What the Cubs Might See in Tyler Chatwood
Right-hander Tyler Chatwood signed with the Cubs today for three years and around $40 million, according to Jon Morosi. The contract is about what one might expect. Dave Cameron, for example, called for Chatwood to receive $10 million a year for three years. The Cubs have given him more annually than Cameron expected. But for one of the youngest pitchers on the market, it’s not absurd.
But there’s also another reason for optimism regarding Chatwood’s near future besides just his relative youth. Given the tools at our disposal, there appears to be evidence that Chatwood’s stuff hasn’t fully translated into results.
Using what we know about the relationship of movement and velocity to outcomes, we can rank each of Chatwood’s pitches against league average. Below are his percentiles. Higher is better in each case, and his strengths are highlighted.
Pitch | Spin Rate | Horizontal Move | Vertical Move | Velo |
---|---|---|---|---|
Four-Seam | 93 | 51 | 86 | |
Curve | 96 | 33 | 83 | 59 |
Sinker | 8 | 22 | 88 | |
Change | 38 | 38 | 69 | |
Slider | 29 | 96 |
You’ve heard of “spin-rate guys,” right? Well, Chatwood is absolutely a spin-rate guy. What’s interesting, though, is that he hasn’t converted that high spin into plus movement. Why? Well, it might have something to do with useful spin. Over time, Chatwood has dropped his arm slot to get more movement on his sinker and more ground balls, probably because he pitched in Coors. That robs his fastball of ride, though, and his curveball of downward movement.
An easy fix might be to just throw the curveball more. He only threw it 11% of the time in 2017. It got over 70% ground balls and above-average whiffs. Batters had a .164 slugging percentage against it last year. And that fits with the spin and movement on the pitch.
A tougher fix may be for him to raise his arm slot, thus converting more of that spin into movement on the fastball and curve, a little like what Rich Hill did. Is it simple? Depends on the pitcher and his willingness to experiment. When I last spoke with Chatwood, he was willing to talk spin and was using his four-seam more to “mirror” his curveball spin. That sounds a lot like what Hill said after he broke out. In fact, I’ve included only his second-half percentiles above because he altered his pitching mix after our conversation and improved his changeup.
There’s a real potential for Chatwood to be more of an over-the-top, four-seamer/curveball/slider guy going forward. You see it in his strengths — velocity and spin — and also in his home/away splits. No, not the results-based ones. Coors affects movement, too. Consider how much better Chatwood’s pitches look when you use his second-half away movement and avoid that Coors effect.
Pitch | Horizontal Move | Vertical Move |
---|---|---|
Four-Seam | 35 | |
Curve | 44 | 94 |
Sinker | 23 | 34 |
Change | 59 | 44 |
Slider | 27 |
Everything looks better. He gains a half-inch to an inch on all of his pitches, all in the right directions, and a whopping three extra inches of drop on his curveball. Coors hates curves in particular.
Using these tables as a guide, we can summarize the present version of Chatwood as follows. He’s a high-velocity, high-spin four-seam guy with a plus curveball by spin, movement, and velocity. He’s gotten good grounders from a high-velocity, meh-movement sinker in the past. His high-velocity slider has gotten him whiffs. In the second half last year, his changeup looked average based on average movement away from home and above-average velocity gap.
Here’s the shorter version: above-average four-seamer with upside to go plus with arm slot change, average two-seamer, plus curveball, above-average slider, changeup that flashes average or better.
And the even shorter one: average major-league starter with some upside beyond, even at 27, with nearly 650 innings under his belt.
You don’t need to know his away ERA to think that this is a pitcher who’s worth his current contract and has upside beyond.
With a phone full of pictures of pitchers' fingers, strange beers, and his two toddler sons, Eno Sarris can be found at the ballpark or a brewery most days. Read him here, writing about the A's or Giants at The Athletic, or about beer at October. Follow him on Twitter @enosarris if you can handle the sandwiches and inanity.
A number of caveats:
1. He has never thrown more than 160 innings
2. He still walks a lot of guys
3. He has has not 1, but 2 TJ surgeries.
There is more risk there than some want to let on.
