Yes, Samantha, Your Cleveland Indians Have Been Unlucky, Not "Worst"

The Cleveland Indians are a team that's loaded with talent. That talent is deployed more successfully at some times than others.

But who would expect them to have traded places with the Kansas City Royals for the worst record in the American League, while the Royals are near the top of the American League Central?

Certainly not a particular writer who asked rhetorically if her Cleveland Indians were the worst team in baseball. This, my answer to her, is that indicators of raw talent show that your team has been at least partly unlucky.

One of these indicators is runs scored (RS) versus runs allowed (RA) and the relationship between them. So far on the season, the Indians have scored 208 runs and allowed 231.

That suggests that they should be below .500, but not to the tune of 150 points. In fact, the theoretical result is that they should be about 50 points above, rather than below, .400, or 18-22, leaving them tied for third place with the Minnesota Twins and still in the running for the division title.

Specifically, the so-called Pythagorean formula from "sabermetrics" (baseball science) is:

Winning Percentage (WP) = RS-squared/(RS-squared + RA-squared).

More runs scored mean more winning chances. More runs allowed mean more losing chances, and what matters is the difference between them (although it is in a "squared" rather than "linear" relationship).

The Indians' negative difference of -23 is high, but similar to that of the Twins (-22) and lower than that of several other teams, including the Oakland Athletics (-27), the Seattle Mariners (-31), the Chicago White Sox (-31), and the Baltimore Orioles (-41), all of which have better records than the Indians.

How did this come about? The Indians have lost too many close games, like last night's 6-5 heartbreaker to the Kansas City Royals—not to mention one-and two-run losses of 4-2, 7-5, and 8-7 to Tampa Bay in the previous series.

On the other hand, their record-setting 22-4 romp over Chien-Ming Wang and the New York Yankees earlier this year was good for only one win. Over these five games, the Indians are a robust 41-29 in points, and only 1-4 in victories.

Moreover, Samantha's intuition that Carl Pavano is better than his 6.33 ERA is supported by other statistics. According to FanGraphs, his "FIP," or what I call sabermetric ERA, is 4.06, or 2.27 lower. (Cleveland's other pitchers exhibit much smaller discrepancies that basically cancel out.)

These so-called "fielding independent pitching" statistics are based on the three things that a pitcher can control: home runs, walks, and strikeouts. Other factors are assumed to cancel out. It's a formula that works for most pitchers, but only over the "long run" (several seasons).

Pavano has given up only five home runs (10 percent of his team's total) and 10 walks (six percent), and he has more than his share of strikeouts. Yet 30 runs have been scored against him by Cleveland's opponents (14 percent).

That's about 11 runs too many, given his other stats. If you subtract Pavano's 11 "unlucky" runs from Cleveland's runs allowed, and plug RS=208 and RA=220 into the Pythagorean formula above to get a theoretical WP, the Indians ought to be about 19-21 on the season.

It's worth noting, however, that the Indians have historically been a sabermetrically "unlucky" team. In 2005, for instance, the Indians outscored the Chicago White Sox 790 to 741 while allowing a similar number of runs, 643 vs. 645.

Yet the White Sox won the division race with a 99-63 record for a win percentage of .611, while the Indians had a 93-69 record for a lower .574 win percentage. (Pythagorean percentages were .601 Indians, .569 White Sox, almost the reverse of the actual.)

And in 2006, the Indians scored 88 more runs than they allowed, yet finished below .500.

Some teams can't seem to catch a break.