However, the claim that they are the "greatest" implies that some comparison to other teams will be made. There is one paragraph devoted to that concept. The paragraph mentions that only two teams -- '06 and '09 Cubs -- had better road winning percentages, at .800 and .740. Selter then concludes the Yanks were better on the road, because their victory margin was 3.9 runs per game, whereas the '06 Cubs had a margin of 2.5 and the '09 Cubs had a margin of 2.0.
I can scarcely believe the BRJ published the article. First of all, it's just an article about what a great team the '39 Yanks were, and barely makes any attempt to measure them against other teams. Second, the numbers are not even correct. The '06 Cubs margin of victory was 2.66 (not 2.5) and the '09 Cubs margin of victory was 2.09 (not 2.0). Seems the Yanks got the benefit of some rounding up in the article, and the Cubs did not.
Finally, it is hard to understand how a larger margin of victory on the road by the Yanks supersedes the better road winning percentage of both Cubs teams. The margin of victory seems to mean when the Yanks won, they won bigger than the Cubs teams. But the fact is, they didn't win as much. So how were they better?
Let's look at the numbers. The following table shows the road wins, losses, runs scored by the team, runs against, and the league average runs scored:
Team W L R/G RA/G LgAvg.R/G
'06 Cubs 60 15 4.86 2.20 3.57
'09 Cubs 57 20 4.54 2.45 3.66
'39 Yanks 54 20 7.80 3.93 5.21
Even if it is a plausible claim that a greater margin of victory makes you better than your winning percentage -- which of course it isn't -- it is far from clear that the Yanks actually have a greater margin of victory. Of course they have a greater absolute margin of victory, but since they played in the late 30s, when run scoring per team per game was 1.5 runs higher than in the first decade of the century, that absolute margin does not mean much.
Let's reduce the Yanks R/G and RA/G to a league run environment of 3.6 runs per game, to approximate the first decade of the century. What you get is 5.39 R/G and 2.72 RA/G for the '39 Yanks. The margin of victory is 2.67. The margin for the '06 Cubs is 2.66 and for the '09 Cubs is 2.09, as mentioned before. So even if the crazy proposition that margin of victory trumps winning percentage is treated as true, the '06 Cubs were at least as good on the road as the '39 Yanks.
Now look at their projected Pythagorean road records, using 1.83 as the exponent, and how the team performed relative to the projected road record.
Team Pythag Diff
'06 Cubs 61-14 -1
'09 Cubs 58-21 -1
'39 Yanks 58-16 -4
First, I'll point out that using the very runs numbers that Selter used, the Yanks still produce a poorer road record than the '06 Cubs. Second, the Yanks underperformed their expected record by a larger margin than the '06 and '09 Cubs. To me, this indicates the Yankees were not the greatest road team. They did not even meet their expected winning percentage, falling short by 4 games, which is a substantial amount given the typical margin of error for the Pythagorean formula.
