As the night of May 28 drew to a close, future Hall of Famer Mariano Rivera stepped onto the mound for the Yankees at Citi Field as one of baseball's true rarities: The perfect closer.

Handed a 1-0 lead, Rivera was called on to do what he has done best for nearly two decades: Get the final three outs of a close game. It's something he's done 626 times before in his regular-season career, and 18 straight times in 2013 without fail. Rivera's reliability is legendary. It's also the exception that proves the rule.

Relievers are a volatile bunch, and among them, closers tend to burn out or fade away with troubling regularity. The history of the sport is littered with names that, for a few brief seasons, dominated the final innings, only to vanish in the blink of an eye: Eric Gagne, Bobby Thigpen, John Rocker. For every Rivera, there are plenty of relievers who turn in incredible seasons, then fall apart without warning.

Take a pair of last season's bigger surprises: Baltimore's Jim Johnson and Tampa's Fernando Rodney. Johnson and Rodney finished first and second in saves last season, with 51 and 48, respectively. Johnson, who began his career as a nondescript middle reliever, took over as the Orioles' closer in 2011 after the team dumped veteran Kevin Gregg. In 2012, Johnson became arguably baseball's most dependable reliever: Just 19 earned runs, 15 walks, and three home runs allowed in 68.2 innings. He was a big part of why Baltimore went an incredible 29-9 in one-run games.

Rodney, meanwhile, was a cast-off reliever with a big arm who was handed the closer's role in Tampa more or less because there were no better options. He responded with one of the greatest statistical seasons in MLB history: A 0.60 ERA in 74.2 innings, a new Major League record for ERA, to go with 76 strikeouts and just two homers allowed.

Both men are flaming out in 2013, with ERAs well over 5, along with several blown saves. Rodney has walked nearly a batter an inning; Johnson has already given up as many home runs in two months as he did all of last year. Though both have the support of their respective managers, it's up in the air whether either will be able to turn things around.

Brilliant closers seem to appear out of nowhere. Three years ago, Jason Grilli was a mop-up reliever who looked to be at the end of his career after a serious knee injury. Now he's the National League's top closer, with 21 saves and an incredible 37 strikeouts in 23.2 innings for Pittsburgh. Tom Wilhelmsen retired from baseball in 2005, then spent the next four years as a bartender in Arizona. In 2010, he decided to give the sport a second shot and signed a Minor League contract with Seattle. Last year, he saved 29 games with 87 strikeouts in 79.1 innings.

Teams drive themselves crazy trying to find closers, pay them extravagantly at the first sign of success, and stick with them past effectiveness. Why? How does a position that seemingly anyone can do — and seemingly anyone can fail at, no matter how much past success they've had — have such a hold on general managers and managers alike?

Conventional thinking in baseball is that getting a save — pitching in that final inning when a team has such a thin margin for error — is a skill that only a few pitchers possess. As the usage of relief pitchers evolved after World War II, managers began to designate the ninth inning to their best reliever, believing it to be the most important inning of the game. Modern statistical studies have called into question just how smart it is to limit your most effective bullpen arm to a single inning regardless of context. But a number of teams still stick to the idea that there's something about the ninth inning that requires a mindset most pitchers don't have.

Thus, you get teams shelling out big bucks or prospects for any pitcher who's demonstrated even a modicum of ability to close games. The Marlins gave Heath Bell a three-year, $27-million contract in 2012 after he saved 134 games for the Padres in five seasons. Bell had such an awful year that he was unceremoniously sent to Arizona in a salary dump after just one season in Miami. In 2006, the Blue Jays gave B.J. Ryan nearly $50 million over five years after he turned in a 36-save season for Baltimore. For that, Toronto got just two good years before injuries ended Ryan's career. The Red Sox, burned by acquiring Oakland's oft-injured closer, Andrew Bailey, before the 2012 season, picked up the Pirates' closer, Joel Hanrahan, before 2013. He lasted all of a month before an elbow injury ended his year.

Teams are constantly in search of the next Rivera, the endlessly durable and totally reliable closer who never falters or fails. But relievers fail and fall away. And trying to judge them on the smallest of sample sizes — 60 or so innings a year — ultimately leads to some expensive mistakes.

And even the most rock-solid closers fail. Let's return briefly to the evening of May 28. Rivera walked off the Citi Field mound, head down, as the Mets celebrated behind him after collecting three straight hits to grab a 2-1 walkoff win. Baseball's best reliever had blown his first save of the year — proof positive that there's no such thing as perfect when it comes to closers.