One of the rites of Spring is being reminded all in baseball is not as it sounds. For example, sometimes you hear about pitchers going through “dead arm,” and your instinct is to freak out because dead surely sounds like an irreversible condition. But then you’re told, no, “dead arm” is a temporary malady , don’t worry, the arm will come back to life, which it inevitably does. A “sore arm,” which you’d figure is just sore, like you might be after a little too much snow-shoveling or vaccine-getting, is much worse than a dead arm. Unless it, too, is a passing panic, though don’t get a baseball person started on a sore elbow, which is worse than a sore arm, even though an elbow is a component of an arm.
Carlos Carrasco  hasn’t pitched in exhibition competition yet this Spring because of elbow soreness . That’s alarming but not an alarm. Carrasco has had elbow soreness before, especially this time of year, and it hasn’t thrown a roadblock into his pitching. Yet it has pushed his timetable back and, though Luis Rojas says his elbow coming along , his right hamstring has been strained , thus it appears unlikely he’ll be ready for Opening Day. (Update: the hamstring has been diagnosed as torn, so Carlos will be a spectator for a while.)
But ready for what on Opening Day? All Carlos would be asked to do on Opening Day is jog to a foul line at Nationals Park and be introduced to a slight murmur of discontent. Jacob deGrom will (knock every piece of wood you have within reach) pitch the Opener for the Mets. Carrasco’s presence on April 1 would be comforting and completist, but from a box score standpoint, unnecessary, unless Cookie is more of a pinch-hitting threat than we’ve been led to believe.
If you’re depth-charting the rotation, you begin with deGrom, then move on somewhat reluctantly to everybody who isn’t deGrom. Ideally, Long Island’s Own Marcus Stroman likely starts the second game, a fully unsore Carrasco is third, and Taijuan Walker is fourth. Or Walker is third. Or Carrasco, hypothetically healthy, is second. Or Walker is second and Stroman, ol No. 0, is third or fourth; how particular can Marcus be about a number if he picked to adorn his uniform a digit synonymous with nothing ?
Though certain guys align in certain places in your sports-loving mind, it doesn’t much matter in a vacuum who pitches when after the first game. It doesn’t really matter for the first game except symbolically, though the symbolism of your ace being front and center as a new season begins shouldn’t be underestimated .
What/who we’re leaving out here is someone connected to another rite of Spring: the battle for the fifth starter. There’s almost always “the battle for the fifth starter,” which in the grand sweep of history falls somewhere between the Battle of Gettysburg and the Battle of the Network Stars. Most years the nominal fifth spot in the rotation is open. Once in a while the Mets are so established in their starting pitching that they perceive themselves to be all set, such as when the Mets rolled out Syndergaard, deGrom, Wheeler, Matz and Harvey in sequence in April of 2018…which lasted two entire turns. When you’re sure you’re all set is when you discover you’re not.
The way baseball commences its action — with built-in off days to protect against bad weather in most places; bad weather beyond Opening Day a frequent possibility; and every team that has a legitimate ace wanting to get the most out of its legitimate ace, — generates a related phrase: “They won’t need a fifth starter until…” The Mets have a fifth game on April 6, but if they decide getting the most out of deGrom is paramount and thus use him five days after Opening Day, they won’t need a fifth starter until April 7, the sixth game of the year. Unless it rains in an inconvenient fashion, in which case all bets are off (if, in fact, anybody bets on the identity of fifth starters).
The Mets have three acknowledged fifth-starter candidates in David Peterson, Joey Lucchesi and Jordan Yamamoto. In the unlikelihood Carrasco comes back very soon, he’ll be on track to be the fifth starter the Mets use in 2021 but not exactly “the fifth starter”. Carrasco’s credentials imply he is too highly valued to be “the fifth starter”. “The fifth starter” is a designation  in a way being the second , third or fourth starter isn’t. Sort of like we have leadoff hitters and cleanup hitters but nobody really makes much of the batter lurking in the six-hole.
