/cdn.vox-cdn.com/uploads/chorus_image/image/45231542/usa-today-8186146.0.jpg)
Wednesday brought some ground breaking news to the baseball world: The pitch clock is coming. At least to Double-A and Triple-A in the short term, but this may be the first step to the clock making its way to the big leagues.
Based on how the rule reads, and how the rule was carried out in the Arizona Fall League, pitchers will have 20 seconds to deliver the ball, with the clock stopping when his motion is started. If he has the ball for more than the allotted time without starting his motion, a penalty Ball will be called.
So what does this mean for the Rays pitchers, if anything?
I wanted to determine if the clock would have any effect on any of the Rays’ vital pitching statistics. My list included pitchers currently on the 40 man roster, and I excluded those that threw less than 20 innings in 2014 (with the exception of Matt Moore) to avoid any sample size issues. Moore was included using 2013 data, as he will likely be making a June return to the rotation. Other than Moore, 2014 statistics were used for all involved to get the most recent sample.
As a group, the Rays pitching staff is averaging 25.2 seconds between pitches, with starters averaging 24.1 seconds and relievers 26.1 seconds. Using the same qualifications (20 IP in 2014), the major league average for both SP and RP sat at about 23.5 seconds, with qualified starters at 22.3 seconds, and relievers around 24.5 seconds.
For the Rays, while it seems the relievers are taking a longer time, the starters are the real culprits of the team’s excess time on the mound, as they are waiting about 8% longer than league average, comparable to the relievers' 6.5%. Highs and lows for the staff were Jeff Beliveau at 29.3 seconds, and Drew Smyly at 21.2. Even the Rays' quickest pitcher is slower than what is allotted by the pitch clock.
Now to the fun stuff. In terms of player performance, the process I used to figure out what the clock could affect led to potentially one key variable.
I tried to correlate what was vital (K/9, BB/9, HR/9, FIP, xFIP), as well as variables that could be affected by the clock (Swing%, Contact%, Zone%), to the group’s 2014 pace. I also took an average of the staff’s pace to get an idea of what the group is bringing as a whole. Samples of the whole staff, starters, and relievers, were all tested separately to get looks at all angles.
The table below shows how well the entire Rays staff’s pace correlated with the previously mentioned statistics. Closer to -1 shows a stronger negative correlation, and vice versa for closer to +1. The word "negative" has a rough connotation, but it simply suggests that as one variable goes up the other will go down, whereas for a positive correlation, both variables would increase or decrease together.
For example, if strikeout rate were to yield a high negative correlation to a pitcher's pace (let's say -0.55), it would suggest that as the pitcher takes more time between pitches, strikeout rate would decrease. A strong positive correlation (let's say 0.75) would tell us that as time between pitches increased or decreased, strikeout rate would increase or decrease with it. Which proves to be true?
Let's go to the data, starting with the Rays' entire staff together (both starters and relievers).
K/9 |
BB/9 |
HR/9 |
BABIP |
Swing% |
Contact% |
Zone% |
FIP |
xFIP
|
-0.16835
|
0.128239
|
-0.48844
|
-0.00283
|
-0.09483
|
0.290392
|
0.275372
|
-0.28586
|
0.075409
|
As you can see, the only remotely significant coefficient is that of HR/9 (and this will become a theme). The other numbers coefficients were too close to zero, meaning there was little to no correlation.
Logically, it is difficult to connect a quicker pace to a higher homerun rate. With a quicker pace, mechanics may not be as crisp, but you would expect other key areas, specifically Zone%, Contact%, and BB/9 to be affected as well if the pitcher’s mechanics slipped a bit.
As we move on, you'll notice a somewhat drastic change to the K/9 correlation. This may be due to the separation of the starters' and relievers' pace when they are grouped together as a whole. Four of the six starters that were tested fall within the quickest five pitchers tested, while four of the five slowest pitchers tested were relievers. The separation of the data will alter the slope of the correlation line in comparison to data points that group together.
The results limited to the starters are below.
K/9 |
BB/9 |
HR/9 |
BABIP |
Swing% |
Contact% |
Zone% |
FIP |
xFIP
|
-0.81951
|
0.264273
|
-0.92173
|
-0.45049
|
0.015215
|
0.610869
|
-0.10358
|
-0.12868
|
0.5241
|
We see a little bit stronger of a case here for a few things, specifically K/9, Contact%, and xFIP.
Despite not having the strongest correlation for this set, K/9 comes in with a very strong result, suggesting that pace and strikeout rate could very well be connected, and logic would dictate this makes some sense.
Again, it’s hard to logically connect the two, but for HR/9 to yield such a strong coefficient for the starters, there seems to be a highly observable connection.
The result for Contact% may have been somewhat expected, as well as the strikeout rate for that matter (although maybe not that strong for K/9). As a pitcher takes more time in between pitches, the batter has more time to settle in and prepare himself for the coming pitch, likely resulting in higher contact rates, and lower strikeout rates.
The correlation we see to xFIP is rather interesting, with strikeouts being a component of xFIP and the statistics going in opposite directions as far as the type of correlation they show, but the result is logical. Rays starters that took less time between pitches had higher strikeout rates and lower xFIP.
And finally, the data for the relievers.
K/9 |
BB/9 |
HR/9 |
BABIP |
Swing% |
Contact% |
Zone% |
FIP |
xFIP
|
-0.62702
|
0.066089 |
-0.5988 |
0.16753 |
-0.16696 |
0.587349 |
0.05088 |
-0.30667 |
0.278769 |
K/9 emerges as the strongest correlation to the Rays’ relievers pace, although, as with the other sets, HR/9 and Contact% still look fairly strong.
As noted above, timing means a lot to a hitter, so it makes some more logical sense that a longer pace would give hitters time to adjust themselves and focus, thus leading to a lower K/9, and a higher Contact%, rather than the correlation seen to HR/9.
Just for a better visual, the following table lists the Rays pitchers, from the longest pace to the shortest, including league average on the bottom:
Jeff Beliveau |
29.3 |
Alex Colome |
28.7 |
Kevin Jepsen |
27.7 |
Grant Balfour |
27.0 |
Jake McGee |
26.4 |
Chris Archer |
25.2 |
Ernesto Frieri |
25.0 |
Kirby Yates |
24.8 |
Alex Cobb |
24.5 |
Brad Boxberger |
22.8 |
Matt Moore |
22.6 |
Jake Odorizzi |
22.6 |
Drew Smyly |
21.2 |
League Average |
23.5 |
Conclusion
So to answer the original question, the pitch clock shouldn't mean too much for the Rays’ pitchers. Chris Archer weighed in on his pace last season, suggesting that the amount of time he takes (highest for a full time starter, as seen above), is simply a product of what the situation or game calls for.
As you might be able to tell, Archer doesn't seem exactly pleased with the idea that someone wants to control the pace of his game.
Correlation does not always equal cause. But the clock and pitcher’s pace only really seem to be having an effect on HR/9 and K/9. It’s something new.
Adjustments will have to be made, but at least through this analysis, the Rays won’t have too much to worry about when the pitch clock makes its way to the majors.
40 man roster information was taken from the Rays' website, with any statistical information (other than the actual analysis) coming from Fangraphs.