K/BB ratios: Does it matter how a pitcher does it?

We all know that K/BB ratio is a good skill indicator for pitchers. It has some small flaws, namely that strikeouts and walks are weighted equally despite having different effects on a pitcher’s ERA. Still, it measures two of the greatest skills a pitcher can possess. I’ve always been curious, however, as to whether or not it matters how a pitcher achieves a good K/BB.

Obviously the 4.52 K/BB (supported by both a strong K/9 and BB/9) of a guy like Johan Santana is excellent, but is the 4.16 K/BB of control pitcher Greg Maddux better than the 2.31 K/BB of power pitcher Tim Lincecum? Or is it better to be like Roy Oswalt and post a 2.57 K/BB without excelling at strikeouts or walks?

Well, I ran a few numbers and found that there does seem to be a noticeable ERA difference among pitchers with good K/BB rates but differing skill sets.

While this might not be the most scientific way of examining it, I think that this should provide a clear, visual depiction of the difference. For the purpose of this exercise, I consider a K/BB over 2.00 to be ‘good’. In actuality, it is league average, but I consider a 2.00 K/BB the cutoff for draftable pitchers except in the deepest of fantasy leagues, so that’s what I used.

National League

Low strikeout pitchers
Qualifications: 2.00+ K/BB, 5.00- K/9, 50+ IP

Year	Qualified	IP	ERA	K/BB	GB%
2007	10	1159.0	4.26	2.55	49.15
2006	10	1170.3	4.30	2.53	46.31
2005	10	1262.3	4.25	2.56	44.34
2004	7	834.0	3.94	2.71	46.86
Total	37	4425.7	4.21	2.58	46.59

Solid in both
Qualifications: 2.00+ K/BB, 2.75+ BB/9, 7.00- K/9, 50+ IP

Year	Qualified	IP	ERA	K/BB	GB%
2007	9	1017.7	4.46	2.06	37.64
2006	9	943.0	4.34	2.17	43.39
2005	9	992.0	4.25	2.19	36.49
2004	8	1146.3	4.11	2.18	41.91
Total	35	4099.0	4.28	2.15	39.88

High walk pitchers
Qualifications: 2.00+ K/BB, 3.50+ BB/9, 50+ IP

Year	Qualified	IP	ERA	K/BB	GB%
2007	17	1881.3	3.57	2.32	37.21
2006	20	1842.3	4.24	2.28	37.68
2005	10	808.0	3.66	2.37	39.11
2004	9	913.0	3.81	2.51	40.44
Total	56	5444.7	3.85	2.35	38.19

As you can see, the ‘high walk pitchers’ have the best ERAs. Not only that, they have lower overall K/BB rates and lower ground ball rates. If we were do this more scientifically and account for these things, the ERA gap would presumably be even larger.

The ‘solid in both’ and ‘low-strikeout pitchers’ have very similar ERAs. The ‘solid in both’ guys, though, have lower K/BB rates. This should be no suprise, though, given the the cutoff points used. A pitcher who is barely ‘merely solid’ in both categories (7.0 K/9, 2.75 BB/9) would have a K/BB of just 2.54, which is less than the average K/BB of the ‘low-strikeout’ pitchers. Still, despite the lower K/BB rates and lower ground ball rates, the ‘solid in both’ pitchers manage to put up essentially the same ERAs.

American League

Low strikeout pitchers
Qualifications: 2.00+ K/BB, 5.00- K/9, 50+ IP

Year	Qualified	IP	ERA	K/BB	GB%
2007	8	1270.3	4.50	2.73	41.82
2006	11	1503.0	4.80	2.51	43.89
2005	15	2419.3	4.18	2.92	43.42
2004	7	962.7	4.22	2.54	48.45
Total	41	6155.3	4.40	2.71	43.99

Solid in both
Qualifications: 2.00+ K/BB, 2.75+ BB/9, 7.00- K/9, 50+ IP

Year	Qualified	IP	ERA	K/BB	GB%
2007	6	730.3	4.49	2.18	43.24
2006	7	1144.7	4.68	2.12	40.99
2005	6	862.0	4.13	2.10	41.21
2004	9	1176.3	4.54	2.15	40.40
Total	28	3913.3	4.48	2.14	41.29

High walk pitchers
Qualifications: 2.00+ K/BB, 3.50+ BB/9, 50+ IP

Year	Qualified	IP	ERA	K/BB	GB%
2007	14	1310.3	4.17	2.43	42.24
2006	11	741.3	3.91	2.33	40.38
2005	8	657.3	3.22	2.39	42.82
2004	11	1014.7	3.42	2.52	41.80
Total	44	3723.7	3.75	2.42	41.85

The ‘low strikeout pitchers’ tend to do worse in the AL than the NL despite a higher K/BB. Goes to show how much more difficult it is to pitch in the AL. Overall, though, we see very similar results. A little more pronounced, but our judgments should be the same. Pitchers who achieve their K/BB rates using pinpoint control and poor strikeout rates are not nearly as good of a bet as those who achieve their K/BB rates by getting a high number of strikeouts, even if they have below-average control.

As a small caveat, it should be noted that a pitcher with good control and a not-so-good strikeout rate can still be successful if he also manages to induce a lot of ground balls. Take a look at these two tables that I posted in an article for MLB Front Office last week. The numbers in yellow are the overall ERAs of pitchers with at least 50 innings pitched (using data from 2004-2007).

GB% Greater Than 50%   K/9 5.0- 6.5+ BB/9 2.25- 3.83 2.74 3.0+ 4.53 3.87

GB% Less Than 40%   K/9 5.0- 6.5+ BB/9 2.25- 4.83 3.36 3.0+ 5.48 4.42

Concluding thoughts

Our overall impression from this article: All K/BB rates should not be treated equally.

Now, of course, most of this isn’t all that earth-shattering, but I do think that this stuff is interesting and important to note. You’re probably already going after guys with high K/BB ratios and high K/9 if for no other reason than because strikeouts are a category in-and-of themselves in most fantasy leagues. This just lends more support to this approach. In addition, it further emphasizes the need to look at the complete package of a pitcher, not just one stat.

If you’re looking for some more stuff on strikeouts and walks, try checking out the team pages in the 2008 Hardball Times Baseball Annual. In the tables that list the players, you can see the relative run impact of every pitcher’s non-batted balls. In other words, you can see the relative runs given up/saved by every pitcher’s strikeouts and walks. Interesting stuff.