A different way of thinking about a pitching staff!

Could we see the bullpen cart in the first, second, or third innings next year — without panicking? Photo by Marlene Koenig for TalkNats

The Tampa Bay Rays won’t make the playoffs. The historic excellence of the Boston Red Sox, as well as the great play of the second-place New York Yankees, has seen to that. But it’s undeniable that the Rays have done something remarkable this year.

Heading into the 2018 season, the Rays were a joke. A team stocked with second-rate talent, which had just traded off its franchise player in Evan Longoria, stuck in a decrepit stadium in St. Petersburg and worst of all, they didn’t even have a real pitching rotation.

From The Comeback‘s Alex Putterman on March 7:

[If] the Rays really do stick with their four-man rotation, their pitchers are going to be heavily taxed. Starters will throw more often, never getting an extra day of rest, while relievers will be asked to enter early and often piece together games by themselves. Unless Tampa is extremely careful, there could be a lot of sore arms at Tropicana Field this year.

From CBS Sports’ Matt Snyder on March 28:

With opening day looming, it’s fair to say the Tampa Bay Rays rotation is in shambles. […]

[Chris Archer, Jacob Faria, and Blake Snell are] a good trio. It’s just that it’s only 60 percent of what the Rays need.

From the Tampa Bay Times’ Tom Jones on March 28:

Oh yeah, the Rays are reinventing the wheel by using the bullpen as a fifth starter. This has disaster written all over it. Rays starters rarely go deep in games, meaning the bullpen is taxed most days to begin with. Then add into the mix that there are no long-term examples of this ever working anywhere else in baseball and you can understand why we’ll have to see it work before believing it can work.

From the Tampa Bay Times‘ Marc Topkin on April 6:

Rays manager Kevin Cash is already so weary of explaining the Rays’ — pick your adjective, within the bookends of idiotic and innovative — pitching plan of three starters and five multi-inning relievers that he joked Wednesday they were going to change the terminology.

“As of today we have eight starters; that’s our new motto now,” he said. Better, they should just scrap the idea.

The Tampa Bay Rays aren’t going to the playoffs. But just for showing up their critics, including those in their hometown paper, they can count this season as a big win. And it just might end up revolutionizing the game of baseball.

Throughout their season, the Rays never fielded a full five-man rotation, and even the four-man rotation they suggested they would put up before the season started was rarely if ever stable.

Only two pitchers have tossed more than 100 innings for the Rays this season; one of those two, Ryan Yarbrough, has only started six of the 36 games in which he has appeared to date. Only four pitchers have exclusively started games in which they have appeared; of them, one (Archer) was traded midway through the season to the Pittsburgh Pirates and another (Tyler Glasnow) was acquired from the Pirates in said trade, effectively just replacing Archer in the Rays’ unorthodox rotation, while a third (Nathan Eovaldi) was traded to the division-rival Red Sox during the season. Young ace Blake Snell leads the team in innings pitched with 169 and starts with 29, working to a phenomenal 1.97 ERA that should present him as a strong candidate for the American League Cy Young Award this year; the second-most starts have been made by Ryne Stanek, who has started 27 games for the Rays this season but only thrown 63 innings, with a 2.71 ERA. All told, a whopping 16 pitchers have made at least one start for the Rays this season.

Most baseball people look at numbers like this and say, “Huh?” Yet the Rays appear on track to win 90 games this season. Even though the dominance of the Red Sox and Yankees mean they’ll be watching the ALDS from the couch, they have to feel good about the results they got with a strategy many were criticizing early this year as short-sighted penny-pinching by an irrelevant small-market franchise.

Far from devastating the pitching staff, the Rays’ strategy paid off to the tune of a collective 3.64 ERA, good for second place in the American League behind only the superlative Houston Astros pitching staff (the league average team ERA is 4.26). While they ended up flipping their perpetually intriguing-but-underperforming “co-ace” Archer, they kept their true ace Snell healthy all year, with outstanding results. Their swingman Stanek and setup man Jose Alvarado have turned in some of the strongest seasons in the league by pitchers we would normally just describe as “relievers”.

Indeed, we have to shift our terminology as well as our way of thinking when we try to wrap our heads around the 2018 Rays.

Stanek and Alvarado have pitched roughly the same number of innings, but Stanek was used as an “opener” in nearly half of his appearances and Alvarado has appeared exclusively in relief. Faria was a dedicated starter up until he hit the disabled list in May; since returning in August, he’s made five of his six appearances to date out of the bullpen. Former starter Yonny Chirinos was also shifted into a long relief role in the second half, and in fact, he earned his only four wins of the season as a reliever (although in all four, he pitched at least five innings of the game). Yarbrough, who owns perhaps the most bizarre pitching line in recent baseball history, has only pitched beyond six innings thrice this season, but also just thrice has he pitched no more than a single inning.

