The problem is, there doesnít seem to be any middle ground when it comes to Gibsonís starts. Either heís lights out, like he was in Seattle (6.0IP, 7H, 0ER, 1BB, 3K) or heís knocked out, like his outing against the Yankees on July 4th (2.0 IP, 6H, 5ER, 1BB, 0K).
One or two starts with such a variance wouldnít mean much. Starters have bad outings, as even staff ace Phil Hughes has thrown a clunker or two this season. Gibson, however, isnít just having one or two all or nothing starts. Seemingly every start is defined by the all-or-nothing principal.
Take a look at this years' splits:
(A blank space in the charts below indicates a value of zero).
|in No Dec.||3||71||68||10||3||2||1||3||13||4.33||.147||.183||.191||.374||13||2||1||.182||14||9|
Thatís a startling difference. Heís essentially allowing opposing batters to hit 170 points higher in average in losses while also seeing opponents' OPS double (!) Obviously, in a loss, starters are going to post worse numbers than they do in wins, but the splits are not typically this wide or dramatic.
Initially it looked like Gibsonís struggles were attributable to home/road splits. In his home starts since the beginning of the season, Gibson has had only two "clunkers"
His road starts have been a bit rougher:
Early in the season, the solution seemed simple. For whatever reason, Kyle Gibson struggled away from Target Field. However, Gibsonís last few starts have made that assessment inaccurate. Heís turned in quality outings in Texas, Boston and Detroit while his last dud came at home.
Itís clearly not as simple as home/road struggles. So, letís dig a bit deeper. Maybe this is all just a matter of pitch selection?
That big block of data does have some interesting points. Mainly, in wins, Gibson is using his fastball between 60%-70% of the time while in losses the fastball usage resides at 51%-57%. Why is Gibson shying away from his fastball in losses? It looks like heís struggling to locate the pitch.
Hereís Gibsonís Pitch F/X data from his 7/4 start against the Yankees:
Look at the gray squares, which indicate Gibsonís sinker. As you can see, Gibson struggled to locate his sinker, leaving many balls out of the zone. That caused him to fall behind and then have to come into the zone with a different pitch (since his sinker control was off) Ė thus resulting in a fireworks performance courtesy of the Yankee bats.
The trend repeats in his 6/24 start against the Angels.
Again, thereís little consistency in the location of his sinker and not surprisingly the results are very similar to the start against the Yankees.
Finally, look at the Pitch F/X data from his 6/29 start against Texas and the start yesterday against Seattle.
A majority of his sinkers are low in the zone and are grouped nicely. Greater control yielded a better result.
Admittedly, thatís a lot of data to comb over to simply reach the conclusion that Gibson is better when he can control his pitches. That policy applies to every starter in the league, outside of maybe Sam Deduno who in fact may be better when he has no idea where the ball is going. Whatís startling is just how different the results are when Gibson is struggling with control and when heís on.
Every starter will battle control issues from time to time and even the best starters get knocked around a bit, but I donít know that Iíve ever seen a player as all or nothing as Kyle Gibson. Hopefully, Gibson can improve on these results as he continues to learn and grow. Heíll need to learn how to pitch when his best stuff just isnít with him Ė even Kevin Correia and Ricky Nolasco can turn in quality outings when their command is evading them.
It seems like Gibsonís struggles may simply be because he hasnít figured out how to work with what he has on any given night. We hope that as he pitches further into the season, heíll figure things out and weíll stop seeing so many "boom or busts" starts from the Twinsí righty.