Embracing The Strikeout

Rich Harden deals against the Royals on Thursday, March 4th.Rangers pitching staffs have ranked no better than 25th in strikeout rates in the past five seasons (26th in 2009, 27th in 2008, 27th in 2007, 25th in 2006, and 27th in 2005). Joey suggested in December that the team seemed to be targeting pitchers with strong strikeout rates while assembling its starting rotation for 2010. Intriguingly, the results in spring training thus far suggest that the strategy is working.

Among the seven pitchers who are likely to be with the organization at the beginning of the season who have pitched the most innings in spring training, only C.J. Wilson and Brandon McCarthy have struck out fewer than eight batters per nine innings (as illustrated in the below table):

MAJOR LEAGUE PITCHING STAFF

Overall, the seven pitchers have averaged 8.3 K/9, 3.2 BB/9, and a K/BB of 2.54. The corresponding averages for the 152 major league pitchers who have pitched at least 10 innings in spring training this year are 6.5 K/9, 2.7 BB/9, and a K/BB or 2.43. Based upon the results in spring training, the Rangers who are most likely to pitch the most innings early in the season look to be well above average when it comes to striking out opponents. 

To get an idea of how different the Rangers starters have performed this year in spring training than in years past, take a look at the table below, which shows the averages for the seven pitchers who have pitched the most spring training innings in each of the past six years. Comparing Padilla/Jennings/Gabbard/Mendoza/Hurley (2008), Padilla/Wright/Tejeda/McCarthy/Loe (2007), and Padilla/Volquez/Dominguez/Dickey/Loe (2006) to Harden/Wilson/Feldman/Harrison/Lewis (2010) is kind of like comparing round steak to tenderloin.

SPRING TRAINING AVERAGES FOR RANGERS STARTERS

Adding the innings pitched this spring by Holland, Feliz, Moscoso, Hunter, Oliver, et al. produces a slightly reduced strikeout rate, but the team’s K/BB ratio remains strong due to an improved walk rate  As with the starting pitchers, the Rangers' 2010 staff has been much more productive than previous staffs during spring training.

SPRING TRAINING AVERAGES FOR ALL RANGERS PITCHERS

But this is just spring training and everyone knows that spring training numbers are meaningless, right? In terms of win-loss records and individual performances, I would agree completely. But there is reasonable correlation between team averages for stats that measure individual performance like strikeout rate and walk rate. Take a look at the spring training (Spr.) and regular season (Reg.) strikeout (K/9) and walk (BB/9) rates from the previous five seasons in the table below. The correlation isn’t perfect, but it is certainly within the neighborhood.

SPRING TRAINING VS. REGULAR SEASON TEAM PITCHING STATS

If the Rangers' pitching staff carries its spring training performance over into the regular season, then it appears possible that the team’s strikeout rate could improve by more than one strikeout per nine innings. That would push the team into the 7.5-plus K/9 neighborhood that was occupied by the Giants, Cubs, Dodgers, Yankees, Red Sox, Marlins, and Braves in 2009. The pitching staffs for each of those teams ranked in the upper half in Major League Baseball, and included the teams ranked No. 1 (Dodgers), No. 2 (Giants), No. 3 (Braves), and No. 5 (Cubs) in ERA. The Yankees (No. 11) and Red Sox (No. 15) would have likely finished at least in the top 10 had they not played in hitters' parks with schedules weighted toward AL East opponents.      

Unsurprisingly, there is a very strong correlation between team strikeout rate and team ERA, as can be seen in the graphs from the 2008 and 2009 seasons shown below. Each point on the two graphs plots the ERA and strikeout rate for each of the 30 major league teams. The circled dots on the graphs show the Rangers' pitching performances from the past two years. The lines on the two graphs approximate the correspondence between team strikeout rates and ERA The outliers on the plot tend to be teams that either have very good or very bad defenses.

TEAM ERA VS. TEAM STRIKEOUT RATE -- REGULAR SEASON 2008 AND 2009

For teams with average defenses, an improvement of 1.0 K/9 corresponds to an ERA improvement of approximately 0.7. For a team like the Rangers with relatively pedestrian strikeout rates, emphasizing strikeouts among members of the pitching staff appears to be a very effective way to improve the team's run prevention.

Comments

[James Mason]

David: what a wonderful way to wake up today and see some *good Rangers news* for once. Thanks for the article.

If we are to believe the numbers, they also seem to show that a pitcher is not fat ull-strength during the spring and that spring batters seem to be less patient. (Good news again, all the way around for Rangers' faithful.)

[rob m]

Ironically, when the Rangers were talking "pitch to contact" they deployed a terrible defemse behind their pitchers. Now, with some pitchers that can get a few Ks, they have a better defense.

[jd21]

Wow, what a great analysis. Nice work.

[rooster]

When Nolan says he's looking for the staff to cut their ERA some more, it looks like he knows how to go about that.

And, so, the baby steps of the Rangers rotation:
2009: reduce walks, rely on defense
2010: repeat 2009, add K/9

And, some good news in regards to repeating 2009 (at least it's good news for pitchers not named McCarthy):
ST groundouts/air outs
Feldman: 3.50
CJ: 3.00
Nippert: 2.00
Harrison: 1.57
Harden: 1.10
Lewis: 0.91
McCarthy: 0.42

[jwtyler]

As a side bar what does the LA in LAIE stand for????

[Rob M.]

jw: League Average Innings Eater

[david]

Thanks for adding the GO/AO numbers, Rooster. I had that in my original draft but dropped it because I couldn't find groundout/air out rates for pitchers from previous spring trainings. Having a rotation of groundball/strikeout pitchers would be most satisfying.

[T.E. Legraf]

David, do you have the K% and BB% instead of K/9 and BB/9? I think that give us a better idea of how strikeout-happy pitchers are. For example, McCarthy has good K/9, but that's partly because he faced a lot more hitters per inning than somebody like Nippert.

[david]

TE - These numbers are imperfect because the spring training stats that I have don't include how many batters each each pitcher has faced. Assuming that there have been no double plays, then the K% and BB% for each of the seven pitchers considered in the article are:

Wilson - 18% K, 2% BB
McCarthy - 15% K, 10% BB
Harden - 20% K, 12% BB
Lewis - 25% K, 3% BB
Nippert - 26% K, 0% BB
Harrison - 28% K, 3% BB
Feldman - 20% K, 11% BB

[Dave H]

david - do you think it's encouraging that Colby Lewis's K and BB rates have compared favorably to what he posted in Japan? We all know that ST numbers won't necessarily transfer over to the regular season, but it does look like he could be at least a dependable league average starter.

[Texas_Dawg]

Good stuff.

One suggested slight adjustment for future calculations: regress the K rate against runs/game (not just earned runs).

[david]

Dave H - I am very encouraged by Lewis' spring training. We can't be sure until he's gotten 10+ starts against major leaguers under his belt, but I will be surprised if Lewis isn't at least as good as Feldman. The two have similar repertoires but it appears that Lewis might have slightly better command.

[Old Tom]

Thanks for this. Really good analysis and really good news.