Not just later in the season. Any time a guy has appeared in >70% of his team's games, there's a fatigue penalty on top of any overuse fatigue and the 'recently-pitched' fatigue for guys who are close to being on track to use up their innings. In theory a guy can pitch in ~110 games, but even that is tough to get to at 90%+ unless the guy has a massive RL inning total.
I understand why you would want your lefty to always be the long guy, but I don't think it's going to work out that way. Even in the best-case scenario, if he pitches in 110 games, you'd need him to average 3 innings/game to reach his RL inning total. Sparky doesn't like to leave relievers in to bat, regardless of pull settings, unless you totally turn off pinch hitting for the guy in player settings. Even then he'll start to defer to, say, a right-on-right matchup if it gets into the later innings. Your only shot at getting close to that 3 inning AVERAGE is to completely give up a lot of strategic advantages, and then you're probably not ahead in spite of getting significant left-handed innings against your opponent's vR lineup. Particularly given that many managers use very similar, if not identical, lineups against both sides anyway.
In order to use up everyone's innings I think you're going to need to rotate guys in and out of the rotation. I would assume you can probably get something like 200 innings of long relief from one guy at a time listed as Long A over the course of a season - that might be a little optimistic, but I'm confident you can cruise past 150. You've got a total of 1140 innings between your 4 guys, so you're gonna want something like 950 innings from guys starting games if you can get there. That's about 6/game, so I'd give all of them something like 90/105 pitch counts as starters unless they are under 6 RL IP/game. Have one guy at long A with pitch counts of 50/60 or so. If you're willing to be particularly hands-on, I'd go in after every game and rotate the guy who just was in the Long A spot into the next-up starter spot. Assuming he pitched in relief, he should be xx(100) and ready to pitch after a 1 game rest. Then put whoever from the rotation was in that spot, who will likely be at or near 100% already, into the Long A. So that would look like
| Game |
|
|
|
|
|
|
| Current Starter (locked in) |
A |
B |
D |
C |
A |
... |
| next starter |
B |
D |
C |
A |
B |
… |
| 3rd starter |
C |
A |
B |
D |
C |
… |
| Long A |
D |
C |
A |
B |
D |
… |
In this way you'll always have a starter at 100% and a reliever at or near 100%; everybody gets 1 day of rest after every appearance and everyone gets equal numbers of starts. As the season goes along you may have to sometimes squeeze an extra start out of the guy with 330 innings (or, depending how things are going for the other guys, just increase his pitch counts by 5 or 10). If this sounds like too much work, you can always just set them all as starters and check the relief box next to each. This definitely doesn't seem to work as well - I don't think starters allowed to relieve are treated at all the same as Long As by Sparky, he seems much more likely to guys listed primarily as relievers in situations that would seem to call for the long guy. If the rest of your staff gets fatigued, Sparky will use your starters in relief, but that means you'll be living near a fatigue cliff with your bullpen all season. So I'd definitely prefer the hands-on approach, but I'll admit that much of the time I would still take the lazy way because I don't personally like having to do hands-on work on a daily basis, much less a game-by-game basis.