Phone-camera ball counts and action clips for junior pace bowlers at academies
TL;DR Spellcount is a phone mounted behind the nets that counts every ball each junior fast bowler sends down and saves a short clip of their action. Coaches see weekly loads and sudden spikes. Academies pay per bowler and pass the cost on to parents, who'd like fewer teenage back injuries, thank you very much.
Young fast bowlers break down a lot, and the main reason is load. In a 2026 study of 98 bowlers aged 10 to 20, 16.3% had a lumbar bone stress injury in one season, and 28.6% had had one at some point. Spikes in workload showed up in the six weeks before diagnosis[1]. A 2026 meta-analysis found that a sharp rise in recent load against the usual level more than doubled injury risk (pooled relative risk 2.45)[2].
Rehab for cricket back pain can run past eight months[3]. And yet most Indian academy coaches count balls from memory, if at all, while a dozen bowlers rotate through three nets.
What it does, in regulatory terms: training-load logging and coaching video only. Spellcount does not diagnose or predict injury, and it says so to parents. Medical questions go to a doctor or physiotherapist.
A coach fixes a phone on a tripod behind the bowler's end. The app recognises each bowler from a short enrolment clip and their kit, counts deliveries and keeps a two-second clip of every sixth ball from the back-foot landing to the follow-through. You get a weekly table per bowler with balls bowled, sessions, the change from their four-week average and a red flag when the jump is large.
The clips give you a quick side-by-side view of front-foot landing and shoulder alignment over the season. Joint angles from a single phone aren't lab-grade though, so Spellcount shows clips for the coach to judge and makes no angle claims until it has its own validation data.
Bowlers use it and their parents fund it through academy fees. The buyer is the academy owner or head coach, who pays a monthly fee per enrolled fast bowler and can bill parents for a premium tier. I couldn't find a published survey of Indian academy fees, so price testing is part of validation. State associations and school sports programmes that run pace camps are a second buyer.
Phone cameras and on-device pose models are now good enough to track a bowler at 60 frames per second without sensors. Research on affordable motion capture for beginner and intermediate cricketers is active but still unsolved[4], which leaves room for a narrow, counting-first product. The biggest amateur scoring app added AI match highlights in 2025, so amateur cricketers clearly accept automated video[5].
Nobody keeps a public count of Indian cricket academies or junior fast bowlers. The nearest adjacent figure is one amateur scoring app, which claims more than 5 million cricketers on its scoring app[5].
The injury evidence shows the need[1], but you'll have to size your market bottom-up, one city's academies at a time.
Scoring apps cover scoring, streaming and highlights but describe no academy load tools[5]. One Bengaluru startup sells a bat sensor and simulators aimed at batting[6]. Abroad, elite teams log bowling loads with wearables and staff, which junior academies can't afford.
The gap is a cheap, sensor-free tool built for the bowler's side of the net.
| Difficulty | Moderate (39/100) |
| Starting capital | ₹35–65 L |
| Time to launch | 6–9 months |
The tough technical task is counting reliably when several bowlers share a net and bowl in quick turns. Selling is the other headache, because academy owners run on thin margins and coaches may see a counter as a judgement on their work. Handling minors' video also adds consent work up front.
No sector licence applies, but nearly every user is a minor, so data rules come first.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI make medical calls. Spellcount logs load and shows clips, and a red flag is a prompt to talk to a physio, not an injury warning. Joint angles from one phone aren't lab-grade either.
Want to work out the AI layer for Spellcount? Talk to Yash →
Here's what can go south -
I'd back this for a founder who has coached or played age-group cricket and can build computer vision. Without access to academies it's a weak bet, honestly.
Prove one thing first - that the phone counts deliveries within 2% of a human tally across five busy net sessions. Then test whether owners will pay.
My estimate is ₹35–65 L to launch, mostly computer-vision work and academy pilots.
No sector licence, but the DPDP Act requires verifiable parental consent for under-18 users, and you must avoid any injury-diagnosis claims.
Yes, in any country with junior cricket academies, like England, Australia or South Africa. Children's data rules differ though, so check them first.
The free Spellcount validation kit has the line-by-line capital breakdown, the effort scoring, the child-data consent checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for coaches and parents, and 5 assumptions to test. Download the kit →
Thinking about building Spellcount or something close to it? The make-or-break work is accurate counting in crowded nets and an academy price that sticks, and I'm happy to think both through with you. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.