With Evan Marshall’s Stock Set to Rise, Now the Time to Buy

You might look at Evan Marshall‘s 9.85 K/9 last season and middle-of-the-road walk rate last season and conclude he was very good, but not elite. But if you did that, you’d be wrong. In a season when the team’s ERA ranked 26th in the majors, any Diamondbacks pitcher who did well deserves a second look — but Marshall should cause you to do a double take.

A quick word of caution: Marshall didn’t quite pitch a full year last year, finishing with 49.1 IP. In terms of writing for the site, I’m attracted to unusual things. You probably are, too. So just note that although Marshall made the group of 142 “qualified” relievers last season, he did so just barely (cutoff was 49).

Marshall struck out just over a quarter of the batters he faced. That 25.7% K% ranked him 51st out of 142; good, not great. His 8.1% walk percentage tied him with three others at 69th; that’s dead average.

Nonetheless, those skills made him special in combination with another stat valued by us sabermetrics types (maybe more than we should?): ground ball percentage. As Jeff and I just talked about in reference to ERA estimator statistic SIERA on Episode 15 of The Pool Shot, ground balls are important because: 1) although they fall in for hits about as often as fly balls when they’re in play, unlike with fly balls, ground balls don’t go over the fence for home runs; and 2) when a pitcher has a particularly high fly ball (think: Addison Reed) or ground ball (think: Brad Ziegler) rate, the percentage of those hits that fall in for hits is actually even lower.

Ziegler is without question the best example. Get a ground ball rate like his, and it means that not only is he probably letting up fewer homers (and getting more double plays), but he’s causing batters to swing over the ball more, in addition to more often. The average ground ball might fall in for a hit 12%-20% of the time. Cause a batter to hit a slow roller, however, and your chances of getting an out are even higher.

A high ground ball rate is rare for strikeout pitchers. For the most part, either you’re getting guys to whiff, or you’re getting guys to swing on top of balls in counts earlier than two strikes. And this is how Evan Marshall set himself apart.

Marshall’s ground ball rate last year was 10th among the 142 qualified relievers, at 60.7%. That’s very good, and in most of the ten most recent seasons, that would have ranked in the 5-7 range. But the key is the combination of ground ball rate and strikeout rate. Of the nine relievers ahead of him in GB%, none had a strikeout rate that sniffed 25%; only one even had a rate above 20% (Zach Britton, who just had one of the best reliever seasons in recent memory, was at 21.8%). You have to slip down to 17th on the GB% list before finding a strikeout rate better than Marshall’s (Zach Duke, who just got a pretty sweet free agent deal).

This is where SIERA really flexes its muscles:

That SIERA would put Marshall in the top 20 percentile of qualified relievers this season (which is, itself, a subset of relievers that are particularly good, as the relievers who tend to not qualify tend to be worse). As a rookie, Marshall looked like he would be the best reliever on at least some teams. And one of those teams was the Diamondbacks.

You might say “oh, that’s a small sample,” and you’d be right. Steamer projects Marshall with a 3.32 ERA next season, and ZiPS at 3.51. Chances are very good that he’ll be worse this season, not better. But consider that the Marshall statistics we’re talking about came at a time when most of the rest of the pitching staff suffered a kind of apocalypse of mediocrity. Based on what happened to the rest of the staff, we’d have expected Marshall to pitch worse than his true talent last season, not better.

Those 142 qualified relievers? 32 were rookies, and there were some damned good rookies. Marshall tied for 5th in SIERA among them, still looking particularly strong — and I’m going to venture a guess that there are few seasons in which we get performances from rookies like Dellin Betances, Brad Boxberger and Ken Giles. Marshall tied with Chris Hatcher, who did not have a helpful ground ball percentage (47.5% GB%), but who made up for that with very good control (5.2%). Yes, that’s the same Hatcher who was just acquired by the saber supergroup known as the new Dodgers front office.

