[joefromchicago]

I'll start things off by addressing some of RayFowler's points:

What is pitcher stamina? This is a statistic that essentially controls how many pitches a player can throw before he gets "tired" and have to be pulled from the game to avoid injury or getting excessively shelled.

I could be wrong, but I think that stamina measures two different things in OOTP: (1) how many pitches a pitcher can throw before he gets tired (in-game stamina); and (2) how fast a pitcher recovers after pitching (between-game stamina). The manual states that stamina "is a measure of how many pitches a pitcher can throw before tiring." The manual also describes the league "pitcher stamina" setting in the Stats & AI page as both "a measure of how quickly pitchers tire in your league" and "a league-wide adjustment of how quickly pitchers tend to become fatigued." Seriously, the manual lists "pitcher stamina" twice and describes it in two different ways. So there's some ambiguity here as to what "stamina" actually is, but it appears that the individual ratings and the league setting together control both in-game stamina and between-game stamina.

What is the point of pitcher stamina? I'm not the dev, but I suspect that the stamina stat is a brute-force method to prevent players from cheesing the game running a starting rotation full of closers.

That's certainly part of it, although there are two main determinants for a pitcher being a starter in OOTP: (1) stamina; and (2) number of pitches. A pitcher needs at least three pitches in order to be a starter (except for knuckle-ballers, who can get by with two). So even if you have a pitching staff composed entirely of pitchers with top-rated staminas, you still can't make them into starters unless they also have at least three pitches.

Why is it a bad design flaw? Well, it doesn't mirror reality very well. While some pitchers in reality obviously have more stamina than others, the majority of pitchers that you normally think of as a "bullpen" specialist started games regularly before coming up to the majors.

I've had a lot to say about the way that OOTP assigns stamina ratings to pitchers in the pre-reliever era who started but who didn't log a lot of innings. In short, my conclusion is that OOTP punishes those pitchers in two ways: by giving them bad pitching ratings and by giving them low stamina ratings. That, in effect, punishes them twice for being bad pitchers. But it's not that they were bad because they didn't pitch a lot, it's that they didn't pitch a lot because they were bad. In other words, their lack of skill is not necessarily correlated to any lack of stamina. So OOTP can definitely improve in this respect.

How is this a problem? Well, the main issue is that pitcher stamina is derived from usage data (how their team chose to play them) but is treated in the game as an unchangeable physical measure (like speed or fielding ability). This means that if you are playing a historical sim and a real-world manager decided to move Dennis Eckersley to the bullpen, OOTP requires you to make the same decision as well (because Eckersley's stamina drops like a rock the next season).

Despite my problems with OOTP's handling of pitcher ratings in the pre-reliever era, I will diverge somewhat from RayFowler on this point. As I see it, his focus is on the era from about 1960 to around 1990. That was a transitional period where a definite split developed between starters and relievers. OOTP rates pitchers so that the game will produce realistic results. That means rating relievers so that they relieve and starters so that they start. So if you're playing in this period, you just have to accept that some pitchers are going to be relievers and some are going to be starters. If you want to change that, then you have to accept that you're not playing historical any more, you're playing a fictionalized version of history.

I think RayFowler's gripe here is that he can't use pitchers in ways that they weren't used historically. I think that's a fair point, but it's a problem that's largely on the margins. As I see it, OOTP does a bad job of assigning stamina ratings to starters who didn't pitch a lot of innings. But for pitchers who spent most of their time in the bullpen, there has to be a mechanism in place to prevent them from being used in unrealistic ways. For OOTP, that mechanism is, in part, the stamina value that each pitcher is assigned.

Granted, that may be a "brute-force" way of achieving the desired result, and the gamer is necessarily constrained by decisions that were made by the on-field personnel at the time. So yes, Dennis Eckersley goes from being a high-stamina starter to a low-stamina reliever because of a decision made by Tony LaRussa and not because Eckersley suddenly became too frail to start. But OOTP is trying to get as close to being historically accurate as possible, and if you want Eckersley to remain a starter, then you have to acknowledge that you're no longer playing a historical simulation. That's not a problem with the way OOTP handles stamina ratings, that's a problem with the gamer wanting to play both historical and fictional at the same time.