Given the return to the 70s era offensive environment, has the chance of someone winning a batting Triple Crown gone up?
Asked by: julesig
Answered: 4/29/2012
Why would it?
Hey Bill, I have some thoughts on the Triple Crown issue. In any area, the more people competing, the less easy it is to dominate, especially in diverse skills. In my circle of friends, I can be the best tennis player and the best racquetball player, but as the circle expands, very few people can be the best at both. With rising strikeout rates, don't you sort of have to make a choice between pursuing batting average or home runs? It seems like you either have to specialize in batting average (Carew, Rose, Gwynn, Suzuki) or power (Schmidt, Sosa, McGwire, R Howard). Basically, your only candidates have to be power hitters who don't strike out. I'd guess Triple Crown possibilities are closely tied in with the strikeout trends you've recently discussed.
Asked by: rtayatay
Answered: 5/2/2012
I think the effect is more the opposite one. Triple crown winners could disappear (a) because you didn't have the same hitters hitting for high averages and with power, or (b) because you had so many hitters hitting for high averages and with power that none of them dominates the others. I think it is more (b) than (a).
In all of baseball history up to 1967 (the Triple Crown era) there were 177 players who hit .300 with 30 homers. Since 1968 there have been 308 players who did this.
Regarding Triple Crowns in low-offense environments, I'm wondering whether it is easier for one batter to separate from the others in all three categories when overall offense is down, conversely, in high-offense eras, there are more potential competitors in each category. Or is it just coincidence/small sample size that the last two Triple Crowns came in 1966 and 1967?
Asked by: julesig
Answered: 4/30/2012
Neither. Something changed about 1970 that I don't understand fully.
There were also several triple crowns about 1930-1933, a very high-scoring era, and I don't see any reason to believe that the level of scoring offense predicts the likelihood of a triple crown. But in that era teams had one or at most two power hitters per team--generally one. There were a limited number of players who COULD lead the league in home runs and RBI.
This changed gradually over time and shifted suddenly about 1970. We are, in a sense, much LESS specialized (in baseball) than we were in the 1920s/1930s. In the 1920s/1930s there were "hitters" who hit .350 with 45 homers, and there were middle infielders and catchers who hit .240 or sometimes .210 with no homers. In modern baseball we don't have offensive and defensive specialists to that degree. Everybody is expected to hit, and the difference between what the best hitters hit and what the lesser hitters hit is nothing like what it was at that time.
At one point I assumed that the disappearance of Triple Crown winners was simply a function of the expansion of the leagues, which created a lot more candidates for each "crown". To demonstrate that this was true, I tried to show that the number of players leading their own TEAM in home runs, RBI and batting average was the same as it always was.
Only, to my shock, that wasn't even CLOSE to being right. The number of players leading their own TEAM in home runs, RBI and batting average nose-dived about 1970, for reasons that I don't really understand. So. ..if you don't have players leading their own team in the triple crown categories, obviously you're not going to get players leading the league in all three.
Making the Triple Crown discussion more concrete, what would you say are Matt Kemp's chances of winning it this year? 1000 to 1?
Asked by: julesig
Answered: 5/3/2012
More like 20 to 1. First, although no one has won a triple crown in 45 years, this is probably a fluke, and the odds would probably be that there SHOULD have been Triple Crown winners in there. In other words, if "shook up" the data, you'd probably get Triple Crown winners. Second, Kemp was very close to a Triple Crown last year. Third, he has the advantage of a torrid start this season. The odds are against his winning the Triple Crown, but they are certainly nothing like 1000 to 1.
Couldn't the disappearance of Triple Crown winners be the result of essentially the same effect to which Stephen Jay Gould attributed to the disappearance of the .400 hitter? That "variation in batting averages must decrease as improving play eliminates the rough edges that great players could exploit, and as average performance moves towards the limits of human possibility and compresses great players into an ever decreasing space between average play and the immovable right wall."
Asked by: Scott_Ross
Answered: 5/1/2012
That could contribute. It couldn't cause the disappearance of Triple Crown winners, absent other causes.
There were actually three Triple Crown winners in the 1970s: Secretariat, Seattle Slew, and Affirmed.
But none since then.
--evanecurb
Regarding the drop in triple crown team leaders around 1970: if you look at the frequency of a player leading his team in home runs and RBIs only (not looking at batting average) is there still a big drop?
