AQX SportIQ - AI Performance Analytics

AI platform that analyzes player stats to predict performance, injury risk & winning strategy. For coaches, not just analysts.

Comment

Inspiration

What it does

How we built it

Challenges we ran into

Accomplishments that we're proud of

What we learned

What's next for AQX SportIQ - AI Performance Analytics

Inspiration

In Indonesia, 90% of grassroots football clubs (SSB) still coach by feeling. We saw talented young athletes in Semarang getting injured because training load isn't monitored. We watched a U-16 tournament where a winger cramped at minute 60 for the third game in a row.

In the 【entity-Premier League¦canonical_name=Premier League】, every sub is data-driven. In Liga 1, 【entity-Persib¦canonical_name=Persib】 Bandung with Bobotoh has massive fanbase but still lacks affordable analytics for academy. And for 【entity-Timnas Indonesia¦canonical_name=Timnas Indonesia】, we have to wait for overseas analysis.

Why can't a local SSB coach in Semarang or Bandung have the same power as a 【entity-Premier League¦canonical_name=Premier League】 analyst? That's why we built AQX SportIQ - Indonesia & 【entity-Persib¦canonical_name=Persib】 Bandung Edition.

What it does

AQX SportIQ is an AI Performance Analytics platform for grassroots coaches, built for Indonesia & 【entity-Persib¦canonical_name=Persib】 Bandung.

For 【entity-Timnas Indonesia¦canonical_name=Timnas Indonesia】:

  • Passing Network 3-4-2-1: Visualize 【entity-Thom Haye¦canonical_name=Thom Haye】 key passes vs Arab Saudi (412 passes, 83% accuracy)
  • Heatmap 【entity-Witan Sulaeman¦canonical_name=Witan Sulaeman】: Distance 11.5km, Top speed 33.1 km/h
  • Performance Score Rafi Pratama: AI Score 0-100

For 【entity-Persib¦canonical_name=Persib】 Bandung:

  • Passing Network 4-3-3: 【entity-Marc Klok¦canonical_name=Marc Klok】 x Dedi Kusnandar 35x combo, 456 passes 87% accuracy
  • Heatmap 【entity-Beckham Putra¦canonical_name=Beckham Putra】 #7 & Ciro Alves #77: Action in final third 48%
  • Smart Substitution: When 【entity-Marc Klok¦canonical_name=Marc Klok】 stamina 61%, AI recommends 【entity-Beckham Putra¦canonical_name=Beckham Putra】 98% in

Core Features for All:

  • Player Load Monitoring: Input training minutes & RPE, AI calculates acute:chronic workload ratio to predict injury risk.
  • AI Video Analysis: Upload match clip, we detect sprints, distance covered, and fatigue patterns.
  • Smart Recommendation: Not just "player is tired", but "Rest winger for 48h, risk of hamstring strain is 78%".
  • Stasiun Bola Command Center: GBLA Bandung & GBK Jakarta full analytics dashboard with crowd analytics, turf monitoring, and tactical grid.

How we built it

  • Frontend: Next.js on Vercel for fast, mobile-first dashboard - because coaches check from the field. Designed for low-bandwidth SSB.
  • Backend & AI: Python + Pandas for data processing, OpenAI API to translate raw metrics into natural language coaching advice in Bahasa Indonesia & Sunda.
  • Data Integration: AQX Sports Analytics Data + manual data collection from Persib vs Bali & Indonesia vs Vietnam cases.
  • Visualization: Custom heatmap, passing network, and performance score built with D3.js style overlay.
  • Deployment: Vercel + Serverless Python API. Live at aqx-sportiq.vercel.app

Challenges we ran into

  1. No open dataset for Indonesian SSB & Liga 1: We had to create synthetic data based on interviews with 3 local SSB coaches in Semarang and match footage of Persib Bandung.
  2. Making AI sound like Bojan Hodak, not a robot: We had to prompt-tune OpenAI to speak like a real Indonesian coach: "Jaga tengah, dobrak sayap".
  3. GBLA & GBK full stadium rendering: Adding 77.193 penonton Bobotoh & Garuda with realistic crowd analytics was heavy.
  4. Balancing Timnas & Persib: Making one platform relevant for both National Team 3-4-2-1 and Persib 4-3-3.