Crappy Counterpoints:
1. Good point
2. Contreras throws down–BAM!–JAVY TAG TO THE FACE
3. I feel like I read in Passan’s THE ARM that post-TJ guys are generally a-OK for 5 or 6 years, so maybe Chatwood is double aa-OK
So any piece that doesn’t mention those three things is somehow hiding the risk?
Maybe it’s because that information isn’t exactly a secret, and Eno (and others) are giving their audience the benefit of the doubt of already being aware of it from the numerable other places that have mentioned it, and would rather spend their time writing about what the Cubs are hoping for here, which, if you’ll scan up, you’ll notice is more or less the title of the piece to begin with.
Fangraphs just doesn’t write articles critical of mediocre free agent signings anymore.
Yeah I said the same thing this morning. I need to spend a lot more time reading about mediocre players signed to mediocre/bad contracts. Do you have a newsletter I could subscribe to?
there are like three top of the line free agents and we are writing about Ohtani and Stanton like once a day, forgive me for being excited about ACTUAL NEWS.
Forgiven!
But hate the thumbs options. You know I/we enjoy your stuff. But many would give you a thumbs up for penning an article about avocados. Still…..keep em coming.
I didn’t say Eno ignored the risks, just saying that there are reasons to be skeptical that he can reach that upside. Look at the history of 2 TJ pitchers. Not much to be excited about.
No, no, I did! Because I wrote this piece and not “why 29 other teams didn’t sign Chatwood” haha
Hello,
It’s a good piece, useful and informative. A writer doesn’t not write about something because he believes the reader already knows the information. We all love Eno’s stuff. But I think he missed this. Still…..keep em coming Eno. Enjoy them all.
His BB% is twice as high (20.2%) with the fastball as the next nearest pitch (sinker at 10.4%). His BB% with the curve is 4.5%.
There is a real chance that simply by having him throw the curve twice as often that that alone would fix his walk rate, even before any other mechanics changes.
Like most other pitchers, Chatwood probably throws the fastball on a 3-ball count. The reason his walk rate is so low on curveballs is because he likely rarely throws them unless it’s a 3-2 count to a dangerous hitter.
As a Cardinals fan, I am thrilled that the Cubs are paying a bunch of money for a mediocre pitcher. I’ll face a 17% strikeout rate and 12% walk rate guy any day.
I’m trying to understand exactly what this stat says, especially because I don’t see it on his Fangraphs page. Is it saying that 20.2% of PA’s that end with him throwing a fastball are walks? Or is it saying what percent of each pitch that he throws is taken for a called ball?
The math doesn’t seem to add up for the second one, because Chatwood’s Fangraphs page says that a bit over 40% of the pitches that he throws are called balls, both for 2017 and his career.
I think that what you’re quoting is therefore the former. If it’s limited to how PA’s end, it’s a stat that has big selection bias and mainly shows that Chatwood throws a lot of fastballs on 3-ball counts.
i wanted the yankees to sign him, just so he could throw 30-40% curveballs
He’s a slightly worse than 1.5 K/BB guy over his career, whether at Coors or away. His recent success with a low road ERA is BABIP driven- he had 60 Ks and 42 BBs last year on the road but a .197 Opp avg in 77 IP. Similar stuff but slightly better K:BB rate the season prior, .196 road Opp Avg, but 61Ks to 36 BBs in 80 road IP.
It’s not like he is striking people out, avoid walks or even consistently limiting HRs on the road (10 home and 10 away last year, although to be fair an 11 home and 4 away split the season prior).
He’s a guy who will turn 28 in a week, and ran a great BABIP twice over 80 inning stretches on the road. In his last two seasons, he’s had a .217 and .227 road BABIP. In his 3 colorado seasons before that, he posted .327, .296, and .324 road BABIP. He also ran strand rates of 84.3% and 77.8% on the road the past two seasons, compared to 68.6%, 78.4%, 64.1% in the 93.2 road IP in his 3 prior colorado season. He simply got lucky on the road the past 2 years.
https://www.fangraphs.com/splitstool.aspx?playerid=4338&position=P&splitArr=10&strgroup=season&type=2&startDate=all&endDate=all&filter=
He has a career 5.1 WAR in 647 IPs, and projects for 1.6 next year- which is a touch better than market value on the inflated FA market (i.e. $8.5M/WAR), but above the actual max value of an added win ($6.5). He has had 2 TJs as mentioned, and projects to have the Cubs’ 4th best WAR/200 IP (2.3), beating out Dallas Beeler by a bit (1.9). The spin rate is interesting but doesn’t highly correlate with success.