--kgh
Regarding the near disappearance of Triple Crown candidates after 1970: I think you were too quick to dismiss the divergence of batting-average hitters and power hitters as a cause, and I don't know why you think it began in 1970. Just looking at the NL from 1960-1970, the lowest avg. to win a batting title is .325 (Dick Groat in 1960) and the highest avg. of a home run champ is .319 (Aaron in '63--when he came with .007 of winning the Triple Crown--the only near miss of the decade). In fact, only one guy to finish in the top 5 of HR (Frank Robinson in 1962, when he hit .342, losing the batting title by .004 but also finishing 10 HR behind Mays) had an average that would have won a title in any year between 1960 and 1970. What we're looking for is not guys who his 30 HR with a .300 avg but guys who hit 40 HR with a .325 avg. (Only 2 home run champs in this period finished with less that 40: Aaron with 39 in '67 and McCovey with 36 in '68).
--wwiyw
The Triple Crown Winner, like the T-Rex, is an animal that used to exist but no longer does. I am not sure why we are interested in this, but I’m interested in it and some of you seem to be interested in it, so. …
This article repeats and updates some research that I did some time in the past, I think in the 1990s. I was asked about the disappearance of Triple Crown winners, and I started to reply that it happened mostly because of the expansion of the leagues, consistent with the theory advanced by rtayatay. In an eight-team league there are 64 candidates to lead the league in each category, 64 regulars. In a 14-team league with a DH there are 126 candidates; in a 16-team league with no DH there are 128. It is a great deal more difficult to lead 126 or 128 players in all three categories than it is to lead 64.
I was going to illustrate that this was the only thing that had changed, and to illustrate this, I was going to point out that the percentage of players leading their own TEAM in all three categories was just the same as it always was. Except that, when I went to document that that statement was true, I discovered that not only was it not true, it wasn’t even close. The percentage of players leading their own team in all three categories actually dropped sharply in 1970.
So I remembered that, but I never know where I printed anything or even if I printed it, so I couldn’t really cite the research. I decided to repeat the research so that I would know what I was talking about in specific terms.
OK, in the 1960s there were 198 major league teams. Of those 198 teams, 56 had one player who led the team in all three categories. That’s 28.3%. I haven’t repeated the research for previous decades and I’m not going to, either, but I believe that this percentage is consistent with previous decades, before 1960.
From 1970 to 1989 there were 506 major league teams. Of those 506 teams, only 86 had one player who led the team in all three categories. That’s 17%.
We haven’t had a triple crown winner since 1967, an East Coast player named Yarsewlenski or something. The percentage of major league teams which had one player who won the team Triple Crown dropped very sharply in 1970. I don’t really understand why.
Mr. Ross asked if this could be a manifestation of the effect noted by Stephen Jay Gould many years ago, of the disappearance of the extremes in baseball? Well. ..yes and no. Gould. . .I assume most of you know this. .. Gould explained the disappearance of the extremes in baseball—the disappearance of .400 hitters, for examples—as the normal compression of a competitive species. You don’t get members of the same species who outweigh one another ten to one, as a rule; you don’t get members of the same species, one of whom weighs four ounces and the other of whom weighs 20 pounds.
Certainly the compression of the population around a norm contributed to the disappearance of the Triple Crown winner, yes. But that effect is so small that you would normally have great difficulty establishing that this compression had actually occurred, within a 20-year window. You can establish that it happened within a 40-year window, sort of, and you can clearly establish that it has happened over a longer period of time. But that effect is not strong enough or sudden enough to explain a change of this nature, in my opinion.
Also responding to Mr. Ross on a different level. .. we want to be careful about interpreting new information as merely a manifestation of things we already know, right? I’m as guilty of that as any of you, but. ..it’s a bad habit. If, when something new is pointed out to you, you immediately explain that as merely a new manifestation of something that you already know, then you’ve passed up an opportunity to learn from the new phenomenon.
Well, then, what did happen?
Regarding the near disappearance of Triple Crown candidates after 1970: I think you were too quick to dismiss the divergence of batting-average hitters and power hitters as a cause, and I don't know why you think it began in 1970.
1) I don’t think something happened about 1970; I know something happened about 1970. I just don’t know exactly what.
