/cdn.vox-cdn.com/uploads/chorus_image/image/69965371/1340431081.0.jpg)
It’s difficult to know exactly what to make of Rays pitching in 2021.
We know that what they did was radical, and we know that it worked. With both a rotation and bullpen largely devoid of big names and established starters (particularly after the Tyler Glasnow injury), the Rays posted a 3.67 ERA, the fourth best in baseball, and top in the American League.
After adjusting for stadium, the Rays numbers were still very impressive: fifth best ERA- in the AL, third best FIP-, and second best xFIP-.
They did this by blurring the line between starter and reliever more than they ever had before. Both the long starter and the short reliever were de-emphasized, and in their place they elevated a whole range of good pitchers who could effectively get between four and twelve outs on any given night.
:no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/22905722/2021_MiddlePitchers.png)
The result was that they both aggressively shielded their starters from the times through the order penalty while aggressively shielding their relievers from overwork.
The most heavily used Rays reliever, All-Star Andrew Kittredge, pitched 71.2 innings, which would have placed him sixteenth overall in total relief innings pitched. But more clarifying than total innings, which can be effected by all sorts of external factors (like injuries and trades) is the overall distribution of clumps of appearances.
Simply put, while asking their relievers to do among the most work in baseball in total, the Rays were also among the best at not making those relievers work in back to back days, or in bursts of appearances over several days.
:no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/22905480/2021_Team.png)
But why did they do it?
It could be that they were forced to by a poorly sequenced offseason shuffle. At the beginning of the offseason, they declined to pick up Charlie Morton’s $15 million option, but then attempted to resign him at a lower cost. Morton said “no thank you” and took his curveball to Atlanta.
But then the Padres called with an unexpected offer for Blake Snell that Erik Neander couldn’t refuse, which cleared $10.5 million in 2021 salary (and more in future years) while finalizing the Rays catcher position at league minimum.
It’s likely that if the Snell trade had happened earlier, Morton would have been a Ray in 2021. But because it happened when it did, the Rays front office was left holding a bag of money with no big-ticket pitchers to spend it on. After signing Chris Archer, who was coming back from injury, they arrived at a point late in the offseason where, to my eye at least, the only remaining free agent options obviously able to help a championship caliber team were Rich Hill and Collin McHugh. So the Rays signed both Hill and McHugh — pitchers who were a good bet to provide quality rate stats, but a bad bet to provide full starter innings.
The 2021 pitching plan could also have been a considered response to the pandemic-shortened 2020. There’s a rule of thumb in baseball these days (which Rays executives and coaches have repeated multiple times in interviews) that you shouldn’t ask a pitcher to pitch more than 20 innings more than he has the previous year. The idea is that arm strength is gets built up and maintained over time, and that asking a pitcher to make a big jump in innings puts him into dangerous overexertion territory.
The 2021 Rays were always going to be forced to rely on pitchers who, because of age or injury history, had not been fully and methodically built up, but then 2020 happened and nobody in baseball pitched more than 84 innings. What would that do to long-term arm strength? Nobody knew; everyone was guessing.
Both of these explanations are part of the story. But also — and this is the thing that should frighten baseball traditionalists — the Rays 2021 pitching strategy was not purely a reaction to their circumstances. It was the culmination of a multi-year process of experimentation. And it worked.
And the Rays weren’t alone on this path.
When Middle Relief is Plan A
Consider for a second what prompts a three or four inning pitching appearance (for clarity, I’ve called these “Bulk Bridge” appearances on my charts and defined them as times when a pitcher recorded at least seven outs but not more than twelve).
Traditionally, these happen when a starter is replaced early, either because of injury or poor performance. Sometimes it’s that starter who’s recording the bulk bridge appearance; other times it’s a swingman long reliever (often the least established pitcher on the roster) thrown in to emergency relief duty to soak up innings in a losing effort. Either situation is likely to mean that the pitcher is giving up runs.
Below are MLB teams ranked by the number of times their pitchers have had an appearance between seven and twelve outs in 2021, with the average ERA of those appearances.
