Expansion Dilutes Hitting?
by Asher B. Chancey, BaseballEvolution.com
June 28, 2008

I'm working on a theory that has resulted from dwelling for too long on Bill James' insistence that expansion does not dilute pitching any more than it does hitting.

If this is true (and it makes logical sense, but I've always thought the empirical evidence seems to say otherwise), then there must be some other explanation for the performances of Roger Maris and Mickey Mantle in 1961, Harmon Killebrew and Willie McCovey in 1969, George Foster in 1977, and Mark McGwire and Sammy Sosa in 1998. It simply can not be denied that there have been many noteworthy individual season accomplishments in expansion years which happened to coincide with expansion.

But maybe this is the key – maybe expansion dilutes hitting and pitching equally, so the leagues as a whole don't improve significantly on offense, because the dilution of hitting and pitching is balanced, but the individual hitters improve, because the pitchers they are facing have been diluted. Thus, while the National League as a whole would not expect to many more runs per game in 1977 than in 1976 (which, for the record, they did do, since scoring jumped from 3.98 in 1976 to 4.40 in 1977), because the bad hitters brought in expansion counter-balance the bad pitchers, George Foster individually could be expected to hit more homeruns than the year before because of the decline of the overall pitching talent he is facing.

(It occurs to me that expansion must impact the caliber of play in every facet, meaning diluted pitching, diluted hitting, and diluted fielding. Two of these things favor increased run production, one of these things favors decreased run production. Maybe this explains the thing Bill James is trying to deny.)

But do you know what that means? It means the same would be true for the pitchers – while pitching would not improve as a whole, individual pitchers would improve because they were facing a diluted pool of hitters. But that didn't happen.

Or did it?

Let's briefly name the Top Ten Pitchers of All Time. We'll probably all disagree as to order, so the following will be in no particular order: Walter Johnson, Cy Young, Pete Alexander, Christy Mathewson, Lefty Grove, Tom Seaver, Randy Johnson, Roger Clemens, Greg Maddux, and Pedro Martinez.

And odd list. Cy Young played half his career in the 19th Century; Alexander, Mathewson, and Johnson played in the first 25 years of the 20th Century, Lefty Grove in the 1930s, Tom Seaver in the 1960s, 1970s, and 1980s, and then four guys who are roughly contemporaries, all coming to the end of their careers in the last few years.

It has always struck me as amazing that four of the greatest pitchers of all time played in the same period, and this era of baseball featured four pitchers better than any two that any other era of baseball had ever featured. What made it even more incredible was the fact that these four guys had all been dominant during a period that was notoriously tough on pitchers; these four guys pitched better in a hitter-dominated era than any other pitchers had ever pitched in pitcher friendly eras.

Maybe the premise alone should have tipped me off.

If hitters excelled in this expansion-filled era (four teams in five years), not because expansion dilutes pitching uniquely but because expansion dilutes both and allows individual hitters to excel, then wouldn't the same be true for pitchers? Shouldn't individual major league pitchers, particularly the best in the league, only get better after expansion dilutes the league's hitting?

Let us take a look.

Major League Baseball expanded in the 1990s in 1993 by adding the Florida Marlins and Colorado Rockies, and then again in 1998 by adding the Arizona Diamondbacks and the Tampa Bay Devil Rays. So 1993 and 1998 should be our guide-posts in terms of when we would expect a pitcher's numbers to improve because the pitcher is now facing a diluted pool of hitters. Let's go blow-by-blow.

Roger Clemens, 1984-2007

Clemens burst onto the scene in 1984, and in 1986 won the AL MVP and Cy Young Awards in his first full season of action. He won the Cy Young again in 1987, deserved to win it in 1990, and then won it again in 1991. So it would be hard to argue that Clemens wasn't a dominant pitcher without expansion. Further, the first bad year Clemens ever suffered through happened to be 1993, when he finished with a losing record and an ERA over 4.00 for the first time in his career. Clemens had a big year in 1998, winning the Triple Crown and the AL Cy Young, but that year was not as good as 1997, when he did the same thing in a more impressive manner. It could certainly be argued that Clemens' rejuvenation in the late 1990s was a result of the expansion era, and that his numbers looked better because they were being compared to a diluted pitcher's pool, but the evidence doesn't support that argument very well.

Greg Maddux, 1986-2008

Maddux started with the Chicago Cubs, and looked very good, winning a Cy Young Award in 1992 before defecting to the Atlanta Braves. While his 1993 season is actually quite similar in numbers to his 1992 season, Maddux's 1994 and 1995 numbers were absurdly dominant compared to the rest of his career – his relative ERAs for those years are fourth and fifth best of all time – and his run from 1994 to 1998 is one of the best of all time. Ironically, though, his last truly dominant season was 1998. Maddux's numbers certainly support an expansion era hypothesis, but he didn't show much after 1998 to merit such an assumption – his numbers from 1999 to 2002 were merely great. But who knows, perhaps the expansion of 1998 kept his career going and put off the mediocrity he is suffering through now for another five years or so.

