What I was going to tell you today was that it was time for a few teams to really mix things up with their approach to pitching staffs. It’s not my idea, originally, but it’s been a while since I’ve heard anything about it, and I was going to tell you that for some teams, its time had come: get away from the five-man rotation, and from the idea of “starting pitchers” altogether.

We know that even the best starters tend to get worse with each time through the batting order — that fatigue is a gradual effect, not a binary, suddenly-happens-at-100-pitches sort of thing — and that pitchers, essentially all of them, fare much better in efficiency terms as relievers asked to go one inning than as starters relied upon for six-plus.

We know that the “win,” the pitcher’s stat, is almost perfectly useless, almost as bad as the save, which itself is a little better than the hold, and that managing a game based on those numbers is just ridiculously backward thinking. So why not toss all that old-fashioned nonsense aside and endeavor to use the entire staff when it’s at its best (that is, when it’s fresh)?

Plan to use five or so pitchers every game, and no one gets more than three innings. Maybe your starter goes three, two more guys go two apiece, and then you’ve got a guy for the 8th and one for the 9th. Everyone stays fresh, everyone feels free to really let loose, and you’re getting the most out of the whole staff.

But I’m not going to argue for that, and there’s a simple reason why: it’s an incredibly dumb idea.

That’s not a slight against anyone but me, since I was the one who was about to argue it just now. And it would’ve been stupid. For one thing, players and some teams and arbitrators still really value wins and saves; for your first year, you’d have a staff of angry, bitter pitchers, and from year two forward, you’d have staffs of angry, bitter, less talented pitchers, as your free agents bailed and others’ refused to sign.

Second, it’d be a logistical nightmare. Those kinds of plans don’t even work well in Spring Training games, when it’s only a month long and nobody cares who wins or loses — guys get blown up too badly to get through an inning, get injured, etc. Over a whole season, when you’re actually trying to win, it’d be abandoned almost immediately.

Third, you lose some ability to play matchups, with more pitchers likely to be unavailable at any given time. No, the whole-pitching-staff-by-committee thing sounds pretty good in theory, and maybe it would work in a video game or something, but the more I think about it, the more I’m convinced it’d be a disaster in practice. It’s awfully dumb to manage with the idea of getting your starter the win or your closer the save, but there are really good reasons for sticking with more or less the current system that have nothing to do with that.

I suppose it happens in baseball just as everywhere else, doesn’t it? You get older, you gain experience, and you start to realize that most “revolutionary” ideas are revolutionary because stupid. If they were any good, somebody would be doing it by now. But they’re not any good, so no one is.

I’ve been into sabermetrics, or whatever you prefer to call it, for twelve years now, and I’ve been writing about this stuff on a regular basis for a bit more than three. That’s a lot of time to invent or be introduced to, tout, and then be embarrassed by the short-sightedness of different ideas. I remember once being convinced that teams were going to go back to four-man rotations; another time, that differences in defensive abilities were basically negligible, so teams were better off just putting the best offensive teams they could and forgetting about the defense.

Dumb, silly things to think, the kinds of things one thinks when one becomes convinced that he or she is smarter than everybody else. Sabermetric types as a whole are getting better at avoiding that sort of thinking, but there’s a long way to go.

“Relievers are fungible” is a big thing now, a serious overstatement based on the basically correct ideas that (a) relievers are fickle and unpredictable and (b) many of the best ones are paid more than they’re probably “worth” to their teams in terms of total wins added. It’s the same basic problem as with the staff-by-committee idea; the “everyone should build a bullpen from nothing like the Rays did” thing sounds fine, until you realize the Rays’ pen had the AL’s 10th-worst FIP and 13th-worst xFIP in 2011, and is off to a disastrous, 7.49 ERA start in 2012.

It’s a strategy that certainly can work, and it’s probably the only one that makes any sense for a team like the Rays, but it’s terribly risky, and if you’ve got the cash to buy yourself a little more certainty, you probably want to do that — even if it means overpaying for an arm or two.

That’s not to say that all revolutionary ideas are bad, or that the kind of thinking that leads to them is ever bad. The changes that do happen in the way teams are built and the game is analyzed — and there have been a ton over the last ten years or so, and will probably be as many over the next ten — happen because smart people are thinking just that way. There’s also a lot of traditional thinking that I hate and will keep hating: too many bunts and intentional walks, that ridiculous save-based use of closers, the inexcusable delay in taking advantage of instant replay, and so forth.

The problem is this: while revolutionary thinking is good and productive, the vast majority of ideas that come out of it are not. And most, but certainly not all, traditional ways of doing things are done that way for a pretty good reason. The round-table, twelve-men-with-no-starters pitching staff addresses all sorts of real issues, but creates all sorts of bigger problems. The real gains to be made are in the margins, small adjustments here and there, incremental improvements.

Don’t change the whole way a pitching staff is used, but maybe stretch that ace reliever a bit more, and maybe don’t send Jason Marquis back out after 97 pitches. Those types of changes may not be nearly as exciting as the big, sweeping kind, but they almost certainly make more sense. That realization sucks — especially as a writer who’s trying to come up with something interesting to write about — but it’s the right one, and it’s part of growing up. Revolutionary thinking good; actual revolution, at this point in the game, probably not.

(Unless we’re talking instant replay, in which case, Robots Now, viva la revolution, etc., etc. But that’s a talk for a different time.)