Spring Training Results
I put up a version of this post every year. For those who have been reading for a while, I apologize for the redundancy, but it unfortunately remains necessary.
As you are no doubt aware, Cactus League games began over the weekend. The good news is that means that we can stop hearing about who looked good jogging and stretching. The bad news is that now we have to hear who looked good because they put together a nice week or two against minor league pitching in a minor league ballpark in games that don’t count.
The usefulness of spring training statistics have been examined a couple hundred ways, and the result is always the same – they hold no information of value. Whether a guy has a good spring (Munenori Kawasaki hit .455 in Arizona last year) or a bad spring (John Jaso didn’t get a hit until the final week in March), the data has no predictive value. It is completely worthless, for all the reasons Jeff laid out a few weeks ago.
Every year, though, decisions are made based on how players do in March. The decisions are justified by claiming that it they aren’t based on the results, but on how the players look to experienced coaches and scouts who are paid to evaluate players in an up-close-and-personal atmosphere. The problem is that human beings — even experienced scouts and coaches — are pretty terrible at evaluating the difference between “how a guy looks” and what his results are.
In other words, it’s really hard to look bad if you’re hitting a bunch of home runs. It’s really hard to look good if every pitch you throw ends up as a rocket off the bat. Our opinion of how a player looks is informed by the outcome of the plays he is directly involved in. Yes, even trained MLB coaches. Despite the appeal to authority that people like to make, they are simply not immune to the biases that are inherent in our human reaction to watching people perform.
Last year, Eric Wedge watched John Jaso do poorly at absolutely everything, and his evaluation was that Jaso couldn’t help the team in any real capacity, so he stuck him at the end of the bench and never used him. He watched Hisashi Iwakuma give up a bunch of hits and decided that he wasn’t ready to pitch in the Majors, so he made him the backup long reliever and never let him pitch either. Meanwhile, Blake Beavan and Hector Noesi locked up their rotation jobs with strong springs. Alex Liddi played himself onto the team by showing a significantly improved approach at the plate, edging out Carlos Peguero for the final roster spot, despite the fact that Peguero was also deemed to be quite impressive in March.
When the games started counting, it became clear that Noesi and Beavan didn’t belong in the rotation, Jaso was the team’s best catcher, Iwakuma was being completely wasted in the bullpen, and Liddi still had no business in the Major Leagues. Of course, neither did Peguero. Pretty much every decision about playing time that was influenced by what the coaches saw last March turned out to be incorrect. And those decisions haunted the Mariners all year long. It wasn’t until the second half of the year that Jaso and Iwakuma finally got the roles they deserved, and half a season of good performances weren’t enough to change the coaches predetermined minds about what kind of skills Jaso possessed.
It wasn’t just the fringe guys either. The biggest story coming out of spring training last year was Ichiro’s rebirth as a #3 hitter, as he hit .415/.479/.634 in Cacus League play. A close second was how strong Dustin Ackley looked, as 10 of his 13 hits went for extra bases and he only struck out five times in 45 at-bats. You know what happened to those two once April rolled around.
Early on, it became clear that Saunders was hitting for power to to the opposite field, which he had never really been able to do before. I wrote about that last March 15th, for instance, as we followed with some amazement as Saunders drove double after double to left and center in Arizona. While he was still a pull-heavy hitter during the regular season, he took a big step forward in his results on hitting to center field, which was one of the primary reasons he had a breakthrough 2012 season.
That said, Saunders didn’t exactly have a monstrous Spring Training from a results perspective. He hit .356/.396/.533, which is pretty good until you remember that everyone hits well in Arizona. Of the 10 guys who got at least 40 at-bats in spring training last year, Saunders’ .929 OPS ranked seventh, just barely ahead of Jesus Montero (.923). The only guys who got regular playing time and didn’t hit as well as Saunders last year were Chone Figgins and Casper Wells. If we expand the list to guys with 30 or more at-bats, Saunders also falls behind Justin Smoak.
So, yeah, Saunders showed something last spring that was worth paying attention to, but it wasn’t a results thing – it was a change in mechanics and a display of a skill he did not previously have. Likewise, Erasmo Ramirez showed an average fastball speed of 94 MPH in games at Peoria (where PITCHF/x cameras are installed), which is well above what he’d shown previously in the minors. Like with Saunders, the results weren’t overly spectacular — one walk, four strikeouts in 10 innings — but the stuff was significantly better than had been seen in the minors, and Ramirez’s uptick in velocity allowed him to move from being a suspect to a real prospect.
Saunders and Ramirez exemplified the kinds of changes that actually matter in March. Big changes in velocity can matter, though as Felix showed, velocity loss can also not really matter, so don’t read too much into guys working to get their fastballs up to normal speed over the next few weeks. I’d say a velocity spike — recorded by a PITCHF/x camera, not a radar gun, and adjusted for the readings other pitchers were getting that day — is more important than velocity loss in March. For hitters, we don’t have such an easily recordable skill measurement, so there’s going to be a lot more of the BS fluff stories about so-and-so changing his swing. With Saunders, it was real, but it also resulted in a pretty obvious change in the direction and trajectory of the ball coming off his bat.
The rest of it, though, was total garbage. And pretty much 90% of what you’re going to read and hear over the next month is going to be total garbage. Jason Bay is going to “look good” when he hits home runs, and he’s going to “look old” when he strikes out. Unfortunately, the organization and the coaching staff have shown that they’re going to make decisions based on how guys look in Arizona, and so as long as Jason Bay stays healthy and hits a few more home runs, he’s probably going to make the team, while Casper Wells will be shipped off to someone who has a better grasp of Wells’ skills. This is an unfortunately predictable outcome, and I’m preparing myself for the inevitable dump of Wells at the end of camp, while we read about how Bay’s rejuvenation just pushed him off the squad. It’s going to be annoying, and it’s going to happen because the Mariners put a value on how players “look” in spring training.
That doesn’t mean you have to. Ignore the BS that filters down over the next five weeks. Whatever you think about the players today, you should think that about them on April 1st. Spring training performances simply don’t matter. We’d all be better off if they just had the entire exhibition season in private. What matters is what happens when the games count. None of these games count, and none of what happens actually matters.
If there’s a Saunders or Ramirez situation that suggests that further evaluation is required, we’ll talk about it, as we did with those two last spring. Those are the exceptions that prove the rule, however. By and large, you can basically ignore everything that happens between now and Opening Day and you’ll be no worse off for having skipped all of it.