Kyle Gibson story on Fangraphs

[Mike Sixel]

A nice interview with Kyle is on Fangraphs. Recommend it.

 

http://www.fangraphs.com/blogs/kyle-gibson-on-pitching-to-contact/

 

[Seth Stohs]

Glen Perkins just tweeted it. Very good stuff.

[drjim]

Wait, the Twins have an analytics department?

[drjim]

On a more serious note - I think Gibson makes an interesting point on fip and xfip. He really is the type of pitcher that breaks that model.

[HitInAPinch]

Very good article.  Put's it all together on what Gibson is doing on the mound.

I'm wondering when the author interviewed Gibson.  "The other day David Ortiz hit a hard line drive up the middle and Danny Santana was waiting on it."

Also liked the reference to "elevating pitches".  I remember when Tommy Lasorda told Orel Hershiser to start elevating his sinker.  Turned out pretty nice for Orel!

[Brock Beauchamp]

 

 

I have this debate with Glen quite a bit: How effective an out on two or three pitches is as opposed to a strikeout. For me, a soft ground ball is just as effective if it’s within the first two or three pitches. If I’m able to keep my hard contact rate low, ground balls are almost automatic outs for me.

 

...

 

If you get a ground ball right where the shortstop is usually going to be, and no one is there, you can get frustrated in the heat of the moment. But in the grand scheme of things, if he hits a ground ball there 10 percent of the time, and ninety percent of the time he hits it on the other side of second base, it’s smart to put an extra guy over there. You have to trust the process. Baseball is a process-driven game, and you have to understand that at certain times the numbers might get you, but most of the time the numbers are going to hold true.

 

...

 

I don’t go out there trying to prove analytics wrong, I’d benefit from increasing my strikeout rate, but the same time, you have to know what your strengths are. Going after strikeouts for 27 out of 27 guys isn’t my strength right now. Sure, there are times where I’d like to get more strikeouts, but 27 outs are 27 outs. Whether it’s 27 strikeouts or 27 ground balls doesn’t matter to me.

 

Great stuff.

 

Who loves Glen Perkins?

 

*enthusiastically raises hand*

[TheLeviathan]

 

Great stuff.

 

Who loves Glen Perkins?

 

*enthusiastically raises hand*

 

Agreed, the more I hear him the more I like him.  (And the more I understand how his personality offended our previous regime)

[Brock Beauchamp]

 

Agreed, the more I hear him the more I like him.  (And the more I understand how his personality offended our previous regime)

Yep. I'm not trying to diminish Jack Goin's influence but his job of getting pitchers to buy into analytics has to be easier when the team's best pitcher is such a student of the subject.

[drivlikejehu]

 

On a more serious note - I think Gibson makes an interesting point on fip and xfip. He really is the type of pitcher that breaks that model.

 

Kyle Gibson, career:

ERA - 4.50

FIP - 4.20

xFIP - 4.15

 

If he breaks the model, it would be because he's worse than the metrics predict. But to me it looks like he's just been unlucky, along with probably some questionable defensive assistance.

 

I liked what he said about not giving in.

[Brock Beauchamp]

 

Kyle Gibson, career:

ERA - 4.50

FIP - 4.20

xFIP - 4.15

 

If he breaks the model, it would be because he's worse than the metrics predict. But to me it looks like he's just been unlucky, along with probably some questionable defensive assistance.

 

I liked what he said about not giving in.

His career xFIP is *slightly* misleading because of his 2013 season. xFIP kinda breaks when a pitcher is really bad (or sometimes really good). You often see a 6+ ERA pitcher sport an xFIP of 4.5-5.0 and I think it's because the metric doesn't factor in that bad pitchers absolutely get drilled when contact is made.

 

But your overall point is spot-on. According to xFIP, Gibson was a bit unlucky last season and he's quite lucky this season. I can't fully explain this season's discrepancy (though I'll kinda meander my way to a point) but last season's "unlucky" status was almost surely due to Gibson's Jekyll & Hyde routine. Again, I think xFIP breaks when a pitcher is really bad. We've seen it time and time again with the likes of Blackburn, Pelfrey, Worley, etc. Their ERA is terrible while their xFIP is merely bad.

 

Gibson ended 2014 with a 4.47 ERA, 88 ERA+. Slightly below average but not alarmingly so... but when you dig into his game log, it tells a very different story.

 

Gibson made 31 starts in 2014.

 

Gibson gave up 5 ER or more in ten (!) of those starts.

 

Gibson gave up 1 ER or less in 13 (!) of those starts.

 

In 75% of his 2014 starts, Kyle Gibson was either stellar or horrible, bad, awful.

 

That's kinda crazy but when talking about a young pitcher, it's very promising and indicates a pitcher that will turn into a very nice piece once he establishes himself a bit (as he has this season). Will he continue to be this good? I'm not convinced of that but he should be above average.

[halfchest]

Ya know who else broke the model?  Nick Blackburn, at least for awhile. 

 

In all honesty though that's a really bad comparison.   Gibson's struck out more guys both in the minors and majors and has a much better pedigree than Blackburn ever did.  (although wasn't Blackburn our #1 prospect one year after he tore up AA?).   While I don't expect Gibson to be a 3.00-3.50 ERA guy every year or possibly any year.  I think he could settle in nicely and have a career at just under 4.00 ERA. 

 

Also, while he was almost scared of a strikeout the first 6 starts and had a backwards K/BB ratio, he's turned it around and has struck out 44 over 49 IP with a 44/13 K/BB ratio over his last 8 starts.   He's probably somewhere in between those two guys but I think he's still developing and has a very good chance to improve his K/9 numbers over last year and what they are over the first half of this year. 

