Answering The Cole

The A's might have gotten a lot more than anyone thought from him.

In the world where 100 mph pitches and 12 Ks in a game are both commonplace, Cole Irvin’s numbers looked, well, unenthusiastic. Coming into the 2021 season, his MiLB/MLB numbers were these:


Cole Irvin, pre-2021 numbers.


In his time in the majors (with the Phillies) and the IP he threw in the minors, Irvin had kept a very similar trend in K% and BB% while doubling his homers per nine innings (HR/9) and ERA. Very unenthusiastic.

So, I was very intrigued on what was that the A’s front office saw when trading for Irvin during the off-season, as, a) I could not see the upside on this trade, and b) I’ve grown skeptical of said front office through the years (yes, I’m an Oakland’s fan). Usually, trades for cash considerations are not the type of trades that you expect would yield the kind of initial success Irvin has experienced in Oakland so far:


Cole Irvin, 2021 stats.

He’s provided the A’s with 35 mostly great quality IP: a career-high 23.6% K% and a career-low BB% of 3.5%, both preponderant factors in his success.

His below-average CSW% is confusing because he is still getting a good amount of Ks, even though his four-seam fastball is sitting at barely 91.0 mph, almost two ticks lower than last year.

So it’s obvious he is doing something else to achieve these results, but, what is it? Well, it looks like it’s more a combination of small things than just one huge adjustment.


Command & Control


Irvin’s 3.5% BB% ranks 6th among all the qualified starters in the league, behind some guys named deGrom, Cole, Buehler, Ryu (and Eflin). This low walk rate has helped him maintain a K-BB% in the top 25, with 20.1%, it is not elite, but it is definitely very good.

This characteristic is also reflected in two important stats: O-Swing% and Zone%. O-Swing% shows the ability to deceive batters and making them chase and swing to pitches out of the zone, while Zone% is how frequently the pitcher locates his pitches inside the strike zone.

I like to combine O-Swing% and Zone% in just one summed stat, that way, Irvin is 15th in the league with 80.5%, better than Clayton Kershaw, Tyler Glasnow, and Joe Musgrove, to name a few. This bump in O-Swing% + Zone% helps him reach position 46 in the speX leaderboard, in the same tier as Brady Singer:


Cole Irvin – speX numbers.


Singer has a way better CSW%, but Irvin relies less on power and more on deception, getting the strikeouts that way.



There has to be a good level of confidence to throw these changeups outside the zone to get the strikeouts.

Now, we have established that Irvin’s superb control and location are helping him obtain better results, but how is he achieving this?


Repeat, Repeat, Repeat


Digging into all the possible variables for Irvin, I found that he is doing something a lot better this year, and it is the repeatability in his delivery.

He is very consistent in the way he releases the ball from his left hand in the direction of the batter, but this is an improvement this year from previous seasons. To evaluate this, let’s compare side to side his 2019-2020 and 2021 release points graphics:



We are looking at them from the batter’s perspective; those are the points where Irvin released each one of his pitches in the noted timeframe (2019-2020 on the left, 2021 on the right), and the distances in the axis are in feet.

Highlighted and crossed by a vertical and a horizontal line, there are also the centroids in each graph; that’s just a fancy name for the average release point for each one, the theoretical optimal point if all the balls were released at the exact same point each time.

The first thing to notice is that, in 2021, the horizontal displacements are smaller, which means that Irvin is not extending or “opening” his arm action in the direction of first base (being a lefty), so he is releasing the ball a little more centered from the batter’s point of view. There are a couple of ways this can be done, one of them is by the initial placement he is having on the rubber on the mound.

Although the camera’s placements are less than optimal, we can see this by comparing one pitch from him in 2019 on the left and another in 2021 on the right, in the following pics:



This is actually a common and very simple tweak that pitchers do all the time and can sometimes help with ball placement, as it is a way of correcting for the displacements and breaks their pitches have. This has very probably influenced the ability Irvin has to command his throws and, although it still a little early to talk about it, we can be very optimistic that his BB% is close to stabilizing, as it usually does around 40-45 IP, depending on if he can repeat these mechanics consistently.

Talking about this “repeatability”, there is a relatively simple (although imperfect) way of comparing how consistent Irvin is now in terms of releasing the ball from the same spot, which is also desirable for tunneling purposes, and that’s to check how much each of his pitches deviates from the average release point (the so-called centroid in the previous graph) as that indicates if he is able to consistently do it at the same spot, or as close as possible to it.

Skipping on the math, I calculated that deviation through the Root Mean Square Error (RMSE), where the smaller it is, the better as it means that the throws are not deviating that much from the same release point. The results for the pre-2021 period and this season are as follows:


Cole Irvin – RMSE at release points.


The RMSEy column shows the deviation on the height of the release point and for 2021 it is smaller than for the previous seasons, meaning that Irvin can maintain a more consistent height when releasing the ball this season. This, as I mentioned earlier, is extremely desirable for any pitcher because it adds an extra layer of difficulty for the batters to guess and decide when and how to swing as every pitch comes from the same spot.

With a few tweaks (and the hard work they imply), Cole Irvin has transformed himself into an incipient version of Eddie Plank, striking out just enough hitters while keeping good command and control of his pitches.

Without going deeper in other stats, but going back to the mechanical and placement basics, we can find signs about this “new version” of Cole Irvin that point to the possibility of a continued success; it is still too early to forecast how will it end, but his journey has a good probability to continue to be a good one, based on the foundations he is showing so far.

These foundations make me a believer that Irvin will surpass the projections from the major systems and will finish the season with an ERA of around 3.90, a 20% K-BB%, a 1.25 WHIP, while also being able to work a good quantity of innings, a thing that is precious in this high uncertainty season.


Photo by Leslie Plaza Johnson/Icon Sportswire | Adapted by Jacob Roy (@jmrgraphics3 on IG)

Carlos Marcano

Just a Venezuelan, not living in Venezuela. Intrigued by most of the things that can be measured in baseball, football, basketball, soccer, and life. I love to try to estimate performances.

2 responses to “Answering The Cole”

  1. Travis says:

    With the simple windup, and pitch type, he seems like the next dallas keuchel

Leave a Reply

Your email address will not be published. Required fields are marked *

Account / Login