Baseball teams largely went along without fifth starters for a century. The old four-man rotation lingers in the baseball subconscious as an extended “when men were men” moment of inner toughness to which we quit aspiring  in our quest to keep arms from deadness and soreness. At some point, the Mets and everybody else will meander into a debate about having six starters. That’s not based on an appraisal of personnel. It’s just what happens every year. You have more than five starters and you worry. You have fewer than five starters and you grow apoplectic, even in this age when relievers starting games for an inning is considered clever. A fifth starter should be the “just right” in a Goldilocks context. Instead, perhaps because of our vestigial reverence for the biggest of the big fours (the ’71 Orioles, the ’93 Braves, the ’11 Phillies), fifth starters are cast into the role that Benjamin Franklin envisioned for the vice president of the United States — “His Superfluous Excellency ” — rather than treated as equitably vested 20% partners in any given rotation.
Fifth starters tend to have their utility lopped off come postseason. Remember what the Mets, innovators in the religion of four days’ rest  under Gil Hodges and Rube Walker, have always done with their fifth starters in their playoff years? They’ve assigned them to the relief duty. In ’69 we used only three starters. In ’73, George Stone, the fourth starter of his day, got only one start, and you know it wasn’t in the World Series. Rick Aguilera’s robust second half in 1986 (9-4, 2.64 ERA from July 12 forward after a horrid first three months) drew him a seat in the bullpen. Bobby Ojeda’s hedge-trimming made the fifth-starter point moot two years later. Orel Hershiser, Glendon Rusch and Bartolo Colon each went the Aguilera route in their respective postseason Met years of 1999, 2000 and 2015. The Mets barely had four starters to get them through October 2006 and only one postseason game altogether in 2016.
We should be so lucky to have fifth-starter issues in the 2021 postseason. We don’t know if we’ll have our very own postseason this year. We’re reasonably confident we’ll have a “this year,” however — and have it with people ! Eventually we’ll have “the fifth starter”. As for who will be our plain old fifth starter, which is to say the fifth starting pitcher the Mets use in a season, that will be up to the elements, injuries and array of unknowables one encounters in trying to deduce too much in advance .
The Mets have played what we’ll call fifty-nine discrete seasons, counting the two 1981s individually. Twenty-three of those seasons incorporated five starting pitchers in the first five games. No off days or contingencies came into play, just one starter after another toeing rubbers and taking aim. It sounds so normal you’d think it was the norm. Recently it kind of has been, with seven of the past twelve fifth games being started by a fifth starter (if not “the fifth starter”). Just last year, which was abnormal as a year could be in every other respect, David Peterson made his season and career debut in 2020’s fifth game, following Messrs. deGrom, Matz, Porcello and Wacha, and that was without making the delayed Opening Day roster. With respect to his current battle with Yamamoto and Lucchesi, you can say Peterson should have an edge from being a seasoned fifth starter already.
Jason Vargas was 2019’s 5/5 man, though if you called him that in the clubhouse, he might threaten to “knock you the fuck out, bro .” Dillon Gee was the fifth starter in the fifth game of 2012 and 2015. In between, in 2014, he was the Opening Day starter, leaving Gee to ponder status whiplash. Hopefully Jay Horwitz got him in touch with Craig Swan, whose Opening Day starts in 1979 and 1980 were bracketed by fifth-game first starts in 1978 and 1982.
Two of your odder-appearing chronological fifth starters to start fifth games were pitchers you automatically associate with starting much sooner. Both cases speak to how injuries and the healing they necessitate can rearrange best-laid plans. In 1992, Dwight Gooden, who started eight Openers as a Met, was coming back from arthroscopic shoulder surgery and needed the extra few days to physically ready himself. Al Leiter, who threw the first Met pitch of 1999, 2001 and 2002 , got pushed back after being hit in the head by a line drive toward the end of Spring Training 2004 (during a game in which the Mets and Marlins combined for 34 runs, so we can assume there were lots of line drives). Instead of being that season’s third starter as slated, he waited in line behind not only T#m Gl@v!ne and Steve Trachsel, but emergency starter Dan Wheeler (in for late scratch Scott Erickson) and rookie Tyler Yates. I don’t know about you, but I’d take Leiter ahead of both of those guys. Ahead of Gl@v!ne and Trachsel, too.
