So on a hunch after the loss today, I did some quick calculations to find out how the run support AJ receives relates to how well he pitches. To do this I divided his starts into "Good AJ" starts and "Bad AJ" starts, and calculated his ERA in those starts (I'm sure someone with more sabermetric expertise could find a better statistic to use, but this is what I'm going with). I also calculated his "RSA" (Run Support Averages) for those two groups. I'm not sure if that's an actual statistic, but I calculated it the same as ERA, only using the runs scored by the Yankees in those starts:
RSA= (Run Support/Innings Pitched)*9
Of AJ's 24 starts, I placed 9 in the "Bad" category because he either went less than 6 innings with 3 or more runs allowed or went 6 or more innings with 5 or more runs allowed.
15 fell in the "Good" category because he either went less than 6 innings with no runs allowed or went 6 or more innings with 4 or less runs allowed.
Results (as of the game on 8/15):
Bad AJ has an ERA of 12.46 with RSA of 4.38.
Good AJ has an ERA of 1.68 with RSA of 5.56.
At face value, AJ receives over 1 run more run support when he is pitching well than he does when he is pitching badly, which make sense.
Digging a little deeper, Good AJ received more run support than he allowed runs in 10 games, he received the same amount of run support as he allowed runs in 2 games, and received less run support in 2 games.
By contrast, Bad AJ received as much run support as he allowed runs in only 1 game, but he received less run support in the other 8 games. Now this is probably because of the small sample size since Bad AJ (by my calculation) has only pitched 39 innings, and went 5 or less innings in 7 of the 9 games, giving the offense less time to pick him up.
So what conclusions should we draw from this? To state the obvious, Good AJ wins games; Bad AJ does not. Good AJ also receives run support; Bad AJ does not. Most importantly, Good AJ's ERA is phenomenal compared to that of Bad AJ, and he shows up almost twice as often. Finally, AJ is in fact a good pitcher about 2/3 of the time; it just feels like he's awful all the time.