Team Non-Tender Could Outhit the Orioles, or Love Me Non-Tender

If we were to build a team from this offseason’s non-tendered players, would that team be better than the Orioles? How about the Marlins?

Due to the quirks of the arbitration system, players who are very much worthy of a major league roster spot may be jettisoned simply because their projected production is not worth their projected cost. Though their pitching would have their non-tendered fans shaking their heads almost nightly, Team Non-Tender could assemble a decent lineup.

Let’s look at who we’re working with for our non-tendered position players. These projections use Steamer600, which assumes all players get 600 plate appearances. Some of our non-tenders are projected for part-time work, but because we’re putting them on an imaginary team, we are envisioning them as starters. Several of these players have signed, but for the sake of consistency, we’ll stick with their projected salaries.

That adds up to 12.1 wins—not bad for a team of castoffs. The catcher projection is roughly an even split between Herrmann (2.0 WAR) and McCann (1.4). Herrmann never has logged more than 265 major league plate appearances in a single season, so I feel better calling this an even split than a standard starter/backup relationship with McCann, who was the Tigers’ primary catcher last year.

Our bench isn’t terrible, either. When injuries and slumps rear their heads, Team Non-Tender can turn to Yangervis Solarte (1.2 projected WAR over 600 plate appearances), Gorkys Hernandez (0.8), Tim Beckham (0.8), and Chris Owings (0.7). The total output from a given position is closer to 700 PAs, so to get a total team projection, we’ll tack on 300 PAs from Hernandez, 400 from Solarte, and 100 from our catching duo. Add that all up, and we have 13.5 wins from the position player side.

How does this set of position players match up against the rest of the major leagues? Two caveats: The offseason has many transactions left, and these projections will change. (There are almost 60 projected wins left on the free-agent market.) Secondly, we’re mixing Steamer600 for Team Non-Tender and regular Steamer projections for everyone else. Regular Steamer is based on actual playing projections, and we’d create more confusion than clarity by switching to Steamer600 for our real baseball teams. So, with all that in mind, here are the 10 worst projected teams, with Team Non-Tender assuming its spot.

Hey, that’s not bad! Team Non-Tender is clearly better than two teams, in the middle of a clump with four more, and within spitting distance of a few squads that fancy themselves contenders. Those teams, however, have some impressive pitching, and that’s where TNT falls apart. Here’s the starting rotation.

Team Non-Tender would have to convert a reliever or play around with bullpen games or—never mind, no amount of strategic innovation can save this pitching staff. Also, handing 200 innings to Shoemaker is a tad optimistic, given he managed only one start in March and six in September in 2018, and only 77.2 innings the previous year. The same goes for Miller, whose 2018 output amounted to 16 innings, over which he gave up 21 runs. Sampson threw 149.2 innings last year, but only 23 of those were in the major leagues, the rest coming in Triple-A. Only Fiers has managed a starter’s workload over the last four years, but even he topped out at 180.1 innings. While the 5.8 wins projected for this quartet, at 200 innings apiece, is nothing special, the standard Steamer projections, which cuts them down to a cumulative 2.7 wins, are probably more accurate.

Things only get a little better in the bullpen. Blake Parker misses bats, and Hunter Strickland and Brad Boxberger were reliable closers not long ago (not so much last year). Matt Bush’s control abandoned him in 2018, but if that returns, he can make better use of his impressive velocity. Cory Gearrin provides middling quality for the middle innings. That’s pretty much what we’re working with for this team’s bullpen. The entire unit projects for two tenths of a win using Steamer’s inning projections and 1.2 WAR if we assume 65 innings apiece (or 1.5 WAR if we bench Zac Curtis). Any leads Team Non-Tender was able to carry into the latter third of the game would be in danger until the final out.

Surely no real baseball team projects as badly as this, right? Well, it depends on how you count. If we give every starter 200 innings and every reliever 65, only the Orioles stay in single digits for their pitching staff WAR projection, and their 9.4 WAR still clears TNT’s staff. (We’re actually cheating just a tiny bit—these figures only count positive WAR projections. If you don’t do that, every team plunges into negative numbers due to a large group of relievers projected for a few terrible innings suddenly having their output multiplied many times over). Using Steamer’s estimates for playing time, the Orioles, Royals and White Sox still beat out TNT’s pitching staff as long as we use Steamer’s playing time projections. Otherwise, Team Non-Tender edges them out.

