Core Dump: AAA vs. AA Differences

A frequent commenter, allfrank, (it's all relative, comment once or twice and you're frequent :^) asked a good question of me the other day: what's the difference between AA and AAA, "... AA doesn't seem all that much lower than AAA." (Sorry this is a late response, I've been busy with work and personal issues, plus wanted to give some more thought to this)

To start, I want to make it clear that I'm no expert on the minor leagues but I do read a lot so I will share what I know in the first of what I will call my "Core Dumps".

Obviously, there must be some some sort of difference, else why have the different leagues and advance the players from AA to AAA. But beyond that, what other differences are there?

Why They May Seem Similar As Well As Different

First, one thing I find that is indicative of the difference - and the similarities - between the leagues is the study I read that was done for Ron Shandler's Baseball Forecaster book (which I recommend to anyone interested in baseball analysis, though it is harder to justify any subsequent ones unless you are into playing baseball fantasy leagues, like I do; much cheaper via Amazon too).

For the similarities (my interpretation, BTW), they found that the success rate (they unfortunately did not define what "success" meant, so we'll have to take their word) for hitters was virtually the same whether or not they played a full season in AAA or not. If they had a full season, 56% performed well, 38% performed poorly, and 7% had a second half drop-off. If they had less than a full season in AAA (amassed at time of call-up), 57% performed well, 21% performed poorly, and 21% had a second half drop-off. Obviously, if they did not amass a full season of AAA, they they most probably have played most of their time in the lower levels, like AA or below, because most prospects never make the jump to AAA, let alone the majors, and most do not make the majors the next season, meaning most prospects spend more time at below AAA than in AAA.

For the difference, it was found that AAA experience was almost critical to a pitcher's success. For pitchers with a full season amassed under his belt in AAA before his call-up to the majors, 56% performed well, 33% performed poorly, and 10% had a 2nd half drop-off. However, for those with less than a full season in AAA, only 16% performed well, 77% performed poorly, and 6% had a 2nd half drop-off. What this means is that one year of amassed AAA experience improved the chances of success by over three and a half times!

Thus there must be something about the competition in AAA that enables pitchers to make the leap from prospects to MLB pitcher. Otherwise, the success rate would not be so much greater after a year of experience in AAA. I guess it could also be because of GM incompetence in judging AA talent but

The book also noted that if you are looking for a player to speculate on for fantasy baseball, which has the same goal as any fan, to find a good player, it notes that you should look for a player with at least two seasons of experience at AAA because career AA players are generally not good picks. Nothing earth shattering but just another piece of info on the difference between AAA and AA.

My Data

Through my work for another website, I got into the habit of collecting the stats for each of our minor league affiliate team's respective leagues: Pacific Coast League (AAA), Eastern League (AA), California League (A+), Southern League (or SAL or Sally; A-), and Northwest League (Short Season-A). So I decided to compile the season ending data (from a great source of minor league information, The Baseball Cube but it was mind-numbing work!) and find out where each of our prospects ranked within the league by a variety of parameters (overall, by age, etc.). I looked over the data for differences between the leagues in terms of the age of players and the following is what I found.

For hitters, the average age in the Eastern League (EL) is 25.1 years, whereas the average age in the Pacific Coast League (PCL) is 27.1 years. Basically, a hot prospect's shelf life is pretty much over at 27; according to the book, a player reaches his peak skill at 26, which means that it is either plateau or downhill after that. So a lot of the players in the PCL are basically what they call a AAAA-level player: not good enough for the majors but nothing much to prove in AAA. The Giants always seemed to get players like that (particularly 1B), with Todd Linden appearing to be the latest, but others included Brian Dallimore, Damon Minor, Rick Lancellotti, J.R. Phillips (was he related to Emo?), Randy Elliott.

And for good prospects, basically it is best if they are only 22 (according to Shandler's book) or younger by the time they reach AAA, only 21 when they reach AA. These are the "ideal ages for prospects at the particular level." Obviously, if he is younger and still is able to hold "his own against older and more experienced players", he is considered better as a prospect, while if he is older, his status as a prospect suffers.

The distribution for the hitters show this a bit more clearly. In the EL, most of the players are from age 23 to 27, while for the PCL, most are from 23 to 31. In the EL, 6% is 21 years (the ideal age for a good prospect in AA) or younger; in the PCL, 6% is 22 years or younger. In the EL, hitters 27 years and older (over the hill skill-wise, though experience could help performance, but it's a downhill fight) make up 24% of the players; in the PCL, hitters 27 years and older make up 51% (about half!) of the players. (To be clear, this is just a straight average of each player, it is not weighted by AB or IP)

For pitchers, the ages are similar to the hitters. In the EL, 25.2 years; in the PCL, 27.2 years. The distribution shows the bulk of the EL pitchers being from age 23 to 27, while for the PCL, the bulk being from 23 to 31. In the EL, 5% is 21 years and younger; in the PCL, 5% is 22 years and younger. In the EL, pitchers 27 years and above make up 27% of the players; in the PCL, pitchers 27 years and above make up 53% (again, about half are past their peak skill age point) of the players. Again, clearly, the same distribution pattern holds.