2021 Bulk Bridge Appearances
Team | Number of Instances | Average ERA |
---|---|---|
Team | Number of Instances | Average ERA |
TEX | 75 | 7.85 |
BAL | 74 | 8.16 |
SDP | 70 | 6.25 |
DET | 69 | 4.82 |
LAA | 69 | 6.82 |
TBR | 67 | 4.53 |
MIN | 58 | 8.61 |
NYY | 57 | 5.59 |
MIA | 57 | 7.29 |
BOS | 54 | 5.42 |
CHC | 54 | 7.46 |
PHI | 53 | 6.08 |
ARI | 53 | 7.77 |
NYM | 52 | 6.25 |
CLE | 49 | 8.05 |
PIT | 49 | 7.65 |
TOR | 48 | 7.80 |
KCR | 47 | 7.20 |
HOU | 45 | 4.82 |
STL | 42 | 6.60 |
SEA | 41 | 7.84 |
COL | 40 | 8.14 |
CHW | 39 | 5.61 |
MIL | 39 | 6.08 |
ATL | 35 | 8.70 |
SFG | 34 | 6.52 |
WSN | 34 | 8.56 |
CIN | 34 | 9.09 |
OAK | 34 | 7.48 |
LAD | 28 | 4.52 |
The two teams with the best ERAs in these appearances are the Rays and the Dodgers — both teams who value fluidity in their pitching roles — but the Rays have posted that ERA at high volume, while the Dodgers, thanks to their excellent star-filled rotation, have recorded bulk bridge appearances the fewest times in baseball. The only other teams to post ERAs below 5.00 are the Astros (not unexpected, they’re a championship-caliber team run by a former-Rays baseball ops executive in James Click) and the Tigers.
Why do some teams have so much more success in these short pitcher appearances than others? Because not all bulk bridge pitching appearances are unplanned. Sometimes, when the bullpen is well-rested, it makes sense to pull a back-of-the-rotation starter early, even before the fifth inning. And if the pitcher knows ahead of time that his team doesn’t need him to pitch deep into the game, he can pace himself appropriately for the innings his team does need.
When a three or four inning appearance is the plan going in, it’s more likely to be completed with a good ERA. I’ve defined a “blowup” as any pitching performance with an ERA over 9.00, and have listed teams out below by the percentage of non-blowup bulk bridge pitching appearances.
2021 Non-blowup Bulk Bridge Appearances
Team | Total Appearances | Non-blowup | Percent Non-blowup |
---|---|---|---|
Team | Total Appearances | Non-blowup | Percent Non-blowup |
LAD | 28 | 24 | 86% |
DET | 69 | 56 | 81% |
TBR | 67 | 53 | 79% |
CHW | 39 | 29 | 74% |
PHI | 53 | 39 | 74% |
HOU | 45 | 33 | 73% |
BOS | 54 | 39 | 72% |
SDP | 70 | 48 | 69% |
NYY | 57 | 39 | 68% |
NYM | 52 | 35 | 67% |
LAA | 69 | 45 | 65% |
STL | 42 | 27 | 64% |
MIL | 39 | 25 | 64% |
SFG | 34 | 21 | 62% |
CLE | 49 | 29 | 59% |
TOR | 48 | 28 | 58% |
MIA | 57 | 33 | 58% |
COL | 40 | 23 | 58% |
KCR | 47 | 27 | 57% |
CHC | 54 | 31 | 57% |
TEX | 75 | 43 | 57% |
PIT | 49 | 28 | 57% |
SEA | 41 | 22 | 54% |
MIN | 58 | 31 | 53% |
BAL | 74 | 39 | 53% |
ARI | 53 | 27 | 51% |
WSN | 34 | 17 | 50% |
CIN | 34 | 16 | 47% |
OAK | 34 | 16 | 47% |
ATL | 35 | 15 | 43% |
Who Isn’t Blowing the Bridge?
The Dodgers posted the highest percentage of non-blowup bridge appearances, while also recording the lowest number of bridge appearances overall.
That’s important, because it tells us that the shortened start and/or the lengthened relief outing is not just a gimmick pulled by teams without the ability or desire to put together a competitive starting rotation. Even the Dodgers, overflowing with established star-level starting pitchers, do this on purpose. It’s worth looking more closely at what pitchers actually made the bulk bridge appearances.
Leading the way with eight appearances was Tony Gonsolin, a good young starter who’s been on and off the Injured List this season with shoulder inflamation. The Dodgers limited the length of his appearances to protect that shoulder, but also to get major league value while he worked back from injury — there’s only so many rehab appearances a pitcher can make, and anyway, why have your best pitchers throw minor league pitches (while lesser pitchers throw in the majors) when you can create a situation for them to win you games while rehabbing at the major league level.
