Strikeouts and Walks
“Strikeouts and walks? After spending weeks talking about things like BABIP and LOB%, you’re going to talk about strikeouts and walks?” Absolutely. If, this week, I do nothing more than stress to you how important these two categories are, then I’ll be satisfied.
Strikeouts and walks, hands down, are the two most important isolated stats to consider when evaluating pitchers. If you look at nothing else when doing your evaluations, look at these two stats. We touched on them a bit when we discussed DIPS Theory, but they are so important that they deserve their own article. Aside from strikeouts being a fantasy category unto themselves in the majority of leagues, strikeouts and walks have a significant impact on several other fantasy categories.
First and foremost, a pitcher has almost complete control over his strikeouts and walks. The batter, obviously, has some control as well, but the pitcher could accomplish a strikeout or walk without a single defensive player behind him and sans any luck. This makes strikeouts and walks a couple of the most predictable stats for pitchers. Because raw strikeouts and walks vary from year-to-year due to pitchers throwing different numbers of innings, we need to examine them as ratios. When I evaluate strikeouts and walks, I use the following, simple formulas:
Obviously, the higher the K/9, the better, and the lower the BB/9, the better. To roughly measure their combined importance, we can use another statistic.
Strikeout-to-walk ratio, or K/BB ratio, is an excellent statistic for predicting pitcher ERAs. We can run some tests to see just how effective it is. We’ll use data from 2004-2007 and include all pitchers with at least 50 innings pitched.
Correlation Coefficient: -0.48
R2: 0.23
Adjusted R2: 0.23
P-value: 9.59E-74
Level of Significance: 1%
These tests show that there is a negative relationship between ERA and K/BB. In other words, as K/BB increases, ERA decreases – just as we would expect. While a 0.23 R2 isn’t amazing, when you consider how much variability there is with ERA and how difficult it is to predict, being able to predict 23% of ERA movement with this one simple statistic is actually a very good result. The p-value and level of significance further show that these tests are highly significant.
To make it even easier to understand how important K/BB ratio is in predicting ERA, let’s look at the top dozen best and worst Starting Pitcher K/BB ratios from 2007 and each pitcher’s corresponding ERA.
As you can see, 11 of the 12 players on the ‘best’ list have ERAs under 4.00. On the ‘worst’ list, though, just 2 players have ERAs under 4.00 and 6 players have ERAs over 5.00.
Of course, K/BB ratio isn’t the only component of ERA, as we can see some guys who don’t quite fit in with the rest. For example, Tim Redding – with an awful 1.24 K/BB – had a 3.64 ERA in 2007. As I’ve said many times, though, ERA is highly subject to luck-related swings, which is why we’ve spent the past few weeks explaining some of these indicators of luck.
Looking deeper into Redding, we can see that he had a .285 BABIP, 9.52 HR/FB, and 82% LOB%, all of which are above-average and should regress in 2008.
Still, on the whole, you can easily see how powerful K/BB ratio is for evaluating pitchers. Just be careful that the K/BB ratio isn’t artificially inflated, or isn’t in danger of falling off. This is far less common than it is for stats like BABIP or LOB%, but it still can happen. For example, look at Joe Blanton’s stats over the past three years:
That 1.57 BB/9 is very low, and only a finite number of pitchers are able to maintain walk rates that low. It is certainly possible Blanton will be able to maintain it, but the possibility that his BB/9 will rise above 2.00 in 2008 needs to be considered and accounted for.
The K/BB of a guy like Jake Peavy is much more stable, as it is supported by years of similar walk rates and a high K/9.
As always, the message I’m trying to get across to you is to do your homework thoroughly, examining all components of a pitcher’s performance before making a decision.
Strikeouts and walks, hands down, are the two most important isolated stats to consider when evaluating pitchers. If you look at nothing else when doing your evaluations, look at these two stats. We touched on them a bit when we discussed DIPS Theory, but they are so important that they deserve their own article. Aside from strikeouts being a fantasy category unto themselves in the majority of leagues, strikeouts and walks have a significant impact on several other fantasy categories.
First and foremost, a pitcher has almost complete control over his strikeouts and walks. The batter, obviously, has some control as well, but the pitcher could accomplish a strikeout or walk without a single defensive player behind him and sans any luck. This makes strikeouts and walks a couple of the most predictable stats for pitchers. Because raw strikeouts and walks vary from year-to-year due to pitchers throwing different numbers of innings, we need to examine them as ratios. When I evaluate strikeouts and walks, I use the following, simple formulas:
Obviously, the higher the K/9, the better, and the lower the BB/9, the better. To roughly measure their combined importance, we can use another statistic.
Strikeout-to-walk ratio, or K/BB ratio, is an excellent statistic for predicting pitcher ERAs. We can run some tests to see just how effective it is. We’ll use data from 2004-2007 and include all pitchers with at least 50 innings pitched.
Correlation Coefficient: -0.48
R2: 0.23
Adjusted R2: 0.23
P-value: 9.59E-74
Level of Significance: 1%
These tests show that there is a negative relationship between ERA and K/BB. In other words, as K/BB increases, ERA decreases – just as we would expect. While a 0.23 R2 isn’t amazing, when you consider how much variability there is with ERA and how difficult it is to predict, being able to predict 23% of ERA movement with this one simple statistic is actually a very good result. The p-value and level of significance further show that these tests are highly significant.
To make it even easier to understand how important K/BB ratio is in predicting ERA, let’s look at the top dozen best and worst Starting Pitcher K/BB ratios from 2007 and each pitcher’s corresponding ERA.
As you can see, 11 of the 12 players on the ‘best’ list have ERAs under 4.00. On the ‘worst’ list, though, just 2 players have ERAs under 4.00 and 6 players have ERAs over 5.00.
Of course, K/BB ratio isn’t the only component of ERA, as we can see some guys who don’t quite fit in with the rest. For example, Tim Redding – with an awful 1.24 K/BB – had a 3.64 ERA in 2007. As I’ve said many times, though, ERA is highly subject to luck-related swings, which is why we’ve spent the past few weeks explaining some of these indicators of luck.
Looking deeper into Redding, we can see that he had a .285 BABIP, 9.52 HR/FB, and 82% LOB%, all of which are above-average and should regress in 2008.
Still, on the whole, you can easily see how powerful K/BB ratio is for evaluating pitchers. Just be careful that the K/BB ratio isn’t artificially inflated, or isn’t in danger of falling off. This is far less common than it is for stats like BABIP or LOB%, but it still can happen. For example, look at Joe Blanton’s stats over the past three years:
That 1.57 BB/9 is very low, and only a finite number of pitchers are able to maintain walk rates that low. It is certainly possible Blanton will be able to maintain it, but the possibility that his BB/9 will rise above 2.00 in 2008 needs to be considered and accounted for.
The K/BB of a guy like Jake Peavy is much more stable, as it is supported by years of similar walk rates and a high K/9.
As always, the message I’m trying to get across to you is to do your homework thoroughly, examining all components of a pitcher’s performance before making a decision.
0 Comments:
Post a Comment
<< Home