[Mike Sixel]

I find the high in the zone part interesting, since I've been begging him in posts on this site to try more of those with 2 strikes on guys.....*

 

* and no, I don't think for 1 second he read any of that

[JB_Iowa]

 

 

I'm wondering when the author interviewed Gibson. 

 

Gibson talked about his approach when the Twins visited Fenway Park earlier this month.

[Otto von Ballpark]

 

His career xFIP is *slightly* misleading because of his 2013 season.

Since the start of 2014:

 

ERA 4.10

FIP 4.01

xFIP 4.03

 

The "breaking" appears to be occurring this year, and it's less than half a season so far (although I obviously hope it lasts).

[Otto von Ballpark]

 

Also, while he was almost scared of a strikeout the first 6 starts and had a backwards K/BB ratio, he's turned it around and has struck out 44 over 49 IP with a 44/13 K/BB ratio over his last 8 starts.

Gibson has definitely been trending up in that regard.  He's leapfrogged Pelf, who's K/9 has held pretty steady at 4.2.

[Brock Beauchamp]

 

Gibson has definitely been trending up in that regard.  He's leapfrogged Pelf, who's K/9 has held pretty steady at 4.2.

Yep, Gibson's K% is up to 15.2. You'd like to see it creep toward 17-18% when all is said and done but that number isn't embarrassing.

[diehardtwinsfan]

I'd note that he's really dialed the Ks up the last month or so.  Not sure if it's just the percentages playing themselves out or if he's made some adjustments... but his K rate the last few starts has been very very good.

[kab21]

Let's wait and see if he can break the model for a relevant sample size (like multiple seasons).  That isn't to suggest that there aren't players that can break the model but players have single seasons where their ERA is lucky based on previous seasons.  In Gibson's case we don't have a lot of previous MLB data to work with though. 

 

In some ways I think Gibson's rollercoaster 2014 season is in due partly to the low strikeout rate.  He puts a lot of balls in play so he is going to have some days where the balls find holes or fall in for hits leading to big innings.

 

I actually find this troubling that people are suggesting that a quote like this suggests he is a FIP/xFIP rulebreaker.

 

On succeeding with a low K-Rate: “Maybe someday someone can explain FIP and xFIP to me, and I’ll get it. It’s pretty interesting to see how analytics have worked their way into the game, and are taking a prominent role.

 

“I don’t go out there trying to prove analytics wrong, I’d benefit from increasing my strikeout rate, but the same time, you have to know what your strengths are. Going after strikeouts for 27 out of 27 guys isn’t my strength right now. Sure, there are times where I’d like to get more strikeouts, but 27 outs are 27 outs. Whether it’s 27 strikeouts or 27 ground balls doesn’t matter to me.”

 

There is nothing hard to understand about what is wrong with this quote.  27 K's are 27 outs every time unless the catcher screws up (a very small chance).  27 groundballs result in 6-7 hits (.240 BAPIP on GB's).  There is nothing in this quote that suggests he can sustain a better ERA than FIP/xFIP.

[Brock Beauchamp]

There is nothing hard to understand about what is wrong with this quote. 27 K's are 27 outs every time unless the catcher screws up (a very small chance). 27 groundballs result in 6-7 hits (.240 BAPIP on GB's). There is nothing in this quote that suggests he can sustain a better ERA than FIP/xFIP.

Is take more issue with the quote if Kyle didn't also say this, indicating he understands the nature of putting the ball in play and the risks of doing so:

 

"When hitters puts balls in play, they’re going to get a hit sooner or later. But if I get five ground balls in a row, I’m more than likely going to get out of the inning. Offensive numbers will tell you that. If they only get a hit three out of every ten times they put the ball in play, that’s less than two out of five, so I like my chances.”

[diehardtwinsfan]

yeah, ground ball pitching with Ks is the way to be an Ace.  Gibson had a nice K rate in the minors (just over 8/9 IP), so I think that as he adjusts to ML hitting you'll see his current rate climb (which it has been doing lately), but if there's a runner on first, a grounder is what I want more often than not as that turns one out into two very often... unless of course there are two outs. 

 

That said, imagine how good he'd be if he got is Krate up closer to his minor league rate.  

[kab21]

 

Is take more issue with the quote if Kyle didn't also say this, indicating he understands the nature of putting the ball in play and the risks of doing so:

"When hitters puts balls in play, they’re going to get a hit sooner or later. But if I get five ground balls in a row, I’m more than likely going to get out of the inning. Offensive numbers will tell you that. If they only get a hit three out of every ten times they put the ball in play, that’s less than two out of five, so I like my chances.”

Except Kyle doesn't have a 100% GB rate.  that's impossible.  His really good GB rate still makes it 53/47 that it is a GB or a FB.  If he is getting 5.5 K/9 then I'm not optimistic that he is ever going to be better than a 4-4.25 ERA guy and that guy has an unlucky season with a 5+ ERA every once in awhile. Hopefully he continues to K more like he has in June because then things get interesting.

[gil4]

 

(although wasn't Blackburn our #1 prospect one year after he tore up AA?).  

I think that says more about the one list that had him #1 and about the prospects the Twins had a the time than it says about Blackburn.

 

I did find this on Gleeman's site - a May 2013 article "Revisiting the best Twins prospects of the 2000s" 

 

I disagreed so much with Baseball America ranking Nick Blackburn as the Twins' top prospect in 2008 that I made a bet with the magazine's editor, John Manuel, that Blackburn wouldn't get 70 career wins. I'm feeling pretty safe about the bet now with Blackburn stuck on 43 wins and his career at a crossroads, although in retrospect he did turn out better than I expected even if 819 innings of a 4.85 ERA is nothing special.