his career. He was 29 when he started playing that position regularly - 3 years older than Nishioka will be on opening day. There may be an adjustment to grass & dirt infields and more aggressive baserunners. But I won't look for Nishioka to do any worse at the keystone than Matsui did. I expect him to rate as an above-average defender at 2B.

Conclusion

Casilla will probably never hit as many HR as he did in 2008 again, especially not at Target Field. There will be some loss of power from the SS spot. But his superior foot speed should enable him to stretch an extra base out of some balls where Hardy would have had to hold up. As he showed in his limited PA in 2010, you don't have to put a bunch of balls over the fence to slug about .400. Casilla should definitely have a higher OBP, way more steals, and score more runs than Hardy did.

Nishioka will have a tough adjustment to make coming to the US. Not just to the quality of play, but to the English submersion, and to being an ocean away from his home. However, he's shown himself to be capable of making adjustments. If his 2009-2010 NPB numbers represent his ceiling, they should translate into at least the sort of numbers Hudson put up last year.

Defensively, we're almost certainly going to see a drop-off. Not because Casilla and Nishioka aren't good defenders, but because Hudson and Hardy each had such outstanding seasons in the field last year. But defense can still be a plus for the Twins, and Casilla and Nishioka may be able to improve on the one area in which Hudson and Hardy were lacking last season: turning DPs.

I expect the Twins' new MI combo to provide at least the same production at the plate as their predecessors, while vastly improving upon their performance on the bases. As a result, they will not be a detriment to the offense. Defensively, while perhaps not exceptional, they should also be an asset. The loss of the Hudson and Hardy shouldn't result in a net loss in the overall run differential when all is said and done.

As long as Casilla and Nishioka don't get hurt, that is...