So, by now, you’re probably wondering: What the heck does all this have to do with the Washington Nationals?

Maybe nothing. Maybe a lot.

The curious case of Jeremy Hellickson

Throughout the Nats’ 2018 season, rookie manager Davey Martinez was frequently pilloried on this site, by beat writers at MASN and The Washington Post, by rival broadcasters, by national sportswriters and pundits — for leaving his starting pitcher in the game too long.

It wasn’t a new phenomenon to see the Nats leave a struggling starter in the game well past an advisable pitch count in what can variably be explained away as either an attempt not to go to an often-creaky bullpen or an effort to let the man earn a “win”, as previous manager Dusty Baker was also well known for the tendency. But it was nonetheless frustrating to watch games like the Nats’ win on June 29 at Citizens Bank Park, in which rookie Erick Fedde threw 64 pitches through a laborious fourth and fifth inning, finishing with five earned runs on his line and 115 pitches thrown in order to be eligible for his first career win after the offense put up 11 runs in the early going. (Fedde exited his next start early with an injury and ended up missing several weeks with shoulder inflammation.)

We watched Martinez push Fedde, fellow rookies Jefry Rodriguez and Austin Voth, and veterans Max Scherzer, Stephen Strasburg, Gio Gonzalez, and Tanner Roark. But we saw the complete opposite for most of the season with journeyman back-end starter Jeremy Hellickson.

Acquired during spring training as a minor league free agent signing, Hellickson took a few weeks to ramp up and join the team. In the interim, the prevailing concern seemed to be building up Hellickson’s workload so that he could handle the stress of a typical 90- to 100-pitch start. As it turned out, that wasn’t how Martinez (who previously worked with Hellickson when both were in, oh hey, Tampa Bay) planned on using Hellickson at all.

Hellickson was a valued member of the 2018 team. Despite missing time with no fewer than three freak injuries over the course of the season, the latest of which (a wrist injury suffered while batting) has him out for the rest of the month, Hellickson typically gave the Nats a good chance to win when he pitched.

And yet, of Hellickson’s 19 appearances on the year all of them starts he only posted two “quality starts”. That isn’t because Hellickson was bad; it’s because the “quality start” stat requires a starter to have gone at least six innings, one more than the minimum five needed for a starter to be eligible for the pitching win. (Both of these stats, as you can see, are fairly stupid.) The Nats won 11 of the games Hellickson started; Hellickson himself took the loss just three times, twice in starts that saw him give up three or fewer earned runs. He topped 90 pitches thrice and never reached 100 pitches in a start. He will finish the season with a 3.45 ERA over 91 innings pitched.

For all intents and purposes, when he was healthy, Hellickson was considered a full member of the Nats’ rotation. He wasn’t considered a swingman or an up-and-down guy. But at the same time, it was clear that Hellickson was different from the other starters. Martinez rarely pushed him through the fifth inning, even with a lead, to make him eligible for the win. Regardless of his pitch count, once he had gone twice through the opposing lineup, he was subject to a quick hook.

Why was Hellickson treated so differently? Well, there’s no real mystery to it: He stinks third time through the order. For Hellickson’s career, batters post a .235/.283/.385 triple slash against him in their first plate appearances, a .240/.305/.419 triple slash in their second, and a .278/.341/.506 triple slash in their third. The Nats figured out that in signing Hellickson to a cheap one-year deal, they were getting a very reliable and effective pitcher for the first four to six innings of a game, but one with a tendency to hit the wall hard if asked to go deeper than that. For the most part, Martinez managed Hellickson accordingly. As a result, the Nats got good results out of Hellickson, and it seems very likely Hellickson will find a major league guarantee somewhere as a free agent this winter.

So what does this season’s usage of Hellickson have to do with the Rays, again?

A new view of pitching

The Rays showed that a revolutionary approach to managing a pitching staff that broke down many of the barriers between “starters” and “relievers” can actually work in modern Major League Baseball. The Nats’ deployment of Jeremy Hellickson showed that even a traditionalist team can still maximize the output of a limited pitcher by understanding his limitations and working to his strengths.

Heading into 2019, the Nats face a great deal more uncertainty than they have in the previous two winters. Stalwart second baseman Daniel Murphy is gone, traded midseason to the Chicago Cubs for a fringe prospect. Frontline catcher Matt Wieters will be a free agent, with no indication that the Nats are considering either backup Spencer Kieboom or failed prospect Pedro Severino as his heir apparent. Superstar outfielder Bryce Harper, coming off an up-and-down season, might or might not return on a contract that could break records for dollars guaranteed for a position player. But the biggest question mark is probably pitching.

