What Marco Gonzales has done is incredible

Coming into the year, I had written Marco Gonzales off. I hadn’t written him off completely, I just didn’t think he would be good. Last year Marco Gonzales was not good, and he was also barely serviceable. Through 40.0 innings, Gonzales accumulated 0.1 WAR, a 6.08 ERA, 5.06 FIP, and 4.46 xFIP. He was, almost by definition, a replacement level player.

This year, Gonzales has been one of the better starting pitchers in the league, as he has been good for 2.5 fWAR thus far. Given what Gonzales has said about his health, we shouldn’t be surprised about the results of either this or last year.

Last year, I thought I was about 85% back, but looking at it from where I am now, it was closer to 60-70%. Now I think I’m 90-95%.

A lot of us knew that Marco Gonzales was coming off of a (somewhat) recent Tommy John surgery, but it feels different hearing him say that he wasn’t very close to 100%.

You could just read the article, but there are quite a few interesting tidbits about his health, as well as how Gonzales’ injury affected him.

In April, I, with a hint of skepticism, raved about what Marco Gonzales was doing. At the time, he was coming off of a couple encouraging starts, and I was interested in how his pitch mix, release point, and effectiveness had all changed. Then, I was going off of just two promising successive starts, but now we have 113.1 innings to go off of. The title then was Is Marco Gonzales for real?, and it now feels safe to say that Marco is at least not not for real.

Some things have not been surprising. That Gonzales’ changeup has been good — not surprising. Since college, it has been his best pitch, and according to Eno Sarris, it’s “a special pitch.”

What have been surprising are his cutter, sinker, and curveball.

His cutter has been incredible. So incredible that it has been, by most metrics I have looked at, his best pitch in terms of outcome and batted ball quality. What’s interesting is that this is a brand new pitch for him. He introduced it this year, and throughout the year has equally mixed it in with his sinker, changeup, and curveball. However, if you look at how his month-to-month pitch mix is trending, you’ll notice something interesting.

Gonzales pitch mix update

(Try to halfheartedly ignore the month of July, because Gonzales has only made two July starts.) Marco has phased out his four-seam because he knows it sucks, and month by month, he has grown increasingly confident in his cutter as a quality pitch. The 36 times Gonzales threw his cutter against the Angels on July 5th was the most times he’s thrown any pitch in a game this year.

Baseball Info Solutions and PITCHf/x pitch type data vary in terms of pitch classification, but regardless of which you look at, Gonzales has had one of the most valuable cutters in the league (per 100 pitches).

According to PITCHf/x:

gonzales pfx


And Baseball Info Solutions:

gonzales pi

Twelfth! Ignore Tyler Mahle!

Regardless of how you look at it, Gonzales did not have a cutter, and now he has a cutter. When you add a new pitch and it becomes one of your best pitches, surely you’re probably going to improve.

As for his sinker, which he calls a two-seam fastball (same thing!), it has also been one of the most valuable of its kind. Don’t let me tell you, I’ll show you!

sinker gonzales

there are only five sinkers with positive value

While his sinker is interesting, I think it’s enough to say that 1) It’s been good. 2) He has benefited from throwing it less with other pitches mixed in.

What I wanted to focus on is that his curveball, which previously was not good, has been one of the best in baseball.

Lately, I have grown fond of using QOP Baseball. Their website looks like it was designed by a sixth grader in his (or her!) intro to web design class, but it has been cosigned by baseball minds I trust and appears to be very legitimate. What has drawn me to QOP Baseball is the uniqueness of its QOP statistic. Their website describes QOP as follows:

QOP gives a statistic describing the quality of a pitch on a scale of mostly 0 to 10. The MLB average is around 4.5 with median 5. QOP is a patent-pending proprietary regression model that incorporates MPH, location, and movement (vertical break, horizontal break, breaking distance, and rise).

Apparently, Gonzales’ curveball grades out as the best in baseball via QOPA.

QOPA Gonzales

If the threshold of pitches is lowered to 300, then Gonzales falls to second just behind Seth Lugo, but regardless, Marco Gonzales once again has had a poor pitch become a very good pitch. In terms of his curveball’s components:

  • Quality of pitch: 97th percentile
  • Rise: 13th percentile
  • Late break: 90th percentile
  • Horizontal break: 42nd percentile
  • Vertical break: 13th percentile
  • Location: 91st percentile
  • Velocity: 50th percentile

From these numbers, it’s clear what makes Marco’s curveball successful is both his location and the late break of the pitch. This overlay exemplifies why it’s so hard to hit his curve:

This gif couldn’t be a more perfect example of how the late break of his curve benefits him. Gonzales’ curveball is identical to his cutter, until it reaches the hitter’s swing decision point, at which point it drops off the table into the middle of the strike zone. Hitters are left swinging at a cutter that isn’t there and left to brood over the curveball that they are currently swinging over. Oh, and according to FanGraphs?

curveball Gonzales

Not one, but two Mariners!

Turns out, FanGraphs concurs: it’s one of the most valuable curveballs in baseball. It used to feel more prestigious being at King Felix’s level in, well, anything. Nowadays, it’s more troubling than it used to be to rank behind Felix, and here Marco’s curve does. But Felix’s curveball has never been the problem, and it remains a very good pitch.

It’s true that Gonzales has been fortunate. His .299 wOBA is 27 points better than his .326 expected wOBA (xwOBA), but there are 40 pitchers with higher wOBA-xwOBA differentials. Good fortune notwithstanding, he ranks 62nd in xwOBA above Zack Greinke, David Price, and Masahiro Tanaka.

