I’ve set up only two tables this time, like a busboy getting weary at the end of a shift, one of pitchers and the other of batters, no ages listed, though if you really care about their precise ages, you can easily figure that out.  The tables below are copied from Bbref, which can be reordered (as I’ve done here)   to yield players in groups of age, but within each age-group, the players are arranged alphabetically, not chronologically.  So if you start with the final alphabetical grouping (for pitchers, that would be Adrian Devine through Pat Zachry), that will be all the 24-year olds. The previous alphabetical grouping will be the 23-year-olds, and so on. Each list below is, again, of all the players in MLB under 25 years old who got 1.0 WAR or more, this time in 1976.

I don’t see any reason to expect that there would be a significant distinction between leagues, or see any pattern other than random fluctuations between the league totals, but if you want to make the case that such a pattern exists, it’s not difficult to sort them back into AL and NL.  Go at it, and argue that the differences are important, and why you think those differences would show up.  Pitching is pitching, and hitting is hitting, and those are two categories I’m contrasting here. It’s especially insignificant in the career WAR numbers, of course, because you’ve got players like Dave Winfield and Goose Gossage spending large chunks of their careers in each league.

If you feel that I’ve been arbitrary in establishing a 1.0 WAR minimum for inclusion, you’re correct. I’ve been arbitrary, and I’ve been lazy: mostly I set it up that way to limit the amount of work I’d need to do and still arrive at a meaningful number of players to study.  I’ll defend my laziness, however, by repeating my original rationale: I wanted to study players who had put up MLB stats that approach significance, which will not be accomplished if we count every rookie who plays in three games. Below a certain point, I’d be sorting through a lot of noise rather than signal, but you’re right, 1.0 WAR is a totally arbitrary cutoff. I’d consider it a service if someone would repeat each of these studies with all of the under-25 players who earned, say, 0.7-0.9 WAR,  or any positive WAR, or any WAR at all, and let us know if that study contradicts my conclusions.  I think you’ll just be dealing with loads more of raw data, and less meaningful data, but if you think that would be the way to go, have at it.  Or, for that matter, do a study limiting the WAR to 5.0 or more, information you can get from my tables. I don’t think the cutoff point should change the overall conclusion, but if you think so, gopher it.

Another possibility is that I’ve made a mistake by setting a minimum WAR for inclusion in this study but no maximum: maybe if a young player, at whatever age, performs above a certain level of skill, as measured by a certain WAR, 4.0 WAR or 6.0 WAR or WARever, his team should consider him like an established veteran, and he should be excluded from this study. Maybe. But it should be easy enough to segregate young players above and below that WAR-point, if we think later that the distinction is valid. For now, I’ll just keep my mind open to that possibility. For every Roger Clemens here with an MVP-type WAR, there’s a Mark Fidrych with an MVP-type WAR—I don’t think there’s much to be gained by putting a cap on WAR here.

Same with age. I’m using players under 25 because that was where I got this idea, all the TV commentators telling us over and over how the Cubs have achieved greatness by playing six regulars under the age of 25.  If you think that 28 would be a more significant age cutoff, do a study of players 25 to 27 in these years, and I’ll be glad to fold your study into mine. I don’t expect it will change the conclusions very much, if at all, but let’s see. If you raise it too much, of course, you’re no longer studying young promising players with great potential, you’re simply studying players.  The idea here is that by studying young players who have shown some signs, but not overwhelming signs, of MLB talent,  you’ve got a body of players whose teams could A) decide to commit seriously to that player (signing him to a long-term, big-bucks deal extending into his first year or two of Free Agency, or just by clearing out a position for him), or B) decide to gamble that the best use of him is to swap him out for someone else.

The Mets, to pick an example offhand from a roster I’m familiar with, have a young outfielder right now, Michael Conforto, age 23, who has played parts of the last two seasons, and has looked good at times, not so good at other times.  I don’t believe the Mets feel with any sense of certainty that Conforto is a future star or that they feel he is a future washout. They just don’t know—he could go either way.  I pick Conforto because the airwaves are featuring a lively argument just now about whether Conforto’s a keeper or not. Likewise with a slightly older pitcher named Steven Matz, who turned 25 years old last season, also has played parts of 2015 and 2016, also has looked great at times, not so great other times. Matz also is the subject of discussions right this second about keeping him or trading him (some talking head on TV last night was talking-heading about including him a package for Chris Sale). Again, the Mets haven’t seen quite enough of Matz or of Conforto to have any sense of certainty if they’re sitting on a gold mine or a land mine here. 

