Wednesday, April 4, 2007

The Effect of Steriods in Baseball

I ran across this op-ed in the New York Times on Monday. The premise of the article is that the recent increase in home runs in baseball is due to expansion, not to steroids. In the 1990s, MLB added four expansion teams; the theory is that these expansion teams diluted the talent in baseball, allowing the very best players to excel and hit more home runs. Now I don't know much about baseball, so I don't feel that I can judge this theory very well, but my gut feeling is that this seems unlikely.

First of all, the expansion only increased the size of the league by 15%. Are the top players in the minors really that much worse than major-league players that they could cause such a significant increase in home runs? By comparison, 15% is approximately the population growth of the U.S. between the 1977 expansion and the 1993 expansion. I would think that the pool of major-league-quality baseball players would have grown similarly during that time period.

I can understand that there might be a short-term effect, especially if all the "bad" and inexperienced players were initially clustered in the four expansion teams. But I would think that it would have evened out by now with trades and attrition. (The Marlins won the World Series after only 5 years in the majors, so clearly the expansion teams weren't stuck at the bottom with inferior players.)

Now if you look at the home run statistics, there's a significant jump in home runs in the NL in 1993, when the NL gained two teams. Before 1993, the NL seemed to be averaging around 1300 or 1400 home runs per season. Since 1993, it has never had fewer than 1900 (excluding 1994 because of the strike). Of course, some of that increase is due to the fact that there's more teams in the NL now. If you look at home runs per team in the NL since 1977, this is what you get:

(Again, 1994 was special, although given that they were about 70% of the way through the season, they were on course for more home runs than in 1993 or 1995.)

Unfortunately, this graph is inconclusive to me. Sure, the number of home runs per team has gone up about 50% in the past 15 years, and the increase happened around the time of the expansion. There was even a suspicious jump in home runs in 1993. But 50% seems like quite a lot to me, and I still wonder whether the increase is better explained by other factors. It could be drugs. It could also be a rational shift in baseball strategies. (Maybe trying to hit home runs helps you win games? Or maybe home runs attract spectators?)

As I said, I don't follow baseball very closely, so I'm interested to hear what other people think.


dave hiller said...

Thanks for posting, Ben.

One interesting point is that you can tell the quality of a league based on the difference between the lowest ERA and the highest BA. As the article points out, the variance is smaller if you're drawing players from the far right of the skill distribution. Therefore there are fewer extremely low ERAs or high BAs.

The catch is, according to this admittedly rudimentary analysis, the quality of the league has not diminished (figure). (That 2000 outlier is Pedro Martinez and possibly the best year ever by a pitcher).

I think Ben is on to the reason - despite expansion, the pool of players to draw from has increased, both because of population growth and because of a greater influx of players from Latin America, the Far East, etc.

Instead of an overall dilution of talent, there was an increase in offense over defense (figure. I'm not sure why that is - but a number of things happened between 1993 and 1995 - expansion, realignment, the strike. Since then runs have been trending slightly downwards.

Alan Rosenwinkel said...

Exactly what portion of the rise in home run rate can be attributed to steroids is a very hard question to answer, but I would argue that as least some of the added home runs are due to factors other than steroids.

I think league expansion is a factor, but not because "the very best players excel and hit more home runs". Some studies have indicated that it's actually the middle group of average hitters, not a small group of elite hitters, who account for the increase in home runs. The reason expansion leads to more HRs, I would argue, is that there are fewer good hitters than good pitchers, so when the league expands, the pitchers talent pool is diluted more than the hitters. This is only a problem if the population doesn't keep up with expansion, which the overall population has; however, the number of 18-24 year olds actually decreased from 1980-1990. Also, I'm not buying Dave's "league quality" argument. There's too much noise in that metric to learn anything meaningful.

The trend towards smaller fenway-style ballparks over that last two decades has certainly contributed to the rise in HR rate, as well as the drop in triples. Anyone who has watched the phillies transition from the pitcher-friendly VET stadium to their new "bomb box" can attest to this. Citizen's Bank Park gives up about 30% more HRs than the VET did, which would cause about a 1% rise in HRs in the league just from that one new stadium.

Pitchers have also been known to take steroids, so it's not certain that hitters would be more advantaged than pitchers anyway.

Another interesting note is that the jump in HR/AB we have seen recently is by no means unprecedented, leading me to the conclusion that it is certainly plausible that this recent trend in non-steroid related (or non-steroid dominated).

Rapid said...

I am happy to say that I ordered Testosterone Anadoil from and I am very happy with the results! I highly recommend this.The best part is no pescription required when ordering from them!!!

Anonymous said...

life insurance quote treatment of mesothelioma

Anonymous said...

cheap car insurance california chemotherapy for mesothelioma

Anonymous said...

car insurance texas auto insurance pa

Anonymous said...

car insurance in michigan

php hit counter