For all the grief Jerry Dipoto received for trading dinger-hitting Tyler O’Neill for soft-tossing Marco Gonzales, the Mariners have been more profitable thus far. Through 19 MLB games, O’Neill has a 2.1% BB%, 42.6% K%, and 0.2 WAR. That’s really bad! Meanwhile, Gonzales has helped to buoy the Mariners’ rotation with a 3.41 ERA, 3.48 xFIP, and (if you’re into it) team-leading 10 wins. Clearly, it is far too early to determine the “winner” of the trade, but it certainly feels better to be on our end than the Cardinals’ at the moment.

Whether through his interviews or watching him pitch, you can tell Marco Gonzales is a bright, deliberate individual. We have watched him evolve just in the matter of months. Finally, after several years, he is healthy, and for the first time in his major league career, he is good. Marco has always had the plus changeup, and he has always had the plus command. For the first time, his other pitches have seemingly caught up to his change, and at the very least they aren’t lagging behind. You would never mistake Marco Gonzales for Chris Sale, but for the Mariners, you have to be giddy, because you would also never mistake him for Yovani Gallardo.

Yet another Edwin Díaz post

At the present moment, Edwin Díaz ranks first in the MLB in reliever WAR. On May 1st, I wrote an article discussing how Edwin Díaz had changed. At that time, he was behind Josh Hader and Adam Ottavino in WAR. Now, despite only pitching 6.2 more innings, Díaz leads Ottavino by 1.0 WAR, and leads Hader by a not-insignificant 0.4 WAR.

Before I was skeptical because it’s rather easy to fall prey to small sample sizes. Díaz had surely been dominant, but it had just been a month or so, and his 2017 was certainly more mediocre than elite.

Yet here we are, July 10th, and Díaz is among the top of most leaderboards in which you would like to see him leading. In xwOBA (expected weighted on-base average), Díaz is second only to Josh Hader. In actual wOBA, Diaz ranks sixth. In xFIP, Díaz is first. In SIERA, Díaz is second (by only one point!). In K/9, Díaz is fourth.

As I was taking a peek at QOP Baseball, I was surprised to find that Díaz’s fastball and slider both ranked incredibly low in terms of quality. How could someone who has been so elite not have quality pitches?

According to FanGraphs, Díaz has one of the best sliders in the league.

Slider table (7-10-18)

wSL/C: Weighted slider runs per 100 pitches

FanGraphs’ leaderboard falls in line with what most people believe of Díaz’s slider: it’s good! More specifically, it’s the 11th best in the MLB.

According to QOP Baseball, this is how Díaz’s slider ranks in the following categories:

  • Rise: 57th percentile
  • Late break: 47th percentile
  • Horizontal break: 5th percentile
  • Vertical break: 29th percentile
  • Location: 21st percentile
  • Velocity: 97th percentile
  • 3.36 QOPA (quality of pitch average; 4.50 is league average)

For QOP Baseball, their logic is that a good pitch will rank highly in many, if not most, of these categories. Díaz only fares well in velocity. If you hadn’t seen him pitch and only looked at these metrics, you would think he’s a mediocre fireballer.

QOP Baseball does have a point. Díaz’s slider is pretty subpar in terms of movement, and his velocity is very high for a slider.

Here is a table of the top 11 sliders in baseball (per FanGraphs), mostly in order:

Name HMov VMov Velo
Snell 2.75 0.79 88.54
Alexander 1.23 1.11 85.95
Pruitt 1.49 2.00 88.67
Bradley 2.55 3.54 88.18
McHugh 9.51 1.97 80.36
Robertson 4.66 1.09 87.13
Mikolas 2.25 1.7 88.57
Hader 3.44 1.8 82.26
Glasnow 3.08 6.73 85.99
Bauer 6.88 0.99 83.32
Díaz 1.12 2.24 89.76
 averages: 3.541818 2.178182 86.24818

Díaz’s horizontal movement is well below average (even with McHugh single-handedly swaying the mean), and his vertical movement is relatively average. However, he does have the fastest slider of the group by more than 1.0 mph.

This seems like an oversimplified method of measuring the quality of a pitch. Or maybe it’s not! Like QOP Baseball says: it’s measuring the quality of a pitch, not the value of it.

Other than velocity, location, and movement, there are other methods to set yourself apart as a pitcher. Here is where Díaz does just that:

Pitch tunneling is an increasingly popular concept that suggests that the longer two different pitches travel within the same path, the more effective they will be. That seems incredibly intuitive — and it’s likely a reason why Felix Hernandez was successful for so many seasons: his changeup looked like his fastball, until it didn’t.

So if you see a hitter swing at a Díaz slider like this, you’ll know that it must have looked like his fastball:

A more straightaway view may give a better glimpse into how utterly ridiculous it must be to try and hit against Edwin Díaz:


To further visualize this, we will compare Edwin Díaz and Juan Nicasio’s fastball-slider combos to one another.

First, a bird’s-eye view visual:

Diaz-Nicasio top

You can see that Nicasio’s pitch trajectories diverge earlier, and are wider later.

A side visual:

Diaz-Nicasio side

This is just another vantage point, but again, you can see that immediately Nicasio’s fastball-slider trajectories are not intersecting. For Díaz, they are intersecting for the first five or six points.

It still isn’t entirely clear why Díaz has taken such a gigantic leap forward since 2017. As I’ve suggested before, his slider has gained a 1.0 mph uptick in velocity. He’s stopped throwing his two-seam fastball as much in favor of his four-seam. He’s increased slider usage. His release point has changed.

Regardless of what is precipitating his success (which is almost definitely the return of his slider), he’s a major reason the Mariners have had so much success this season. When you have a closer as dominant as Díaz, you’re going to win close games, and boy have the Mariners won more close games than they should have. Edwin Díaz is on pace for 3.2 WAR, and the Mariners are 6.0 games ahead of the Athletics in the Wild Card race. For now, all is well. There’s not a better way to slice it.