A Tradition Unlike Any Other
It’s spring, and the Mariners top pitcher has made a single start. He faced the Rangers, in Arlington, and was extremely difficult to square up. On the other hand, he had awful command, walking 5 and plunking another, and velocity was in the :SIRENS BLARE: FELIX VELOCITY WARNING! IMMINENT INJURY???
This is now a rite of spring. We do this – baseball bloggers, I mean – every single year, and those of us in the Northwest are among the biggest offenders. I get it, I really do: he’s a star player with a large amount of money committed to him, and he looks nothing like the pitcher he was when he entered the league, or, for that matter, in 2009 or whatever. The proliferation of pitch fx data has been revolutionary for baseball analysis, and I will admit that 90% of what I do utilizes or is based upon that data set. It enables reliable trend data on all manner of pitching results, including velocity, and *we cannot look away*. In this specific case, I think we should. To say that Felix’s velocity is down is to state a truism, and something you could say after literally any start in the past six years. To state that it’s a sign of injury or impending ineffectiveness is, to put it mildly, not supported by the results of the past six years. Is THIS the time it suddenly matters? Is this the end of Good Felix? Pitch FX or release points or any of that *cannot give us an answer*. Only batters can do that.
Is this just me being a homer? Look, I’m the pessimist around here. This is my beat; I should be all over a great doom and gloom story. But we’ve all been through this more times than I can count. Here’s old friend Graham MacAree writing about Felix’s slowing fastball in the offseason following the 2009 season. He noted that as it slowed, it produced better results by pitch-type linear weights. Here’s Jeff in the spring of 2012, a bit after Dave wrote a post on Felix’s velocity after his first start of that year. Remember 2012? Good year for Felix, that one. Adam Wong noted Felix’s velocity was down in 2013, but that it was climbing again after a slow start.
As the years go by, the nexus between velocity loss and potential injury get more and more explicit, as with this piece last year, or this one from a few days ago. I don’t dispute the premise of these articles, which is that velocity decline can be a sign of injury, and that velocity decline is generally pretty highly correlated with performance declines. As Felix ages and his margin of error decreases, you might argue, it’s much more important. Going form 97 to 95 is one thing, but going from 92 to 90 may be very different. I don’t dispute any of these arguments, well-supported as they are. What I’d like to see is a bit more understanding that the same exact articles have been written now for years, and that in Felix’s specific case, it is very, very difficult to see any kind of connection between lower velocities and injury and/or ineffectiveness. Felix is *constantly* changing and adjusting, which can set off alarm bells too, as other measures start to look weird, from pitch usage to release point. Picking injury risk out of data works a lot better when you’ve got an extremely consistent pitcher who then varies from the established pattern. It’s much, much harder to do that with Felix, and other guys like him – Zack Greinke, for one. I tried to make this point last year, and I’ll schedule a post about it next year, too.
Not every “Is Felix Hurt?” and “Felix is throwing 89!!!” piece hits the blogs in the spring, but there’s a pretty big glut of them in April and May. Why? Because Felix’s velocity, and essentially every other pitcher in baseball’s velocity, is lower in April. Average April velocities are the lowest of any month, with May next-lowest. If you’d like to get a clear, dramatic velocity drop, just compare a pitcher’s velo in April against their yearly average the previous year. It’s not an apples-to-apples comparison, but it’ll make the gap look huge. To make matters worse, the pitch fx values *themselves* are a little “cold” in April, as Max Marchi wrote years ago at BaseballProspectus. Later in that piece, Marchi notes that the pitch FX systems differ by up to a full MPH or so, with the slowest guns about a 1/2 MPH below average, and the fastest three-quarters of an MPH above. What are the *slowest* radar guns in baseball? They’re in Arlington (#1) and Oakland (#2). In recent years, the M’s have opened against Oakland, Anaheim and Texas, in April (or even March). Look back to many of the articles linked above and notice how many were written following starts in either Oakland or Texas. What’s the point? I don’t think we can accurately, definitively say that Felix’s velocity IS down in any real sense from last year. We can say it’s down from 2010, but that tells us nothing. Did Felix throw 89mph in his first start? No, he threw 91, and touched 93. What was his average velo in April of last year? 92. In 2012? 92. In 2013? 92. Let’s see where Felix’s velocity is in May, shall we?
Why do we keep doing this? Why do we seemingly *want* to point out Felix’s decline, even when his performance doesn’t really decline? I think Patrick Dubuque summed it up best in this piece, which argues that as Felix rate of improvement slows or even halts entirely, we reach for all manner of explanations, partially out of concern for the king, and partly because we have an endless firehose of data pointing at us, and it’s the easiest thing in the world to grab some and turn our worries into evidenced-based arguments. For non-M’s fans, this is even easier. The general pattern is clear, and Felix sets off plenty of alarm bells, and the trend really does look ominous. But *every year* since that first “Felix is losing velocity” piece back after the 2009 season (and that’s just the first I could find last night; there must be earlier ones), Felix has put up more fWAR/bWAR than *every year* prior to it (with the exception of 2015). Felix has been a better pitcher as an average velo guy with a diabolical change than he was when he was a plus-plus velocity FB and plus-plus curve guy. We *know* this, but we’re just scared that it’s going to stop. It WILL stop, eventually, and that’s going to really sting, but the methods we use to suss out that decline don’t work. M’s fans are drawn to this because we’re protective of Felix, and I can’t think of a worse end to his M’s tenure than an undiagnosed, untreated problem that suddenly becomes a career-threatener. Velo seems like it’d be a good way to find something like that, but Felix can’t be reduced to his sinker velocity. Felix is getting older. Felix won’t be royal forever. I know that we can’t, and won’t, but let’s all *try* to relax on the annual velo freak-outs.