IMO it’s not a bad signing but the article does remind me of the ‘Why the Braves Needed to Sign Sean Rodriguez’ article from last offseason
https://www.fangraphs.com/blogs/why-the-braves-needed-to-sign-sean-rodriguez/
cherry picking an uncontextualized micro-trend in a sample where the data hasn’t stabilized, and projecting upside despite having full advantage of the more predictive models that predict mediocrity.
Swartz showed that a win was over $8 million this year, his contract is therefore fine by projections, and I pointed out that there might be a little upside based on the best research we have on how velocity and movement can relate to outcomes. Data has stabilized? Movement, velo, and spin stabilize in wicked fast. https://www.fangraphs.com/blogs/which-pitcher-stats-have-relevance-this-early-with-a-note-on-clayton-kershaw/
Eno, I don’t think we (not just you, all of us) are interpreting that $/WAR number quite right. That’s what it actually was, not what teams wanted. Teams are trying to get much better deals than $10 million/WAR, it’s just that FA turns out worse than teams expect.
I used eight per win and his projections and hit $38.5 on the calculator.
I agree that $8-9m is the free agent cost of 1 WAR- with most of that value coming through high-end and low-end contracts, the worst being the mid-tier.
The value of an added win, however, maxes out around $6.5m on the win curve.
IMO your r-squared threshold for stabilization rate is too low and doesn’t produce predictive value. Further, the link you provided doesn’t show curves as they have less consistent movement.
Chatwood’s curveball vertical movement- the main thing you are saying suggests upside, was all over the place last year- and had substantially less movement as the season went on:
Month 4seam Sinker Change Slider Curve
3/17 10.27 8.10 5.49 2.61 -9.67
4/17 9.45 8.16 4.63 2.68 -7.83
5/17 8.81 7.06 4.72 2.64 -8.46
6/17 8.87 7.65 4.90 2.54 -8.00
7/17 9.18 7.76 5.20 4.01 -6.92
8/17 8.22 7.48 6.98 2.74 -5.12
9/17 9.49 7.37 4.73 3.45 -5.45
http://www.brooksbaseball.net/velo.php?player=543022&b_hand=-1&time=month&minmax=ci&var=pfx_z&s_type=2&startDate=03/01/2017&endDate=12/07/2017&gFilt=&pFilt=CU|CS
I know the above shows both home and away- but to look at only his 2nd half away curve ball movement is a very small sample, around 40 curveballs.
Chatwood has had about a -7.0 drop on his curve for pretty much his whole career, save for the bulk of his first year in Anaheim. There have been some microtrend variances, but his career average has been more predictive of his coming season average than has any microtrend. He had a spike of less movement, then more in ’12: but his average around 7 held true the next year. He lost some movement before his injury in ’13, but again going forward his career average was most predictive. Each year he’s been around 7 with some variance.
http://www.brooksbaseball.net/velo.php?player=543022&b_hand=-1&time=month&minmax=ci&var=pfx_z&s_type=2&gFilt=&pFilt=CU|CS&startDate=&endDate=
It’ll likely be the same (-7.0 or so) next year despite some weird variance of losing a lot of movement at home for his last 40 or so home curveballs this season, and gaining a lot of movement on the road for his last 40 or so road curveballs. If projecting a future -9.0 or so vertical drop on his curveball is what makes him interesting- I disagree that there is reason to consider him interesting.
He’s so uninteresting that you’re still here! All curves move more away from Coors, and so does his. All changes and sinkers more away from Coors and so do his. But please tell me more about his strand rate like that’s some sort of revelation. Thanks for reading.
Phoenix256,
That’s what I like about FanGraphs. The writers don’t just regurgitate common knowledge and call it analysis, but instead give you fresh data and perspectives that make you think and possibly see things a different way.
I get it. It was insightful and thought provoking. Eno did an excellent article. I just wanted some sort of counterpoint in there. However, given the title of the article, that was not really his intention.