Marshall clearly has the skills to close — to get a strikeout when a runner is on third with one out, to get a double play with a runner on first. His home runs per fly ball rate was nothing special last year and there’s no reason to think it’ll be special in the future, but that’s why having a high ground ball percentage is so nice (fewer fly balls to start with). He’s also not particularly predictable, throwing four different pitches between 33% and 16% of the time in 2014.

I think there’s every reason to bet on Marshall — that sexy 60.7% GB% is not that far off the 59.8% GB% rate in a similar environment with Triple-A Reno. And while he had a really weird blip in 2012 with very poor K% in the 11-12% range, his 25.7% major league K rate looks similar to what he did in Single-A and seems to match what our expectations would be after his 30.7% K rate with Reno in 2013.

Marshall’s time isn’t here yet, but it may not be that far off. And while it looks probable that the D-backs can control him for six more seasons (his 146 days of service time fell short of the cutoff for a year, 172 days), there is a pretty good chance that he could end up eligible for arbitration before the 2017 season.

The D-backs’ handling of the Addison Reed situation this offseason was pretty bad. I’m not faulting them for the $4.875M settlement — that was in line with my expectations after figures were exchanged. The problem started before that, when the team grossly overbid what was going to be Reed’s likeliest number (I’m using the $3.8M MLBTR projection as a stand-in there). But that damage is done — Reed is getting what should have been his second arb salary in his first arb year, and it’s only going to go from there, regardless of whether the team settles, signs Reed to an extension, or what have you. If Reed is going to be paid like a closer, then, well, fine.

The Reed situation gives the D-backs cover to pay Marshall like he’s not a closer. Generally speaking, Jeff and I have been proponents of approaching players about extensions before sure money starts to loom in the background. It certainly worked with Paul Goldschmidt, and considering the deep discounts we’re talking about here, frankly, they could explode in the team’s face more often than not and still be a very good policy that saves the team money overall.

Sean Doolittle was one year more established in the majors when he signed his incredibly team-friendly extension with the Athletics just under one year ago, but if my math is correct, Doolittle would have been eligible for arbitration for the first time this winter as a Super Two, just like Marshall may be two years from now. And yet Doolittle will make just $800k this year (compare what his arbitration case would have been like, compared to Addison Reed’s). Doolittle has $8.6M due him for what would have been his final three arbitration years (all for a price slightly higher than what Reed is now likely to make next year). And Doolittle is probably better than Marshall, and all he really got in return was that those sub-arbitration salaries became guaranteed, and he’ll be due a $500k buyout if his team declines either his 2019 option ($6M) or 2020 one ($6.5M).

How was this Doolittle extension possible? Well, he wasn’t closing — that happened right afterward. The Athletics had moved for Jim Johnson and signed him to a $10M, one year deal, and at the time, the industry thought it was very curious that the budget-conscious team would allocate money that way. Once the wheels immediately came off for Johnson and he was essentially unplayable, the Athletics looked pretty dumb. But if you view the Johnson deal in light of how they handled Doolittle… not so much.

Now’s the time to pull this maneuver with Marshall, even though it is a year ahead of the Doolittle schedule. Why? Because of Reed‘s schedule. Reed is now in line for a 2016 salary that will be reasonable in the context of it being a one year commitment (2017, after all, not guaranteed), but it’s not going to be good value. Teams are very unlikely to trade a worthwhile package for Reed a year from now, which makes July 2015, in all likelihood, the best time to unload him. Reading the tea leaves, Marshall could be installed at closer in just five months. And if he becomes the closer before his arb salaries get guaranteed at a discount, the D-backs would be making the very same mistake they (or the White Sox) made with Reed.

A six year, $11M offer would present Marshall with the opportunity to be set for life (although he’d obviously be free to turn it down). And it would do two great things for the D-backs: 1) give them the chance to juice an extra $15M or so more of surplus value from Marshall, if everything breaks right, and 2) make it so that the D-backs definitely have a way to make sure that the closer position — and the saves that come with it — don’t poison a different reliever’s arbitration well. I know it’s early, and I know it would seem strange. But being proactive is the only way to stay ahead.