So what is considered an outlier?
Consider the following. Last year's staff allowed 753 runs, which equates to an average of 4.65 runs per game; the standard deviation for this data was 3.43. So a shutout would be within about 1.33 standard deviations. The 5 high games were 18, 14, 14, 13 and 13. or between 2.5 and nearly 4 standard deviations.
Going back to the 2013 Sox. That team allowed 4.04 runs per game with a standard deviation of 2.84. A shutout was about 1.4 standard deviations. The 5 high games were 15, 13, 12, 12 and 11, again between about 2.5 and 4 standard deviations.
I'm no statistician (the above is about as far as I care to go in that realm), but it seems to me that the true outliers in this case are are the upper end, where the distance from the average is 3 deviations or more. Under 1.5 just doesn't seem that far out to me.
(Note, the above uses runs per game because that data was easy to enter into a spreadsheet; I was not going to examine the box scores of each game to determine earned runs, and I doubt it would have changed the overall picture anyway)
I now leave it to you and moonslav to discuss.