iiinova — where your digital self keeps evolving

Ryan Gunther @ryangunther1.bsky.social · 14/10/2026

Very cool to get this recognition from Kyle Boddy on X last month! I would not be 1% of the analyst I am today without everything I’ve learned from driveline blogs, podcasts, YouTube videos, tweets, open data and a visit to the facility. This means the world to me!

041 4

Ryan Gunther @ryangunther1.bsky.social · 10/06/2026

Otto Lopez train is still so alive baby we’re already at a career high fWAR and yes we are still ignoring the .389 BABIP

000 0

Ryan Gunther @ryangunther1.bsky.social · 03/06/2026

you can break down all 15k swings from the past week into 8 buckets FB / SL / CU / CH, and Same hand matchups (LvL,RvR) / opposite matchups Slider_same generates statistically significantly lower bat speeds compared with every other group No other group gives s.s. bat speeds with anything else

010 1

Ryan Gunther @ryangunther1.bsky.social · 03/06/2026

I don't know why I never noticed this before, but I found that bat speed differences are significant when comparing and inside pitch to an outside pitch, or a center-cut pitch to an outside pitch, but not from a inside pitch to a center-cut pitch. Holds for both LHBs and RHBs

110 1

Ryan Gunther @ryangunther1.bsky.social · 21/04/2026

Otto Lopez train keeps rolling. We're ignoring the BABIP right now and just focusing on him already establishing a new career high MaxEV this year

001 1

Ryan Gunther @ryangunther1.bsky.social · 20/04/2026

My swing video dataset consists of a few hundred swings from the 2025 season. Randomly, the fastest bat speed of any of these... by a whole 3 mph... this swing by 23 year old, 5'10" 150 pound org depth outfielder Jonatan Clase at 85.7mph

100 1

Ryan Gunther @ryangunther1.bsky.social · 17/03/2026

Back after a while of no posting to say that I've been hired as a computer vision engineer for nurivatech.ai - a baseball biomechanics project aiming to grow accessibility to motion capture assessments for athletes everywhere About 7 weeks on the job now, life is great! Tons of smart people here

112 2

Ryan Gunther @ryangunther1.bsky.social · 03/01/2026

Just finished Nate Silver’s book. I was really really impressed. Dives into everything risk from pro poker to +EV sports betting, prisoner’s dilemmas, Wall Street/finance, Silicon Valley/Tech VCs, to AI acceleration and nuclear weapons game theory. Couldn’t stop reading this

030 3

Ryan Gunther @ryangunther1.bsky.social · 05/12/2025

98th percentile @fangraphs.com user, up from 93rd last year

000 0

Ryan Gunther @ryangunther1.bsky.social · 04/12/2025

Now we see why I maxed out at the collegiate level 🤣 just 11 big league homers were hit at or below this bat speed in 2025 (Me, today, measured by blast motion)

000 0

Ryan Gunther @ryangunther1.bsky.social · 02/12/2025

Hire me! - Master's degree in statistics (thesis on hitting biomechanics) - Experienced in computer vision, R, SQL, Python - interned with 2 Indy ball teams, in pitch data (2024) & automation (2025) - former collegiate player - worked 3.5 years in finance/contract structuring - Bachelors in finance

010 1

Ryan Gunther @ryangunther1.bsky.social · 01/12/2025

In today's episode of unintuitive things I'm excited about - Isomap finds Ernie Clement and Ernie Clement's mirrored swing almost identical after dimension reduction! This means that if lefties continue to have higher transformation scores than righties, I've found something significant!

000 0

Ryan Gunther @ryangunther1.bsky.social · 15/11/2025

Here are two swings, same hitter, overlaid on each other, one makes contact with a ball in Zone 10, the other, Zone 11 (up+out vs up+in) Barrels (red/purple) tunnel each other like 2 pitches tunnelling each other. But from overhead view you can see that they break off pretty early in the motion

412 4

Ryan Gunther @ryangunther1.bsky.social · 14/11/2025

Exciting new updates in baseball computer vision - I generated this (and many other clips like it) completely automatically through a pipeline I've built Here's Adrian Del Castillo turning 96 above the zone into a gap double. Auto color-coded too, so the denser centers hold more weight in my model

100 1

Ryan Gunther @ryangunther1.bsky.social · 27/10/2025

Putting the pieces together in my computer vision project - I've now taken Bogaerts swing footage, programmed the ability to automatically drill out "donut holes" i.e. the section between XB's arms at contact, then colour code so the centers of the data hold more weight

