Last time, I wrote about the weak correlation between ERA and a pitcher's W-L record. To illustrate this further, I decided to compare this correlation with some other correlations.

IP/GS is innings per start, K/GS is strikeouts per start, and WHIP is baserunners per inning.

Correlations With ERA
Group	Winning %	IP/GS	K/GS	WHIP
All Pitchers	-0.36	-0.62	-0.60	0.81
3.50+ ERA	-0.13	-0.51	-0.32	0.67

Relative to these other stats, winning percentage has a weak correlation with ERA.

Here's another table, similar to the one in my previous post.

ERA	Winning %	IP/GS	K/GS	WHIP
0.00 - 3.00	0.627	6.8	7.4	1.05
3.00 - 3.50	0.640	6.6	5.1	1.20
3.50 - 4.00	0.528	6.4	4.9	1.29
4.00 - 4.50	0.555	6.4	4.1	1.35
4.50 - 5.00	0.511	6.2	3.8	1.40
5.00 - 5.50	0.495	5.8	3.7	1.49
5.50+	0.500	5.8	4.0	1.56

Look at the trends. IP/GS, K/GS, and WHIP all have consistent increases or decreases when compared to ERA (except for the last K/GS row). Winning percentage clearly doesn't.