Three Reasons to Love the Karns-Miller Trade: Part 1
As mentioned entirely too long ago by my esteemed colleague Monsieur Collette I have been itching to break down this trade. With little to talk about the plan is to go really in depth here to take a look at all of the moving parts. I covered my thoughts on Boog Powell back in July and I have no reason to change my opinion. He has a chance to be a lot like Gerardo Parra, but guys with his skillset rarely become even bench guys in MLB. C.J. Riefenhauser is in a similar boat as a LOOGY with arm issues. Both of these pieces are a step above an A-ball lotto ticket as they have beat their way through the bush to at least get within sniffing distance of the Show, but they’re not much different from fringe prospects.
That leaves three players to discuss. In this first part I will take a look at some reservations I have about the recently jettisoned Nathan Karns before tackling the Rays-side of the haul in Brad Miller and Logan Morrison in the near future. Karns carries a ton of surplus value if you think he is something like a 1.5 – 3.0 WAR starting pitcher with five years of control. A TON. The issue I have is that I don’t think he’ll still be in the league for all of those years, which mitigates some of what the Rays are giving up. Here’s an optimistic view of his surplus value using my homemade methodology:
In an optimistic world where Big Nate stays healthy and productive he is an absolutely strong get as the headline of a trade. Roughly $64M in surplus value is equal to something like an 11 – 26 PLUS a 26 – 50 ranked bat on the Baseball America top-100 rankings. That’s not quite what the Red Sox just gave up to get an elite reliever paid like one, but it’s a lot. However, Mr. Karns presents several red flags that lead me to believe that expecting him to be both healthy and effective for that five-year term seems like the still life daydreams of an insomniac.
Josh Kalk was ahead of his time in using Pitch F/x data to determine injury indicators in a landmark piece that helped him get hired by our very own Rays. It’s a must read, but he found that several variables could be used to get an idea of pitchers that may see declining health in the near term. In order of power those variables were:
- Vertical Movement
- Horizontal Movement
- Vertical Release Point
- Horizontal Release Point
Using Pitch F/x data we can look into these components to gain a better idea of whether or not Karns displays any injury-related flags. I’ll start with velocity as most folks have no problem seeing a correlation between velocity fluctuations and potential injury. In order to gain some context I looked at all Rays pitchers that threw at least 1,000 pitches this year. I then looked at the variation between the pitch number on the season and their overall average for that category:
You can click all images to see a larger version. Here, and with movement, I’m looking strictly at two and four-seam fastballs. We can see a couple of spikes here that might be a little bit worrisome, especially as he was on his way to a third spike to close the year before being shut down for forearm tightness. The team mentioned that they weren’t concerned and that if it was a different point in the year that he could have probably kept going, but words are wind. He profiles pretty similarly to Chris Archer, but Archer was much more consistent over his final quarter of the year. Maybe not a lot here, but maybe there is something. Here I have isolated Karns:
The variation in his velocity corresponds with the secondary axis on the right and I am using a smaller, 100-pitch trend here so that blips will be magnified. You’ll also notice that I’m including his variation in pitch movement for both the horizontal and vertical paths. We can see that he had a vicious spike in his vertical variation around halfway through the year, though he was able to gain more consistency through the rest of his season. His horizontal movement variation shows more volatility with a couple of spikes, but again he closed the year on a lower note. Here’s some context using his teammates:
Starting with just the vertical-side we see his peak and ebb, and while his peak was one of the higher ones on the team, once initial early season kinks were worked out, he did close on a nice note. Here’s a look at horizontal:
We see his multi-tiered cascade, but you’ll note that he displayed more variation than his teammates over nearly the entirety of the season. While his movement spread was tightening up so was basically everyone else’s and he still showed more variation than the other guys. Maybe there is something to that. Here’s what that looks like removing the variation and looking strictly at his rolling average movement:
His horizontal movement was between 1.5 and 2.5 inches for most of the season before that drastic dip towards his close. It should be mentioned that we are beholden to the limitations of the data. Garbage in, garbage out, so if there were calibration issues then that adds to the noise, but with a 250 pitch sample we should be cutting through much of that. Onward, his vertical movement showed mostly downward decline over the course of the year and he closed on a pretty mean stretch of less life in his fastball. Fastballs up are a great weapon, but when they begin to flatten instead of rising as much as you’re used to they get hit a long way.
Enough with movement, let’s move on to release points. Here I have expanded upon the fastball approach above and am now including all pitches. Let’s start with the team-level vertical release point variation:
Karns displayed pretty consistent variation over most of the season, but we see down the stretch that he lost his release point showing wild fluctuations. We see something similar from Jake Odorizzi and Chris Archer, however, and while those guys regained some consistency the spike from Karns might be something more normal than not. Isolating Karns and shrinking our sample lends some insight:
Here we can see those late season spikes that came virtually out of nowhere. That looks like a flag to me. We also get an introduction to his horizontal release point variation, which also seems to show some more variation over his final thousand pitches or so. Here’s what the rest of the team looks like:
We do see that late-season climb. His peers showed more variation throughout the year and with their backdrop this might not be quite as worrisome. If there is something to this then the spike for Archer may be of concern, but that’s for another day. As before we can eschew variation and look solely at the actual points where the ball came out of his hand:
Starting with vertical you can see that over the course of the season he steadily dropped his arm slot. Early on he was throwing from north of the six foot line, but by the end of the year he was releasing from that demarcation. That’s around a 2.5 inch drop over his season and looks, to me, like a concern.
Additionally, we can see that he steadily was releasing the ball from a slot that was closer to first base than third. Typically, when a guy lowers his release point it is because he’s dropping down and coming out wider in an effort to close the angle on same-handers. That isn’t the case here. It looks like Karns was having difficulty getting full extension, or short-arming, the ball. This looks like a pretty large concern. Let’s take another look at his velocity:
I have plotted the velocity of every fastball and included a couple of different trends. The first is the 250-pitch average that we have been using, but the second is a 6th order polynomial. Both show a guy that had declining velocity until a late-season increase helped hide the next batch of fall off. Perhaps him short-arming the ball was a mechanical tweak from Hickey et. al. that helped him tap into a bit more zip, but the following decline, and him ending the season on the pine indicate that perhaps that extra giddyup came with a price.
In totality we’re looking at a guy who mostly lost velocity over the course of the season who seemed to also lose extension on his release. His fastball not only lost velocity, but it flattened out and he showed enough spikes in the variation of his velocity, movement, and release point that I have considerable reservations about his ability to finish the next three years, let alone five.
Lastly, let’s take a look at what a less optimistic surplus value might look like if we think that he’s a candidate to suffer a severe arm injury:
If the Mariners can squeeze two good years out of him before the wheels fall off then they’re still looking at a nice bit of surplus value, but if the injury comes in 2016 or the following season then you’re looking at a significant reduction in what he brings in production regardless of what he is being paid. Bear in mind that while these flags don’t paint a rosy picture for him the absolute greatest indicator of future injury is past injury and that’s a clear portrait of a miserable future.