000 0

Ryan Gunther @ryangunther1.bsky.social · 24/10/2025

No matter how I tune this, tree-based models only care to split for runners advancing from 2B on possible sac flies. From 3B it's pretty simple Hangtime doesn't matter. Neither does OF initial positioning, launch angle, OF arms, runner speed, or where the ball's hit. If it's >260ft, you're scoring

000 0

Ryan Gunther @ryangunther1.bsky.social · 22/10/2025

Finally figured this out on Monday. The below video isn’t perfect (helmet..?), but as long as it works conceptually, I can add more training data + annotations and build it into a smooth pipeline within a week or two

000 0

Ryan Gunther @ryangunther1.bsky.social · 22/10/2025

I’ve struggled recently with trying to get my computer vision biomechanics model to properly segment the entire hitter’s body, but NOT the “donut holes” i.e. background that lands between the arms and bat, circled below

000 0

Ryan Gunther @ryangunther1.bsky.social · 21/10/2025

Insane that the Ms didn’t have Munoz ready in that spot Couldn’t possibly have been higher leverage

000 0

Ryan Gunther @ryangunther1.bsky.social · 21/10/2025

By cWPA, the Springer home run is the 68th most important play in MLB history, and 3rd highest outside the World Series. Never wake me up from this dream

100 1

Reposted by Ryan Gunther

Mike Petriello @mikepetriello.bsky.social · 08/10/2025

The average pitch the Cubs batters have seen this postseason is 92.7 MPH. Not the average "fastball." The average PITCH. All of them. That's faster than the average -fastball- the 2008 Cubs saw in the playoffs: 92.4.

311316 113

Ryan Gunther @ryangunther1.bsky.social · 05/10/2025

Not sure I’ve ever seen a pitcher shake his head as he comes set. Either you shake off a pitch or you’re pissed about a bad call. But a borderline call went his way the pitch before this Last pitch he threw today Is this a “I hate the pitch call but I’ll go along with it anyway?”

000 0

Ryan Gunther @ryangunther1.bsky.social · 29/09/2025

made some 3d visualizations of a motion capture session. batter approaching, making, and then following through his contact point

000 0

Ryan Gunther @ryangunther1.bsky.social · 28/09/2025

ALEJANDRO KIRK MY GLORIOUS KING!!!!!!

000 0

Ryan Gunther @ryangunther1.bsky.social · 28/09/2025

ALEJANDRO KIRK MY GLORIOUS KING!!!!!!

000 0

Ryan Gunther @ryangunther1.bsky.social · 26/09/2025

A common issue with using computer vision for biomechanics is that baggy jerseys/pants, shin/elbow guards and long hair become part of the data. To combat this, I've used color gradients on my data - the edges now hold less importance than the denser centers

120 2

Ryan Gunther @ryangunther1.bsky.social · 12/09/2025

I also got in touch with BaseballCV (open source baseball computer vision project) since they don't currently have a segmentation model in their pipeline. Ideally I'd donate my datasets and model weights there once I've finished this stage of my research

020 2

Ryan Gunther @ryangunther1.bsky.social · 12/09/2025

Here's my FIRST test of my self-trained computer vision segmentation model on completely unseen footage. Overall, SUPER happy with the results! A liiiittle bit of flickering and we lost a lower leg in the follow through. Nothing a bit more training data can't fix. A breakthrough for sure

221 2

Ryan Gunther @ryangunther1.bsky.social · 09/09/2025

I'm about 300 frames of manual annotation in to my segmentation project. Should be enough to begin training and automate this (incredibly lengthy) process

000 0

Ryan Gunther @ryangunther1.bsky.social · 30/08/2025

Andy Pages - Home Run - 101.6 mph EV - 68.6mph Bat Speed Now fully annotated via cvat.ai

100 1

Ryan Gunther @ryangunther1.bsky.social · 26/08/2025

I really like this paragraph I wrote tonight, summarizing a chapter of my work. There are bad / inefficient ways to move, but once you get past that, there are many different ways to get to successful swing outcomes in baseball

000 0

Ryan Gunther @ryangunther1.bsky.social · 23/08/2025

Made with inspiration from ViTPose in Google Colab. Will put the code on my Github this week

000 0

Ryan Gunther @ryangunther1.bsky.social · 20/08/2025

I wonder if Andy Pages knows how much effort I've put into tracing his body to build a training set