After that comes former Cy Young award winner David Price, an aging starting pitcher squeezed out of that role on the championship-caliber Dodgers, but fit into another.
And then there’s Mitch White, a young relief pitcher with no great pitch but three pitches better than average. He’s probably overexposed as a starter, but his skill and repertoire would be underutilized as a short reliever.
Flipping over to the high-volume Rays, you can see that they’re using essentially the same archetypes.
Michael Wacha and Chris Archer fit the Gonsolin model of starters who have struggled with injury and so spent the season working through their rehab and ramping up innings on the major league level. Shane McClanahan, Drew Rasmussen, Luis Patiño are of a similar type, but rather than injured they’re just young, and have spent their year building up their innings slowly, sometimes in the majors. Scheduled short appearances are the way to do this.
Rich Hill and Collin McHugh are good aging former starters in the David Price mold, who the Rays targeted in free agency, happy to build a plan around what those good still had to offer. Injuries forced Hill into a longer role than was probably originally intended, and to his credit he did great work.
Dietrich Enns, Chris Mazza, and Trevor Richards are the Rays versions of Mitch White, and are also all pitchers the Rays explicitly targeted, either in free agency or in trades. They’re not quite MLB starters, but their legitimate arm talent would be wasted in purely short relief.
The one role that the Rays have also used extensively that the Dodgers didn’t this year is the Josh Fleming-Ryan Yarbrough archetype. These are legitimate starting pitchers who can go long when they need to — Yarbrough pitched a complete game! — but the Rays know that they will win more games if they don’t ask the Yarbroughs and the Flemings to go long every time out.
The Rays “Trick” Is To Actually Go For It
The Dodgers built a starting rotation where they don’t need to sharply limit their starters innings for performance reasons (yes, they still limit their starters to some degree, because they understand the TTOP, which applies to everyone). For them, the rotation was a given, and then the main balancing act was between short and longer relief.
But the Rays embraced that balancing act throughout the entire pitching staff — starters vs longer relief vs short relief — and that’s how they were able to build a successful run prevention unit with only one sure thing long starter (Tyler Glasnow, who then got injured halfway through the year).
By fully embracing the dissolution of set bullpen roles and trusting in the power of comparative advantage, they spread out the total innings while asking each pitcher to give as much as he could (whether that be one, two, three, or four innings) but not more.
And by targeting and developing pitchers who are able and suited to give high-quality relief innings in bulk (McHugh, Mazza, Richards, and others), and then fitting their injury and workload-limited starters (Wacha, Archer, Patiño, Rasumussen, McClanahan) into a similar and complementary pattern of bulk pitching on a schedule, they were then also able to pull their full starters (Yarbrough, Fleming, Hill, as much as the full starter label can apply to any Ray) early to maximize performance without forcing the short relief bullpen into bursts of overwork.
The Rays have played around with this pitching idea — that middle relief should be the job of very good pitchers — for years, most prominently with role conversions of Erasmo Ramirez, Matt Andriese, and Jalen Beeks. But the American League leading 2021 Rays are a full culmination of these ideas. There’s interactive value to going all in.
The Future Isn’t Coming, It’s Already Here
There are two charts that should terrify baseball traditionalists. The first, which appears above, is the MLB teams ranked by percentage of non-blowup bulk bridge appearances. It shows that six of the top 10 teams by that ranking are in the playoffs — good teams with good pitchers using them intentionally in middle-length outings.
MLB is a copycat league, and even if a team doesn’t want to go quite as far out onto the limb as the Rays, there are plenty of models for how this strategy works.
The second is a close look at one of the four non-playoff teams on that list, the Detroit Tigers.
In what ended up as their best season by ERA- in eight years, after nearly a decade in the wilderness, the Tigers went full Rays. That chart shows an intriguing mix of young prospects being worked slowly into the majors, and interesting older pitchers who for one reason or another never quite made it as topline starters.
There’s a playbook now for what to do when you don’t have the means or desire to roll out five good, established starters, and it includes abandoning the closer and blurring the line between starter and relief. And it works.
This flexibility in the pursuit of talent, more than marginally impactful sequencing changes like The Opener, is the story of the Rays in the past half decade, and also the story of modern baseball.