We can keep splitting these hairs, but the point is that a real rotation led by the Orioles’ Alex Cobb, Dylan Bundy and Andrew Cashner is only slightly more inspiring than the hypothetical one above. The same goes for Royals hurlers, headed by Danny Duffy, Jakob Junis, Brad Keller and Ian Kennedy. The White Sox at least have some exciting arms in Carlos Rodon, Reynaldo Lopez and Dylan Cease, but the on-field results aren’t projected to be much better than Fiers, Miller et al. As for the bullpens, you probably would take the TNT unit—plenty of bounce-back upside there—over the Orioles’ or Royals’ relievers as they currently stand.

All that brings us to the question of why Team Non-Tender can put together a passable lineup, an almost-interesting bullpen, and a rotation of Mike Fiers and three guys who combined for 225 innings and a 4.56 ERA last year. The imbalance of this team reflects, among other things, how the market has diverged from the MLB arbitration system in how it evaluates talent and compensates players.

The first thing to understand about the arbitration system is that players are virtually guaranteed a raise every year. The Mets’ Travis d’Arnaud, for instance, is projected by MLBTR for a small bump up from his 2018 salary of $3.5 million despite playing in only four games last year. Players expected to miss some or all of a season are obvious non-tender targets, because the arbitration system rewards their past performance more than their future outlook. (This was the fate of the A’s Kendall Graveman, who is recovering from Tommy John surgery and was non-tendered.) As for how past performance is judged by the arbiters, this, as Matt Swartz explains below, is based more on “traditional” stats than modern metrics like wOBA and DRS.

Hitters are typically evaluated using batting average, home runs, runs batted in, stolen bases and plate appearances. There are some positional adjustments, but typically the added defensive value of a shortstop relative to a first baseman is not as important in arbitration hearings as it is on the free agent market…Pitchers typically are evaluated using innings pitched and earned run average. Starting pitchers are rewarded for wins, and relievers are rewarded for saves and holds.”

Power threats like Justin Bour and Jonathan Schoop would have been rewarded by arbiters for their home runs without getting dinged too much for their respective defensive limitations and lack of walks. The White Sox would have had to pay for Avisail Garcia’s history of high-teens home run production and his one-off high batting average season in 2017. Flores and Solarte fit the pattern too: modest power, limited when it comes to defense and patience. While the Aaron Judges of the world are still rare, it is easier than ever to find mid-teens home run power, especially without great defense or strong on-base skills. Billy Hamilton is ever-unique, but his elite speed was more compelling to arbiters than to the Reds in what would have been his most expensive arbitration year. The core of Team Non-Tender’s lineup are players with enough longevity to become (relatively) expensive and a skill set that would have been more prized 20 years ago.

On the pitching side, our closest equivalent to slightly-above-average power is saves. Boxberger, Parker and Strickland have plenty of those to their names, though no front office in baseball cares about the saves themselves, just the skills that helped earn them. That’s why TNT has a potentially serviceable bullpen of former closers.

For starting pitchers, the most obvious place to look would be pitchers who somehow notched a bunch of wins (the back of the baseball card kind, rather than Wins Above Replacement) without good underlying results—reverse deGroms, if you will. The opportunities for questionable starter wins have dropped precipitously as teams employ quick hooks and make use of bullpen games. Starter innings after the sixth frame have been cut nearly in half since just 2014! While the arbitration system likely rewards ERA more than teams would, ERA obviously correlates with its more predictive indicators, and there is no particular trend around pitchers beating their FIPs and xFIPS. Only Fiers really seems to fit the pattern of many past non-tendered position players as a good-not-great contributor who got too expensive due to the arbitrary nature of the system that determines his salary.

There is one more factor dictating the balance of Team Non-Tender, and that’s randomness. Last year’s TNT basically wouldn’t have had a lineup. Still, we see some of the same patterns: slugging first baseman (Matt Adams), weak hitting speedster (Terrance Gore), closer (Hector Rondon), Tommy John patient (Drew Smyly), and Mike Fiers (Mike Fiers). The 2016 group was unremarkable beyond checking some of those boxes—Chris Carter was the slugger, Ben Revere the speedster, Tyson Ross the injured pitcher. Christian Villanueva hadn’t played a game at the major league level when he was non-tendered by the Cubs that offseason and now fits the “decent-power, low-walks” profile of this year’s non-tenders.

Perhaps Team Non-Tender’s strong lineup this year is partly a fluke, but it reflects recent trends in baseball around power hitting and pitcher usage. Instead of the 26th-best players on 25-person teams, we see players who represent the inevitable divergence between a fiercely competitive market and a system that rewards yesteryear’s idea of winning baseball. Middling power without enough on-base or defensive skills isn’t compensated the way it used to be…but it still might be enough to beat the Orioles.