010 1

Ryan Gunther @ryangunther1.bsky.social · 13/08/2025

If you're looking to get started in baseball biomechanics using the OBP dataset, I've also uploaded a file that downloads the c3d files from OBP github and converts them to csv's, more easily accessed from a lot of machines and more easily manipulatable. Check it out!: github.com/RyanGunther

github.com

RyanGunther - Overview

MSc student - Brock University / For written updates of my thesis: https://bsky.app/profile/ryangunther1.bsky.social - RyanGunther

000 0

Ryan Gunther @ryangunther1.bsky.social · 13/08/2025

Pushed a bunch of files to my github today The main one was an R file that ran through the dimensionality reduction pipeline from previous posts, found GLOBAL principal markers i.e. same for every swing, segmenting motions based on cov structure and then completed a cross validated LASSO regression

100 1

Ryan Gunther @ryangunther1.bsky.social · 08/07/2025

I can’t sleep so I’m working on computer vision instead

000 0

Ryan Gunther @ryangunther1.bsky.social · 06/07/2025

green dots are original marker placement, purple dots are the model's "best guess" at where those body parts should be based on the position of the red dots at that frame. This is a spectacular result

100 1

Ryan Gunther @ryangunther1.bsky.social · 06/07/2025

(1/n) After dividing the swing into approximately linear data-driven segments (see previous posts) I've now bundled similar segments back together using k-means clustering and taught an algorithm to "learn" the data

110 1

Ryan Gunther @ryangunther1.bsky.social · 25/06/2025

I don't always watch games with Savant pitch feed on the other tab but for Skenes vs. Mis it's pretty necessary

000 0

Ryan Gunther @ryangunther1.bsky.social · 16/06/2025

Still, it’s tough to see this as anything other than a salary dump by Boston, but if that was the goal, they definitely got a nice return while doing so

000 0

Ryan Gunther @ryangunther1.bsky.social · 16/06/2025

Not really sure what’ll happen with 3B next season for the Red Sox. I’m less convinced than most that Bregman will opt out given he seemed to be playing over his head earlier this season. (that might be because of his pull tendencies and the monster)

000 0

Ryan Gunther @ryangunther1.bsky.social · 16/06/2025

Red Sox despite being 37-36 seem to be throwing in the towel on going all the way this season? Playoff odds currently at 33.7%. Nice return for them, I’m actually surprised they got as much as they did after moving the contract. Harrison+Hicks is still only 28+Tibbs is a 45 FV

000 0

Ryan Gunther @ryangunther1.bsky.social · 16/06/2025

Giants are competing and early in deals with JH Lee, Adames, Chapman, now Devers. Rotation was middling before this and they lost some depth but got better overall. It’s a nice all-in move by the giants

000 0

Ryan Gunther @ryangunther1.bsky.social · 16/06/2025

SF takes on roughly 8.5/$270M of Devers. Taking that contract alone is most of the return for BOS. Harrison is a nice piece, velo spike this year. Giants had issues signing FAs so I think of this trade as an FA signing that also costed players, given Raffy could probably get more than 270M in FA rn

000 0

Ryan Gunther @ryangunther1.bsky.social · 15/06/2025

(1/2) How I broke swing biomechanics into entirely data-driven segments, dynamic to each hitter and stance: - Determine principal features of the swing - the markers that explain the most variance (a lot of biomech data is redundant) - group the non PFs as isotropic (same in all directions) noise

110 1

Ryan Gunther @ryangunther1.bsky.social · 11/06/2025

A few weeks ago I toured @drivelinebaseball.com’s Arizona facility and was introduced to some of their staff - really nice people doing really cool work. I got to watch a hitter mocap session, a Trajekt machine in action, and a swing design session. Really cool shit going on there

000 0

Ryan Gunther @ryangunther1.bsky.social · 05/06/2025

Some research I worked on was presented at the Statistical Society of Canada annual meeting last week. It was an honour to be part of such a thorough and diligent team of statisticians from Brock University! 🦡

010 1

Ryan Gunther @ryangunther1.bsky.social · 23/05/2025

Recently at my video scouting baseball ops internship for the Hamilton Cardinals, I totally automated our data entry pitch tracking process using Apps Scripting and sheets formulas that build the entire game's data off of just a few inputs. This should save 1 hour+ per IBL game per week!

000 0