Randy Johnson

Johnson came up with the Montreal Expos, but struggled to crack the majors until he was traded to the Seattle Mariners. The Mariners immediately put him to work, and he enjoyed three straight seasons of 200-plus innings with an ERA just over league average. His breakout year came in, get this, 1993, when he improved from 12-14 with a 3.77 ERA to 19-8 with a 3.24 ERA. More important, perhaps, was his K/BB ratio, which improved from 241/144 in 1992 to 308/99 in 1993. Johnson was dominant-when-healthy from 1993 to 1997, winning the 1995 Cy Young Award but missing almost all of 1996. His 1997 season was admittedly superb, but it was 1998, when he was traded to the Houston Astros, that he truly became the dominant superman that we think of him as now. After finishing the season with the Astros on a 10-1, 1.28 ERA run in 1998, Johnson went on to post four straight seasons of 245-plus innings, 330-plus strikeouts, and ERAs in the mid-2.00s (ERA+ in the 180s). Randy Johnson showed noticeable bumps around both 1990s expansions, going from solid to great in 1993, and then from great to elite in 1998.

Pedro Martinez

Pedro's first full season in the major leagues was 1993, so we have nothing to compare him to pre-expansion. Pedro was a very good pitcher from 1993 to 1996, but he was also quite young, so it would be unfair to assume he should have been elite during that period. Further, his first dominant season came in 1997, when 17-8 with 305 strikeouts in 241 innings and a 1.90 ERA. And in 1998, after moving from the Montreal Expos to the Boston Red Sox, Pedro's numbers actually suffered a bit (which, admittedly, one would expect from a pitcher moving from the NL to the AL, and from Olympic Stadium to Fenway Park). But from 1999 to 2003, Pedro was as dominant as any pitcher has ever been, posting successive relative ERAs of 243, 291, 189, 202, and 210. His 1999 and 2000 seasons are both in the top ten relative ERAs of all time. And though injuries have had as much to do with Pedro's return to earth the last few years as anything, it is noteworthy that his cool-down has coincided with the cool-down on the offensive side of the ball as well; one would have expected him to become more dominant during that period, not less.

Conclusions None of this is definitive proof of anything. But in the last five years or so, I personally, and a lot of other people generally, have heaped praise on these four players for not only being amongst the elite pitchers in the game during the offense-heavy 1990s, but for actually being more dominant than any other pitchers in the history of the game. For the first time I think I am willing to consider the possibility that this is not the correct perspective here.

For years I have been giving these four guys credit for succeeding against a superior offensive Major League Baseball. But I think we might need to look at this as four guys who succeeded against many individually superior offensive players, but also many very poor ones as well. We should also consider the fact that three of these four guys were strikeout pitchers succeeding in a period in which the additional offense was marked also by additional strikeouts, and the fourth one was a control pitcher who never gave up homeruns succeeding in an era in which the offensive up-tick was fueled by an up-tick in homeruns.

It is an awfully hard hypothesis to accept in the case of Roger Clemens, given his dominance in other seasons and his lack of dominance following the 1993 expansion. But we have other explanations for his play from 1998 forward.

Greg Maddux seems to line up with the expansion era pretty well, but there are certainly plenty of other factors to explain his success – moving from Wrigley Field to go play for the Braves, playing for Leo Mazzone, and especially his perfecting of the six-inches-off-the-plate strike; when Major League Baseball cracked down on the strike zone the umpires were calling, Maddux's career as a dominator was over. But those 1994 through 1998 seasons, immediately following the 1993 expansion, sure do look suspect.

Unfortunately for Randy Johnson, his career trajectory lines up almost perfectly with Major League Baseball's period of expansion. He went from a mediocre pitcher before 1993 to a dominant one after 1998 by relying on one of the principle bad things the new young hitters being rushed into the majors were willing to do – strike out a lot.

Pedro Martinez may be the most scandalous player on this list, because of his relatively short career and the fact that he played exclusively in the expansion era. My inclination with Pedro is to think that he is one of the truly great pitchers of all time, but I think opinion of him as the best per-inning pitcher is probably the incorrect one.

As an afterthought, I think it is worth noting that in 1998, the National League scored exactly the same number of runs per game – 4.60 – as it had in 1997; it was not until 1999 that the league jumped up to 5.00 runs per game. And while the American League jumped from 4.93 runs/game in 1997 to 5.01 in 1998, it jumped a lot more from 1998 to 1999, going up to 5.18. Not sure why this is worth noting. Just thought it would be.

It is also worth noting that the correctness of my entire hypothesis depends on me agreeing with Bill James, which bothers me, of course, but I guess I will need to deal with that.


Questions? Concerns? Comments? Asher lives in Philadelphia, PA, and can be reached at asher@baseballevolution.com.