Bill Denehy was the first Mets fifth starter to start a fifth game, in 1967. The righty would take seven more starts at Wes Westrum’s discretion before being sent as compensation to the Washington Senators for Westrum’s full-time successor, five-man rotation evangelizer Gil Hodges in the following offseason. Hodges himself used a chronological fifth starter, Don Cardwell, in the fifth Mets game of 1969, a year without a built-in off day following Opening Day.
Given the opportunity to get extra starts out of Tom Seaver early in the 1968, 1970 and 1971 seasons, Hodges didn’t hesitate to push back everybody in sight. The Mets didn’t use a fifth starter in 1968 until their tenth game, when a doubleheader provided a spot for Al Jackson. In 1971, it was classic swingman  Ray Sadecki getting his first starting shot also in the tenth game. Cast into a similar “not so fast…” role, fireballer Nolan Ryan had to wait until the ninth game of 1970, April 18, to be the Mets’ fifth starter. Was it worth the wait? Ryan pitched a one-hitter and struck out fifteen, the latter figure poised to stand as the single-game franchise record for fewer than a hundred hours. On April 22, Seaver famously struck out nineteen San Diego Padres, including the last ten in a row. Less famous, but eye-popping to our modern pupils is that the Padre game was the Mets’ thirteenth of the 1970 season, and Seaver was already making his fourth start. (I dare Rojas to try that with deGrom.)
The first fifth starter in Mets history, which is to say the fifth starter the Mets ever used, was Bob Miller…Bob L. Miller, to be exact. Casey Stengel waited until the Mets’ eighth game to get Bob L. Miller into the mix. They were 0-7 and when the manager turned to the righty, after which they were 0-8 and the righty was 0-1. The Mets didn’t have lefty Bob G. Miller to confuse matters until May. Bob G. Miller made no starts. Bob L. Miller made 21 starts in all, fourth-most among the 1962 Mets. He lost that first start and twelve games total that year, neglecting to win until the inaugural season’s final weekend, allowing him to take a 1-12 record to his next stop of Los Angeles, where the now-Dodger won ten games in 1963, thus giving Bob L. Miller every reason to tell the Mets, “It wasn’t me, it was you.”
Sometimes the pitcher who serves as your fifth starter of a given year is a gem yearning to be noticed.
• The 1997 Mets, at 1-3 on their pauseless opening West Coast swing, had nothing to lose when Bobby Valentine handed the ball to journeyman righty Rick Reed. Reed pitched seven scoreless innings and proceeded to install himself as invaluable to the starting rotation for a half-decade to come.
• The second-season Mets of August 1981 were six games deep into their reactivated schedule when they turned for fifth-starter purposes to a 27-year-old longtime minor league submariner who’d been promoted only after the strike. Thus began in earnest the major league career of Terry Leach, who’d rescue the Mets basically every fifth day six years later.
• The fifth starter of 1995 had to wait ten games for his first chance. What did Dave Mlicki do against the Reds on May 6 of this strike-delayed year? He did fine (though his bullpen blew a huge lead), but we don’t recognize Mlicki for how he started in 1995. We recognize Mlicki for how he started one night in 1997, throwing a shutout in the very first Subway Series game, and we don’t let him buy his own drinks for that very reason.
• Fifth starter Harry Parker didn’t get a look from Yogi Berra until the seventh game of 1973. Though his Met future would wind up in the bullpen, it was a good short-term future to begin, as Harry that day won the first of his eight games for the eventual National League champs.
• Unexpectedly deprived of Bill Pulsipher’s services, the 1996 Mets fished around for another starter before coming up with righty Mark Clark of Cleveland on the eve of the new season. The suddenly acquired fifth starter, used to start the campaign’s fifth game, Clark became Dallas Green’s most dependable arm from any direction in ’96, winning fourteen games and recording an ERA under 3.50, the only Met starter to keep his leading indicator so low.
