I started this thinking it was going to be a big deal. I’m now convinced it’s important to do even though it’s a small deal. In my last post, I was discussing Justin Verlander‘s quest for 300 wins and it got me thinking about the 5 inning rule for starters. Let me start by saying that the Win was already a pretty stupid stat — probably the single worst headline stat that baseball has. It’s even worse than Game Winning RBI and they junked that one years ago.

Wins are just about the most team-dependent individual stat we have. They are not only team-dependent, they are absurdly context-dependent, particularly for relievers. If there were a proposal to so with pitcher Wins what we did with GWRBI, I’d gladly sign on. But there is zero chance of that happening. None at all. And while people now have stopped talking about wins for relievers as an index of reliever quality, it is still used for starting pitchers. And if Chip Caray can be relied on for anything, the third out of the fifth inning will terminate a running commentary that usually started about two innings earlier.

But the starting pitcher Win, as we all know, is endangered. I’m not going to dwell on this, because we all know it. Twenty game winning seasons are now rare, Pitchers don’t pitch as many innings. I’ve got graphs to show all this, and I’ll provide them if anyone is interested, but I think everyone already knows it.

So, if we’re not going to kill the Win as a stat, maybe we should expand the definition to include pitching stints by starters of less than 5 innings. I originally did this to see if it would help Justin Verlander, but what I was really surprised to see was just how little change it made. I have recalculated pitching wins for every game in the Retrosheet database on the assumption that the five inning rule is kaput: any pitcher who exits a game with a lead that his team never relinquishes will be credited with a win. Does it make any difference?

The Good Old Days

Let’s start with the old guys. The rule affects them almost not all because they didn’t leave games they were leading in less than 5 innings. Take Walter Johnson. He never got a cheap win as a starting pitcher, and would in fact lose one win for a game in which he relieved a guy who left early. Among others with no cheap wins: Christy Mathewson, Steve Carlton and Gaylord Perry. Phil Niekro had one. Don Sutton had three, as did Tom Glavine. Lefty Grove had four, but he would also lose 5 relief wins. Unless you’re somehow wedded to Roger Clemens‘ 354 wins as a number you’re memorized and cherished, changing it to 356 isn’t going to change your mind about him. Of the old guys, the one with the biggest move is Tommy John, moving up 7 wins from 288 to 295 — still just a smidge short of 300, [PS: Why isn’t he in the Hall of Fame? Somebody remind me.]

The Modern Game

This inquiry started to see if the rule had hurt Justin Verlander. Nope. So far in his career, Justin Verlander has had one game where he came out early with a lead his team never relinquished. Once. It was this game, in which he left with the bases loaded after 4 2/3rds leading 8-6. The Astros ended up winning 9-7. Now you might say that Verlander doesn’t deserve this win. To that, I have two responses: first, that’s a stupid argument. But second, you know who got the win? Héctor Neris, the guy who gave up the other run. After his performance with the Braves last year, he deserves nothing.

And there simply aren’t a lot of pitchers highly affected by this rule at all. And it’s not clear why you should care. Three pitchers would have added 10 wins to their totals: Jameson Taillon, Gio Gonzalez and Jake Odorizzi. Ryne Stanek would have vultured 9 wins as an opener to go with his 21 meaningless wins as a reliever, but we’d have to subtract one win he got for inheriting a win from another starter. So that’s 29 wins instead of 20. Does that make you feel differently about Ryne Stanek?

Even if we ignore the guys stretching for aggregate win glory, it just doesn’t make much difference. In 2025, starting pitchers would have earned an extra 146 wins. The high water mark was 2021 with 175. By way of comparison, no year before 2000 had more than 60, and lots were lower than that. 2005 would have shifted 27 wins. But even in today’s 4-inning starting regime, we get a win change in only about 5 or 6 percent of games.

There is a question of whether this percentage would go higher if the rule were eliminated. We all know of managers who tried to drag the starter to get that critical third out in the fifth inning to preserve his personal chance at glory. But in an era of pitcher injury, my proposal obviously has a net benefit. Pitchers can get removed without risking injury if the situation calls for it. Frankly, given the problems in the game today with the overtaxing of relief staffs, the rule change would probably not impact anything at all. Adding innings of workload to your bullpen is the last thing managers want to so, and the win rule just gives them another hook to hang that decision on.

An Alternative

Just to point out: the real reason that wins are down isn’t attributable to this rule. It’s attributable to the fact that it’s hard for bullpens to hold leads. And maybe you think it’s unfair not to give a pitcher a win who pitches great for 7 innings, followed by a blown save that then eventually leads his team to a win.

If you really still think the rule is worth preserving because, say, you think openers who pitch an inning should never get a win even if their team scores a bunch of runs, then I can propose an alternate rule. Just let the official scorer decide who gets the win unencumbered by the 5 inning limitation for the starter, and/or unencumbered by the rule that leads have to be maintained to have a chance at a win. What’s the argument against that? My only objection, if I were still listening to him, is that it would give something for Chip to gab about for the entire game.