What my thesis-in-progress argues is not that the Mets need to hang onto Conforto for dear life or to dump Matz asap for whatever they can get for him, but simply that there (may be?) is a sound basis to keep young batters longer and trade young pitchers sooner than they’d normally be inclined to. Specifically, to trade a young pitcher at his height of peak performance, if you can find a trading partner in desperate need of pitching who is less scared than you are of the dangers of young pitchers, might be smart. And certainly if someone approaches me with a good young pitcher, seeking a position player from me, maybe I need to be skeptical as hell about pulling the trigger.

The exciting part of this thesis-in-progress, for me, doesn’t even involve this thesis specifically. The exciting part is the concept that there are undiscovered market inequities out there.  I’m sure there are, and I’m equally sure that some team will discover one, and use it to their advantage until every other team has glommed onto it, putting it into practice all through MLB.

It’s pretty unlikely that I’ve glommed onto a major undiscovered market inequity in the past few days. It would be very cool, but I doubt it’s true.  More likely, some team (or all teams) have worked out a formula like mine years ago, and this whole idea has been factored into every MLB trade and contract and personnel decision since 1978. Having zero access to the thinking of MLB GMs, I can’t say.  I can tell you that I do have access, here on BJOL and listening to talk-radio and sports shows on TV and chatting with fans IRL (are these abbreviations bugging you, btw? They seem fine to me), and they’re all violently divided on the issue of trading, or hanging onto, young players of both types, about equally. There’s a lot of shouting, and has always been a lot of shouting, about the wisdom or the stupidity of trading young pitching. Certainly no non-professional is including in his argument for keeping or trading some young player the "fact" that pitchers are quantitatively and clearly more likely than position players are to break down physically.

Whether that "fact" is true, or just BSing on my part, whether Young Pitchers’ll Really Break Your Heart or if that’s just fans whining, is what I’m trying to find out here, so let us move on to the content section of this article, and see if we get any closer to an answer:

Let’s see if we can get all three seasons, separated by a decade, summarized into a single chart. (This is fun, incidentally, writing the article as I’m doing the research. Normally, a sane person would mine the data first and then present his thought-out conclusions. I’m sort of going the other way around here, and fully prepared to do an Emily Litella on you.)

Adding and dividing these numbers (not my strongest suit, but I hope I’m doing ok here), the rough collective average of the three single-seasons’ WAR for pitchers is a little under 8  (2.9 + 2.6 + 2.3) and so is the batters’ WAR (2.6 + 2.4 + 2.8). I get 7.8 for each—noting that these figures have been rounded off to one decimal place from ten decimal places several times, so they’re only approximate, plus the year that has 27 pitchers should be weighed more heavily than the years with 21 pitchers, etc., but the differences I’m about to describe are so gross that a decimal error or a weighting error hardly matters here. The career WAR total averages for the pitchers are somewhere in the high teens, 17.5 or so, while the batters’ career WAR totals work out to something around 30. 

Now I’m being a mathematical slob here, a perfect disgrace to the art of sabermetrics, but I doubt all my errors added together amount to more than a single integer: in this study of 72 young pitchers and 84  young batters in 1976, 1986 and 1996, I’m comfortable positing that the pitchers and the batters had very similar WAR totals for the seasons in question, but over the course of their careers, the pitchers had WAR totals no more than two-thirds of the batters’ WAR totals.

If my math is right (which is possible) or even if it’s close (which is probable) AND if these three seasons are not outlying freaks AND if WAR is a valid measurement of players’ abilities AND if there’s nothing fundamentally wrong with my method of testing my premises AND if my premises have been intelligently chosen, THEN we may be onto something here, the approximate measurement of how much more young pitchers will break your heart than young batters will: about 50% more.