• Mark Bomback had to wait until the dozenth game of 1980 to become Joe Torre’s fifth starter. When 1980 was over, Mark Bomback stood as Joe Torre’s only double-digit winner.
On the other hand, Bill Latham was the fifth starter used in 1985 (the sixth game); he’d have one win on the year and be gone by the next year. Aaron Laffey was the fifth starter used in 2013 (the seventh game); he’d have no decisions and be gone by the next month. Not every fifth starter story is uplifting. But at least Latham and Laffey were chosen before too long. You know who wasn’t? For that answer, we go to the extreme end of the “they won’t need a fifth starter until…” spectrum. It was 1975, the first of six times to date that the Mets went more than ten games before deciding they needed a fifth starter. Berra, as would anybody who could, leaned heavily on Seaver, Jon Matlack and Jerry Koosman. The power trio took fifteen of the season’s first eighteen starts. The three others went to rookie Randy Tate.
Not until the nineteenth game of the year did Yogi opt for a fifth starter, Hank Webb. Righty Webb was regularly talked up, certainly by unbiased Met source Bob Murphy, as a comer in the Seaver mode. His future proved a little more Tate-ish. Tate’s only season in the majors was ’75. Webb at least had sipped coffee during the three previous campaigns, but he didn’t make the Mets to stay until he was given the ball on May 3, 1975, or twenty-five days after Seaver got the season going. Hank etched versus the Expos what we now reflexively refer to as a quality start: seven innings pitched, two earned runs allowed. Pretty good for someone who wasn’t Tom Seaver. Not good enough for a win that day (Woodie Fryman one-hit the Mets), but good enough for Berra to remember Webb’s name. Yogi gave him seven starts total. Yogi’s successor, Roy McMillan, gave Hank eight more. The composite result was a respectable 7-6 record. Alas, under Joe Frazier, Webb was stashed in the bullpen before being sent back to Tidewater. The year after that, Webb was a Dodger for a spell, a Triple-A Albuquerque Duke for much longer.
Let’s go out on a more hopeful note, even if the 1987 Mets didn’t think they’d need hope. The defending world champions were planning to pick up where the 1986 Mets left off, particularly in their rotation. The 1986 Mets had starting pitching so strong that you’ll remember from above Rick Aguilera was deemed unneeded to start in the 1986 postseason. Ah, but just when you’re sure of something, something else come along. On April 1, 1987, less than a week before Opening Day, it was Dwight Gooden’s positive drug test. No, the fates were not kind to the 1987 Mets (save for the opportunity to go 11-1 fate furnished Leach), but at least the schedule gave them a break. Davey Johnson didn’t need a fifth starter until the eighteenth game of the year, and when he finally needed a fifth starter behind Ojeda, Darling, Fernandez and Aggie, he was able to call on merely the steal of the Spring, righthander David Cone. We got Coney in late March for Ed Hearn, and while we’d adored Hearn’s backing up of Gary Carter in 1986, we were told we were receiving in exchange for a caddying catcher a genuine prospect who’d make the pitching-rich Mets pitching-richer.
In his first major league start, against the Astros on April 27, 1987, young Cone lasted five innings and was charged with ten runs, seven earned. Hank Webb could have given Davey a better showing that night…and by then Hank hadn’t pitched professionally since 1979 (for the Miami Amigos of the Inter-American League, where his manager was the very same Johnson). Our hubris kept getting the best of us in 1987. That and everything else. It was a little much to expect this Cone kid to come out of almost nowhere and be the second coming of Gooden, Darling or anybody else.
Of course we know that one start every nineteen games wasn’t Cone’s destiny. By September, he’d be in the rotation full-time. In 1988, he’d win twenty games and be on his way to all kinds of accolades, not to mention a career that would carry into the twenty-first century, when much has changed about pitching. Yet we still check to see how long our team can go without turning to a fifth starter, even when we understand that someday that fifth starter might turn into David Cone.