Scherzer and Strasburg will be back, the former a future Hall of Famer coming off a Cy Young Award finalist season and the latter a great talent still in his prime but looking to re-establish himself after an uneven year. Former rotation staple Gio Gonzalez is gone, traded to the Milwaukee Brewers and unlikely to return in free agency. Hellickson will depart, also unlikely to come back after having politely but consistently expressed displeasure with the short leash Martinez kept him on all season (even though it was for his own good). Roark, who is arbitration-eligible for the third time and has labored through his second straight season of subpar results, may or may not be back.

And then there’s “the kids”. The Nats finally have 25-year-old Joe Ross back after he underwent “Tommy John” surgery on his pitching elbow last July, although his results this month have been less than stellar. Once again healthy, Fedde is trying to make his own case not to be shipped out to Triple-A Fresno to start his age-26 season. Jefry Rodriguez, also 25, has been an enigma in appearances both as a starter and in long relief this year, sometimes brilliant and other times looking completely lost. Austin Voth, who turned 26 this year, just earned his first career win with five innings of one-hit ball. Kyle McGowin, 27 this fall, earned his first major league call-up this month with an incredible year in the minors, although he has yet to be given the opportunity to start.

On the relief side, southpaw closer Sean Doolittle is a mortal lock to have his club option exercised for the 2019 season, despite missing a critical chunk of the season with an injured foot. Impressive rookie Wander Suero, emergent left-hander Matt Grace, and effective journeyman Justin Miller also seem likely to return as low-cost, battle-tested middle relievers. A slew of other young relief arms, including Jimmy Cordero, Austen Williams, Austin L. Adams, and Koda Glover, will likely figure as up-and-down pieces. Longtime “Quadruple-A” relievers Sammy Solis and Trevor Gott, who are out of options in 2019 and haven’t gotten regular usage so far this September, likely won’t be back. Inconsistent lefty Tim Collins probably won’t return. It remains to be seen whether the Nats will try to bring back former closers Kelvin Herrera, who pitched brilliantly for the Kansas City Royals but struggled with injuries and underperformance after his trade to Washington, or Greg Holland, who earned his release by the St. Louis Cardinals in August but has been highly effective in leverage opportunities since then for the Nats.

Basically, there are two ways in which one can generally assess the Nats’ off-season needs:

The traditionalist view

The Nats only have two names they can write in Sharpie for their rotation: Scherzer and Strasburg. They have an All-Star closer in Doolittle and a pretty good stable of middle relievers, most of whom can be rotated between Washington and one or two minor league affiliates (likely Double-A Harrisburg due to the impractical distance of Triple-A Fresno, although that remains to be seen and could vary by situation).

The needs are clear, no? The Nats need at least one mid-rotation starter and at least one quality setup man, and realistically, they’d probably like two of each. Roark could be brought back for probably about $8 million after arbitration, and Holland could be amenable to a return, as he and the Nats have mutually benefited from his addition to the bullpen. Otherwise, there’s the free agent market, as well as the trade market, which general manager Mike Rizzo all-but-entirely neglected last winter (he made exactly one off-season trade, a February deal to purchase the rights to superutilityman Matt Reynolds after he was designated for assignment by the New York Mets).

The Harper situation looms over the Nats’ off-season plans. If he departs for greener pastures and a bigger paycheck elsewhere, the Nats could instead make a run at extending third baseman Anthony Rendon, who has quietly been the more valuable player than Harper over the past three seasons and is set to reach free agency before the 2020 season. They would also be loath to part with either top prospect Victor Robles or cost-controlled outfielder Adam Eaton, one or both of whom could be trade chips if Harper re-signs. A Harper return could also open the door to an opportunistic trade of Rendon, whom the Nats probably won’t be able to afford long-term at the same time they have Harper locked up, although that would indicate a somewhat less than “all-in” approach to 2019.

Basically, the Nats have some shopping to do, and that situation is complicated not only by the Harper uncertainty but also by their reckless financial planning over the past two seasons. Because the Nats are over MLB’s luxury tax threshold, they can’t sign a free agent from another team who receives and rejects a qualifying offer without losing two high-level draft picks and $1 million from their international signing bonus pool. That likely means a pursuit of a star pitcher like Patrick Corbin or Dallas Keuchel would hurt the Nats’ ability to add premium young talent to their minor league system next summer. Would it be worth it? The answer to that question is a distinct “maybe”.