2) Well, but what is the theory that I dismissed, exactly? Can somebody explain to me what it is that you think happened in specific terms? Why exactly would rising strikeout rates cause power hitters and batting average hitters to split? It would seem to me, intuitively, that the opposite would happen. As strikeouts go up, it is impossible to compensate for that by increasing your ball-in-play batting average, since the hitter has limited control over his ball-in-play batting average, which has relatively little variability. The compensation comes from balls hit OUT of play—thus, as strikeouts increase, the power hitters should move to the head of the line in batting average. I would think.
Just looking at the NL from 1960-1970, the lowest avg. to win a batting title is .325 (Dick Groat in 1960) and the highest avg. of a home run champ is .319.
Setting aside the sheer absurdity and (frankly) chutzpah of offering up a "study" based on eleven players and suggesting that we should cheerfully generalize based upon it,
a) actually, 1969 NL home run champion Willie McCovey hit .320,
b) in this population of eleven home run champions, we have batting averages of .320 (Willie McCovey, 1969), .319 (Henry Aaron, 1963), .317 (Willie Mays, 1965), .311 (Orlando Cepeda, 1961), .307 (Hank Aaron, 1967), and .304 (Willie Mays, 1962). It is fanciful nonsense to suppose that a population of eleven hitters, of whom six hit between .304 and .320, is somehow evidence that no players of this type would hit .325.
What we're looking for is not guys who his 30 HR with a .300 avg but guys who hit 40 HR with a .325 avg. (Only 2 home run champs in this period finished with less that 40: Aaron with 39 in '67 and McCovey with 36 in '68).
Oh, I’m sorry; I assumed that the reason I was doing that was obvious. I can see that it isn’t.
The reason for looking at players who hit .300 with 30 homers is the same as the reason for looking at players who led their TEAM in all three categories; same reasoning. It is a common research technique, when looking at something which is rare and therefore difficult to study, to expand the field by looking at the nearest relatives of the target population. If you are trying to study a nearly-extinct butterfly, you would have difficulty trying to capture living examples to study, so you capture and study similar and related species of butterflies. If you’re trying to study pitchers who throw 100 MPH for long periods of time, there are only two you can study (Nolan Ryan and Billy Wagner), so you study pitchers who throw 96 for a sustained period of time.
It’s pretty useless to try to study hitters who hit 40 homers with a .325 average, because those are nearly as rare as Triple Crown winners themselves. I assumed it was obvious that, if the population of hitters who hit .300 with 30 home runs is stable or increases, then the population of players who hit .325 with 40 homers could be expected to follow the same trend line.
Since 1967 there have been 35 players who have hit .325 with 40 or more homers. It’s just that none of them have won the Triple Crown.
Regarding the drop in Triple Crown team leaders around 1970: if you look at the frequency of a player leading his team in home runs and RBIs only (not looking at batting average) is there still a big drop?
I think the best answer would be "no"; the real drop is not in players leading in home runs and RBI; it is in players leading in RBI and batting average. In the 1960s, 69% of teams were led in home runs and RBI by the same player. In the 1970s and 1980s this dropped to 65%.
However, in the 1960s 39% of teams were led by the same player in batting average and RBI. In the 1970s and 1980s this dropped to 31%. So that drop is actually much larger.
There are also a handful of teams which are led by the same player in home runs and batting average, but not RBI. That number actually went way up after 1970, but the number is so small anyway that we couldn’t generalize from it. In the 1960s only two teams were led in home runs and batting average by one player, but in RBI by a different player; those two teams were the 1963 Yankees (Elston Howard) and the 1966 Cardinals (Orlando Cepeda). That’s 1%. After 1970 it is 4%. But still. ..4%.
More information about my study:
The percentage of teams which had a Team T-Rex was 28.3% in the 1960s, 19.9% in the 1970s, and 14.2% in the 1980s. This suggests that there was not simply a downward shift, but a downward trend.
Dividing the decades into halves, the percentage was:
24.7% in the early 1960s (1960-1964),
31.4% in the late 1960s,
22.5% in the early 1970s,
17.5% in the late 1970s,
13.8% in the early 1980s, and
14.6% in the late 1980s.
The percentage of teams which were led in these three categories by three different players was 21% in the 1960s, 20% in the 1970s, and 18% in the 1980s.
Thanks for reading.
Bill