Every Precision Sport Has Shot Data. Archery Still Has a Scorecard.

Since 2003, the PGA Tour has tracked every shot hit in competition — trajectory, distance, lie, result. Mark Broadie spent years building a framework around that data, and when strokes-gained went mainstream in 2011, it didn't just change how analysts talked about golf. It changed how players practiced. By 2015, Major League Baseball's Statcast was running in every stadium, capturing exit velocity, launch angle, and spin rate on every pitch and every batted ball. Coaches stopped guessing. The data told them where the problem actually was.

Now think about archery. You shoot an end. You walk to the target. You pull your arrows and write down a number. Maybe you take a photo. If you're disciplined, you log it somewhere. Then you walk back and do it again.

You know your score. You don't know your pattern.

What the Data Revolution Actually Built

It's worth being precise about what golf and baseball analytics actually accomplished — because the story isn't about technology for its own sake. Both sports built data infrastructure because raw results weren't answering the right questions.

Before strokes-gained, a golfer who shot 72 knew he'd had a decent round. He didn't know whether he'd been rescued by his putter or quietly destroyed by his approach shots. Broadie's methodology — built on the ShotLink dataset — gave coaches a way to isolate exactly where shots were being gained or lost relative to field average. Golf Channel named it one of the 25 most impactful moments in the sport's modern history. Not because the math was elegant, but because it finally let coaches ask precise questions.

Statcast in baseball worked on the same principle, applied to faster events. A hitter who strikes out might have made contact with a well-struck ball at 105 mph that happened to go directly at an outfielder. Or he might have topped a weak grounder. The result — an out — looks identical on a scorecard. The data reveals which one actually happened, and whether the underlying quality of contact is improving.

Both systems were built around a core insight: in precision sports, the quality of individual executions matters as much as aggregate outcomes. A good round can hide a broken swing. A lost tournament can mask improving technique. You can't train toward something you can't measure.

Archery Is a Precision Sport Without a Precision Data Layer

Archery demands the same standard of technical consistency as golf, competitive shooting, or any discipline where small variations compound into significant performance gaps. The body mechanics involved — draw length, anchor point, grip pressure, release timing — have to repeat within narrow tolerances on every shot, across every end, across every session.

Archery Analytics GmbH, the German firm behind the RyngDyng automated scoring system used at elite tournament level including the 2024 Paris Olympics, put it plainly in their research: the difference between good and very good archers is that the latter have less variation in parameters from shot to shot, end to end, day to day. Not better averages — less variance. That's a data problem. You can't identify your variance without measuring it.

Academic research has reinforced this. A 2023 study from the University of Notre Dame developed ArcheryVis, a system demonstrating that automatic shot detection is achievable and that visual analytics reveal patterns that manual scorekeeping routinely misses — grouping drift, directional tendencies, end-by-end degradation. Research by Ertan and colleagues on recurve hit distribution showed that even high-level archers who cluster around center still exhibit spatial patterns that would benefit from analysis. The tools to understand those patterns exist. The infrastructure to put them in a competitive archer's hands has not.

The apps that fill the current market — MyTargets, iArchery, Archery Companion — are digital scoresheets. They replace paper with a phone screen. They track totals and averages. They do not analyze where your arrows are landing, how your groupings shift across an end, or whether your 9 o'clock drift is getting worse as you fatigue. That layer of analysis has simply not existed at the consumer level.

Why the Gap Existed — and Why It's Closed

The gap wasn't a matter of indifference. It was an engineering constraint that has only recently been resolved.

Applying computer vision to archery target detection runs into a specific problem: arrows at distance present a very small target for detection. An arrow nock on a 122cm face at 70 meters occupies a few pixels in a standard camera frame. Until recently, building a model accurate enough to reliably detect and locate impacts — and distinguish arrow positions from target face markings — required hardware and processing resources that didn't fit in a consumer product.

Three constraints have changed. Mobile hardware has advanced to the point where the processing required for real-time image analysis runs locally on a phone without the latency that would make the system unusable in a training context. Model accuracy for small-object detection has improved substantially, particularly for specialized tasks where training data can be focused on a controlled environment. And cost — both of compute and of the development pipeline — has dropped enough to make consumer-grade deployment viable.

Elite facilities and tournament operators have had access to hardware-mounted systems for several years. RyngDyng ran at the Paris Olympics. But those systems are facility-side infrastructure — fixed cameras, controlled lighting, dedicated hardware. The barrier for an individual archer or a small club has remained essentially total.

The Moment Archery Catches Up

The best archery coaches have always worked from pattern recognition. End-by-end grouping. Left drift under fatigue. The way a shooter's 10-ring percentage shifts as wind picks up. This knowledge has existed in coaches' eyes for decades. The limitation has been that it couldn't be confirmed, quantified, or tracked over time without a significant manual logging effort that most archers simply don't sustain.

When a sport crosses the threshold from intuition-based coaching to data-confirmed training, the conversation changes. Not because the fundamentals change — the shot process is still the shot process — but because a coach can show a shooter exactly where their consistency is breaking down, end by end, session by session. And a shooter can see for themselves whether the adjustment they made in the third end held through the fifth.

Golf reached that moment in 2003. Baseball reached it in 2015. Archery is reaching it now.

Okami Vision is built specifically for this. It's a mobile application that uses computer vision to automate arrow detection and scoring, and returns spatial analytics — grouping patterns, directional drift, end-by-end consistency — that no scoresheet has ever provided. It was built for the archer who shoots three times a week and wants to train with the same precision they bring to their shot process.