The unorthodox view

We need to flip the script here. Change the terminology. So we can say the Nats have at least two “horses”, Scherzer and Strasburg, whom they can count on to go deep into games and even have a chance of pitching all nine innings of a regulation win, as they’ve combined to do five times over the last two seasons. They have the rights to Roark for 2019 if they want him, and while he isn’t nearly as good as Scherzer and Strasburg and consequently won’t go as deep as they can, he’s still pitched at least 180 innings in each of the past three seasons. Plus they have their stable of “burros” in Ross, Fedde, Rodriguez, Voth, and McGowin, who have yet to prove they can be consistent six- to eight-inning starters for the Nats but have nonetheless shown that, at least on a good day, they can get outs in the major leagues. If you’re so inclined, you can add Grace and/or Williams to that category, as former minor league starters who have worked in long relief at times this season.

Horses and burros there are, provisionally named, our two “classes” of pitcher so far. Those guys can carry a team in a game. The rest of the pitching staff is made up of guys who won’t really carry you, as they can pitch one or maybe two innings at most, but they’re still a big part of The Show. Let’s call them “bulls”.

The traditionalist view says the Nats are somewhere between two and four good pitchers (one or two mid-rotation starters plus one or two setup men) short of having a complete pitching staff. But the Rays posted a winning season with a mediocre team while having just two or three horses throughout the year. They augmented those horses with a rotating cast of burros and bulls, most frequently calling on Stanek, who never pitched more than two innings a bull, then to start games that weren’t pitched by their ace Snell. The Nats have a Snell in Scherzer, and they have an Archer/Glasnow in Strasburg, and if they want, they have a Faria in Roark. They’ve got the burros, all of whom have options. They’ve got the bulls, many of whom have options too.

The Nats still need to field a winning lineup. What if they don’t have to worry about the pitching staff that much? Sure, they’ll probably need one or two more strong bulls, the way the Rays have had Alvarado to set up closer Sergio Romo, but they can cobble together starts with their horses, burros, and bulls. Whether someone like Ross or Rodriguez is starting or coming out of the bullpen, he’ll typically be used for longer than a one- or two-inning burst. Especially running a regular shuttle to continually bring up fresh arms from the minors, the Nats can build around their horses and maybe get the best out of their 25-, 26-, and 27-year-old burros, all of whom have limited control, repertoire, or both that have prevented them from making the permanent leap from minor-leaguer to major-leaguer.

What’s the conclusion?

Let’s face it: If the Nats re-sign Harper, they’re not going to have a lot of salary space left over to basically rebuild their pitching staff. And even if they don’t re-sign Harper, they’re still going to incur stiff penalties for signing a top free agent (at least, one who wasn’t traded during the season) that will hamper their ability to rebuild a diminished farm system.

Maybe the best thing to do is split the gap between the center field option and the one here that is out of left field. If the Nats re-sign Harper, then they can probably use Robles or someone else in a trade to get another steady starter a horse, in the parlance of this post. If they don’t, then they probably have the money to sign a Corbin or a Keuchel, or at least someone like Eovaldi, who won’t have a qualifying offer attached. That gives them three.

But rather than worry about whether Ross, Fedde, et al. the burros can step up to fill out the rest of the rotation, they could stick with a three- or four-man unit and infill with between one and three “bullpen games” per week. Those “Johnny Wholestaff” outings could be started by either a pitcher expected to get somewhere between two and six outs before being replaced, or by a pitcher who will be managed like Hellickson, being lifted without regard for the outmoded win or quality start stats after he’s gone a couple times through the order, or however deep his manager trusts him to pitch effectively.

Obviously, ditching the five-man rotation is a sea change in both roster management and game management, and it’s one that would require complete and total buy-in at every level, from the front office to the dugout maybe even down to the minor league affiliates. Pitchers would have to trust that they will get their chances to earn wins, if they care about those counting stats, and not take it personally when they are hooked from a game before they’re eligible. If they’re on the taxi squad, they will have to trust that if they perform in the majors, they’ll continue to be called back to the majors, and that they’re a part of this team whether they’re a Washington National or a Harrisburg Senator at any given moment in time. The manager and coaches will have to keep the strategy straight, managing for team wins rather than pitcher wins and trusting their burros and bulls to do the jobs they’ve been assigned to do. The front office will need to keep the team resupplied at all times, taking advantage of options to rotate burros and bulls between the majors and minors and even using the 10-day disabled list strategically when it makes sense.

And it has to be said that any radical upheaval can backfire. The Rays’ own beat writers ridiculed their approach before they saw it work; they had their reasons to doubt it, and they could have been right. If the Nats do something like this, they’ll have to be prepared for the doubters, and they’ll have to be prepared to look silly for at least a month or two before they decide whether they need to change course.

As 2018 winds down, the Nats are playing out the string on a disappointing season, and the future direction of the franchise is in flux. They have a chance to right this wrong in 2019. Maybe there’s something to be said for trying something different next time.

This entry was posted in Bullpen, RizzoForADay, Roster, Sabermetrics. Bookmark the permalink.