At some point this week, we’re probably going to get a new Collective Bargaining Agreement. The current one expires on Thursday, and given MLB’s current revenue situation, neither side can be all that incentivized to screw things up right now. Both the owners and the players are getting very rich off of baseball’s ability to sell their television rights, and to risk that kind of cash cow over something like the qualifying offer or an international draft would be the definition of cutting off your nose to spite your face. There may be hurdles yet to clear, but they’re very likely going to be cleared.

A seemingly settled issue in negotiations is the coming expansion of the active roster. According to Ken Rosenthal, owners have agreed to expand the April-August roster limit to 26 players in exchange for a reduction in the September limit, which will come down from 40 to 28 or 29. This will deliver a more consistent brand of baseball throughout the year, rather than having the final month played under very different rules than the rest of the season.

The concern is that expanding the roster will just give Major League managers the green light to carry yet another relief pitcher. We’re already seeing more pitchers used per game than ever before, and with another available bullpen arm, there would be even more opportunities to pull the starter early. And while there are plenty of valid strategic reasons to prefer relievers over tiring starters, from an aesthetic standpoint, the game’s march towards more pitchers throwing fewer pitches probably isn’t something baseball should be trying to accelerate.

So, how does the league rectify their willingness to add a 26th active rostered player with a desire to cut down on the number of pitching changes each game? Well, there’s one surefire way to make sure teams don’t use the extra roster spot to expand their bullpens; make a rule prohibiting it.

Right now, teams can do whatever they want with their 25-man rosters. Pretty much everyone defaults to the basic construct of five starting pitchers, seven or eight relievers, and 12 or 13 position players, though the fifth starter does go away sometimes when off days allow for skipped starts, and then teams can carry both eight relievers and 13 position players. By adding a 26th roster spot, the fear is that teams would just default to eight relievers all the time, and expand up to nine when they don’t need their fifth starter. But there’s no real reason why MLB can’t also just pass a rule limiting the number of pitchers a team can carry at a given time.

You could write the rule a number of different ways. For instance, you could require that teams must have 13 (or 14, if you want to get crazy) position players active for every game, which would effectively limit the size of a pitching staff to no more than 12 or 13 pitchers. Teams would still be able to carry their normal seven or eighth relievers with five active starters, or get a ninth reliever when they option out their fifth starter, so it would essentially codify the current pitcher usage as a maximum, and force teams to use the new roster spot on another hitter, or at least an extra runner/defender type.

If MLB wanted to be a little less draconian in forcing teams to use the extra roster spot in a specific way, they could simply target relievers more specifically, and say that teams simply can’t have more than eight relievers on the roster at a given time. This would, however, force the league to create some kind of differentiation between “starter” and “reliever” in the rules, which could create some room for loopholes, or might have some unintended consequences when it comes to swing guys who both start and relieve. It’s far easier to differentiate between hitters and pitchers, and as such, targeting relievers that specifically might be less desirable.

By codifying the rule as requiring a minimum of 13 hitters — though I’d probably argue for 14, honestly — you’d not only avoid expanding bullpen usage through this new rule, but then the 26th man would actually help combat the decline in offense brought about by specialization. Part of the reason for bullpen effectiveness isn’t just that relievers are throwing harder than ever, but that the growth of bullpens means that there are fewer roster spots for bench players designed to combat bullpen usage, so as we get more relievers, relievers also get an advantage by having fewer deterrents to their own use.

Because teams are carrying so few position players, they aren’t as comfortable removing their platoon-disadvantaged starters from the game to counter the entrance of a specialist reliever, and they don’t have a one dimensional slugger on the bench available to make that a bad match-up for the pitcher anyway. If MLB simply writes the rules in a way that forces MLB teams to use the roster spot on a position player, teams will either get to add that bat-first kind of player to use as a pinch-hitter once a game, or at least add a depth piece behind a starter that might make them more willing to remove a starter who is less likely to get a hit in a high leverage situation. Either of those uses of the extra roster spot will help push back against the bullpen revolution.

Major League teams have been expanding their usage of relievers because the data shows that it works, and teams want to win. But if we give teams another hitter to use, the match-up sidearm guys become a little bit less effective, and it’s those one-batter pitching changes that the league should be most wanting to avoid. And by forcing them to carry a player who makes those specific match-up options less appealing, the extra roster spot could actually serve as a deterrent to bullpen usage.

In all this talk of the 26th roster spot, the assumption has been that teams will just add another arm to the pile. But the MLBPA shouldn’t care too much what kind of player the new spot goes to, and so the league should simply ask that the expanded roster also come with a requirement that forces teams to use that spot on a position player.