Twelve years ago, in his smash hit, "Back That Thang Up," Juvenile made the following prediction: "Cash Money records taking over in the '99 and 2000."
Mr. Juvenile was wrong, though. In '99 and 2000, the only man taking over was Pedro Martinez. Pedro's performance from 1999-2000 is the greatest two-year peak that any pitcher has ever had.
In both years, he made the All-Star team, won the Cy Young award, and finished in the top five of the MVP voting. As Joe Posnanski wrote a couple of months ago:
I have always loved what Bill James said about Pedro Martinez; he basically said that Pedro was a testament to the exponential power of numbers. Ten plus 10 is only 20. Ten times 10 is only 100. But 10 to the 10th power is 10,000,000,000, which is 10 billion. Martinez at his best had a great fastball, a great slider, a great curveball and a great change-up. Each individually was good enough to make him effective. Multiplied by each other, well... he was something beyond great.
So which season was better? Here's the comparison:
1999: 23-4, 2.07 ERA, 313 Ks, 213.1 IP, 0.92 WHIP, 1.39 FIP, 243 ERA+, 13.2 SO/9
2000: 18-6, 1.74 ERA, 284 Ks, 217 IP, 0.74 WHIP, 2.17 FIP, 291 ERA+, 11.8 SO/9
Martinez was rarely perfectly healthy, but his playing time was similar in both seasons, so that doesn't make the comparison any easier. Martinez spent time on the disabled list in each year, a late-July stint in 1999 and an early July stint in 2000 that caused him to miss the All-Star game. He only made a combined 58 starts in the two seasons, 29 in each year.
He finished second in the MVP voting in 1999 while only finishing fifth in 2000. So the voters seem to have felt that 1999 was more dominant. Certainly, he won more games,
pitched more innings, and struck out more batters in 1999. But in 2000, his ERA was far lower, his ERA+ was far better, he pitched a few more innings, his K/BB was a smidge better because he lowered his walks, and his WHIP was a disgustingly inconceivable .74, the lowest ever by a starting pitcher who qualified for the ERA title. (The previous record was Walter Johnson's 0.78 in 1913.) Pedro was amazing in 1999, but in 2000 it was pretty much impossible to reach base against the man.
Because his FIP was lower in 1999, Fangraphs' fWAR stat favors 1999 over 2000, 12.1 to 10.1; because his ERA was lower in 2000, and the league's ERA was higher, baseball-reference's rWAR stat favors 2000 over 1999, 10.1 to 8.4. In fact, the disparity between his ERA and that of other mere mortals in 2000 was staggering: not only did Pedro's 1.74 ERA lead the league, but it was literally less than half that of his closest competitor, Roger Clemens, who posted a 3.70 ERA. Pedro was on another planet.
In 2008, blogger Steve Caimano found that the FIP disparity was largely due to Batting Average on Balls in Play. Over his career, Pedro had a BABIP of .282, but in 1999 it was .325 and in 2000 it was .237. That discrepancy also more than explains the difference in ERA between the two years. Essentially, his FIP is far lower in 1999 because the FIP formula considers his high 1999 BABIP to have been "unlucky"; more balls fell for hits than would otherwise have been expected, which shouldn't be considered his fault. His FIP is higher in 2000 because, by the same token, his BABIP in 2000 was "lucky," as he gave up far fewer hits than he otherwise usually would have.
Whether you believe that Pedro's 1999 or 2000 season was greater basically comes down to how old-school or new-school you are. As I wrote a year ago:
Basically, the fundamental difference between old-school stats and new-school stats is that old-school stats measure what happened at the surface level — batting average, earned run average, wins. New-school stats try to measure each player's contribution to those surface stats, while filtering out the contributions of their teammates and the random fluctuations of chance.
Pedro's 1999 season was otherworldly, and he probably deserved to win the MVP. But his 2000 was even better on the surface, and I just don't see anything below the surface to contradict it. So the only way to say that his 1999 season was better than his 2000 season is if I decided I didn't want to give him full credit for what actually happened in 2000 — he gave up fewer runs in 2000, and the only way to explain that away is to say that he somehow didn't deserve to.
I just can't bring myself to do that. FIP is a great predictor for future performance, because it measures the components of runs at a more granular level than ERA, but because it isn't run-based, it isn't a great descriptor of the past. I'm not so uncharitable as to penalize Pedro for getting lucky on his BABIP in 2000. Hey, I was lucky enough to watch him. It was the best season I'll ever see any pitcher have.
There's no question that was 1999 was great. But 2000 was better.