OK, look: Once Terry Collins said he was going to use Frank Francisco as his closer before last night’s game, he practically had to use Frank Francisco as his closer in last night’s game. He doesn’t owe Francisco much after the reliever’s performance over the weekend, but a big part of a manager’s job is keeping his players happy, and rolling back on his word with a pitcher with shaken confidence would not be the best way to do that. Plus, pretty much everything in Francisco’s history — not to mention his contract — suggests he’s a hell of a lot better than he has been for the Mets in 2012, so getting him straightened out should be a priority for the team if it aims to have a fully functioning bullpen later in the season.

But saves are stupid, and managing to the stat — as is typical in Major League bullpen usage — is ludicrous. Presumably you know this. Everyone knows this. And yet it still happens everywhere with no signs of stopping, and for all the vitriol spilled over stats that actually correlate to meaningful baseball events. there seem to be shockingly few outbursts about this nagging and ubiquitous silly habit.

Here’s the deal, in case you’ve missed it: Since the dawn of the one-inning closer, teams maintain their ninth-inning leads at exactly the same rate they did before the dawn of the one-inning closer.

The argument for defined bullpen roles is that relievers like to know the jobs they’ll be asked to do when they get to the ballpark. And that makes some sense. If I showed up to work one day and my boss said, “hey Ted, instead of being an editor today, we need you to operate the lights at the studio,” I’d get all worked up, plus I’d probably suck at it. I have no idea how to do that.

But then getting guys out in the ninth inning doesn’t seem like an appreciably different task than getting guys out in the seventh inning, it just happens at a different time. Certainly there are some differences in mental and physical preparation. But maybe if instead of telling a reliever his responsibility is the eighth inning or the ninth inning when his team is ahead, the team told him his responsibility is to pitch in relief if necessary on Wednesdays, Fridays and Sundays, he might still feel comfortable that he understands his task. I don’t know; I’ve never been a Major League reliever.

There’s a money thing too, of course. Closers get paid lots of money, so presumably relievers want closers to exist so they can earn the role and get the big bucks. But it seems likely that if some better stat than saves came into widespread use for assessing relievers, it would soon be used to compensate relievers, and with the amount of money in baseball there’d no doubt be plenty of handsomely paid bullpen arms regardless.

Point is, if the Mets weren’t committed to having a closer to aggregate saves — as basically all teams are — and committed specifically to using Frank Francisco as their closer, they almost certainly would have left their best reliever, Bobby Parnell. in last night’s game to face the middle of the Brewers’ order in the ninth after he retired the two batters he faced in the eighth on only five pitches.

But since it was important to use Francisco as the closer and give him the save opportunity because he’s the closer, they wound up not only using their third reliever of the night for the ninth, but needing to warm up their fourth — Jon Rauch — when Francisco struggled. And from here, it seems hard to figure how whatever advantage gained from having relievers know their roles isn’t more than mitigated by that type of forced inefficiency in their usage.