My Thoughts

Now, what do all this means. At any time, about 5% of the players in either league is considered a pretty good prospect, playing at or below the age appropriate for the league in order to be considered a pretty good prospect. However, only about 25% of the players are past their peak skill age in the EL, whereas over 50% of the players are past their peak skill age in the PCL.

So there's where you might see the difference between the leagues being not so different, allfrank, there are a lot of players past their skill peak playing in the PCL and not so much in the EL, so there could be times where the EL appears to be close to the PCL in terms of skills.

However, I think a Darwinian effect comes into play here as well. If it is clear that the player is over the hill at 27 (or older), then that player is probably not going to stay very long in the minors, he would be released or would retire on his own. So these older players, though past their skill peak age of 26, if they are to survive and produce at an older age, need to gain through experience the ability to improve their performance more than their peak skill has deteriorated.

In addition, while players who are able to improve stay around, the lesser players are dropped, but are not necessarily replaced with better players, just younger, less experienced players, which works to the experienced players who is improving advantage. They get better but the competition gets less experienced (though not necessarily less skilled, that depends on who is coming up and the circumstances of each MLB team's farm system. For example, the Giants have not had a really good crop of prospects for a long time until the past couple of years, so when they moved their AA players to AAA, there could be a wider gap between our prospects and, say, a Colorado, who has been drafting low for a number of years now and stocking their minors with better prospects, in general).

Thus, in some cases, like Brian Dallimore, the player improve by a lot so that they are suddenly beating up the players who are left because the better players would have moved up to the majors already plus there are a lot more of these players (over 50%) than there are the true prospects (around 5%) and new players are moving up constantly, all less experienced than he is, plus some encounter the Peter Principle, where you eventually get promoted to your level of incompetence, if I remember that right. Thus someone like him or Dave Doster, who couldn't hit over .280 when they are pre-peak, suddenly can knock the ball around to the tune of over .330, and in Dallimore's case, .352, because of all these factors benefiting them at the AAA level.

However, once they rise to the MLB level, these advantages cease, because now you are facing the best of the best ballplayers. And thus a hitter - Dallimore - who hit .352/.427/.448/.875 the year before and .324/.396/.447/.843 the current year, only hit .279/.347/.395/.742, which is actually not that bad, but he's a 30 year old rookie, basically having his last hurrah. The next year he deteriorated to .302/.369/.437/.806, again in AAA, as his skills eroded.

And while they sound good, what analysts have found is that these numbers can be translated into what they call "Major League Equivalents", where analysis found that if you had a certain hitter, then he would hit a certain way in the majors, in terms of performance. Thus his .324/.396/.447/.843 translated into .252/.296/.333/.629 and his .302/.369/.437/.806 translated into .231/.268/.324/.592 (all from Shandler's book), both Neifi-ish levels of hitting incompetence, but without the Gold Glove defense . And one of the tenets in the book is that defense is what gets you to the majors but it is hitting which keeps you there (though obviously EME will test that tenet sorely :^).

So while one can say "of course, AAA is different from AA, don't be rediculous, cousin Larry," as Balki might have said it, there is some data to support that. From Shandler's book, there is clearly a difference for pitchers, but not so much for hitters, so that could have contributed to allfrank asking the question. And from my data, there is clearly a different mix of players between the two leagues, and while there are more post-peak players in AAA, I think natural selection makes them the best of the best who are not major leaguers. And, of course, there are always a number of them who once played regularly in the majors for a season or two, but then lost whatever magic they had to compete in the majors, but can still do well enough in AAA to be an option should fate deal him a good hand by having him called up to replace another player, for whatever reason, injury or poor performance or both.

Lastly, it would not make sense for MLB teams to stock their AAA and AA teams with equal talent, you want your best players playing together and against other teams' better players so that they can learn and grow and advance to the majors and be able to contribute significantly there.

Sudden thought, what could be happening, and I'm not sure if your view is skewed by observing only the Giants AAA and AA, allfrank, or if you did view both leagues in entirety, the Giants AAA and AA teams are skewed in that the best of the Giants prospects are generally on their AA team, the prospects who have been struggling to break through are stuck in AAA, stalled in their march upward. Thus the two teams may not appear to be that different to you because of this factor.

But generally, there should be some difference between in talent between AAA and AA unless the team fortunately (from too many good players) or unfortunately (from too many mediocre players; hello Giants!) have their teams populated that way. You want your best playing your opponent's best, you want to test your prospects so that you can better gauge how close your prospects are to making the majors or not. Else we could keep Cody Ransom (name your failed position prospect here) in AA and he'll look REALLY good.