Accomplishments that we're proud of

  • Built a working prototype that can flag high injury risk with 85% accuracy on our test data.
  • Created 5 visual assets: Ciro Alves & Beckham Putra heatmaps, Persib passing network, Marc Klok dashboard, Coach AI 70', and GBLA + GBK Stasiun Bola full house (95% occupancy).
  • Created a product that actually solves a local problem we saw ourselves in Semarang & Bandung, not just another Premier League clone.
  • Successfully deployed live at aqx-sportiq.vercel.app - Created by Indonesia & Persib Bandung.

What we learned

We learned that the best tech for grassroots isn't the most complex one. It's the one a coach in GBLA or a small SSB in Demak will actually open after a muddy training session.

We learned that Indonesia & Persib Bandung fans don't just want stats, they want stories: Why Beckham Putra should replace Marc Klok at 70' when stamina drops.

And we learned how to bridge sports science + AI + Bobotoh passion + real user empathy.

What's next for AQX SportIQ - Indonesia & Persib Bandung Edition

  • Integration with GPS trackers: Polar / Garmin for real-time load monitoring at GBLA.
  • Computer Vision: Automatic player tracking from a single phone camera on the sideline of SSB fields.
  • Pilot Program: 5 SSB in Semarang, Demak & 2 Akademi Persib Bandung in Bandung.
  • Persib Bandung Official Partnership: Offer Stasiun Bola command center to Persib academy.
  • Bahasa Sunda & Jawa Version: For even wider adoption in West & Central Java.
  • GBK Jakarta Integration: Full energy usage, turf health 96%, and crowd analytics 95% for Timnas.

Built in Indonesia, Created by Indonesia & Persib Bandung, For Indonesian Football.

Built With

Share this project:

Updates

posted an update —

u pdate Final - AQX SportIQ: 【entity-Timnas Indonesia¦canonical_name=Timnas Indonesia】 Passing Network

Terima kasih AQX! Final submission sudah live.

What I Built: AQX SportIQ - AI Performance platform yang menganalisis statistik pemain untuk memprediksi performa & risiko cedera, dengan studi kasus 【entity-Timnas Indonesia¦canonical_name=Timnas Indonesia】 Passing Network.

Key Insights dari project ini:

  • Visualisasi passing network Timnas - siapa playmaker sebenarnya?
  • Analisis player stats untuk prediksi performance drop
  • Injury risk prediction based on workload
  • Interactive dashboard untuk pelatih

Why Timnas? Karena data Timnas kita under-analyzed. Padahal dengan analytics yang tepat, kita bisa optimize strategi seperti tim pro Eropa.

Stack: Sports analytics, data visualization, AI prediction

Code 100% open-source, siap di-review juri!

Wish me luck vs 208 participants!

【entity-TimnasIndonesia¦canonical_name=Timnas Indonesia】 #SportsAnalytics #AQXDataBowl

Log in or sign up for Devpost to join the conversation.

posted an update —

Update #2 - Persib Bandung Edition is Live!

We just launched Persib Bandung edition of AQX SportIQ!

✅ Added Persib 4-3-3 Passing Network vs Bali United (456 passes, 87% accuracy) ✅ Added Beckham Putra #7 & Ciro Alves heatmaps with 11.5km / 33.1km/h stats ✅ Added GBLA Bandung Lautan Api full house - Bobotoh blue sea ✅ Added Stasiun Bola command center with crowd analytics 95% ✅ Changed Created By to Indonesia & Persib Bandung

From Semarang for Bobotoh. Built with next.js, openai, pandas, python, vercel.

Live: aqx-sportiq.vercel.app

PersibBandung #Bobotoh #AQXHackathon #Indonesia

Log in or sign up for Devpost to join the conversation.

posted an update —

In Indonesia, 90% of grassroots football clubs (SSB) still coach by feeling. We saw talented young athletes getting injured because training load isn't monitored, and coaches making sub decisions too late. In the Premier League, every decision is data-driven. Why not for local coaches? That's why we built AQX SportIQ - to bring pro-level analytics to every SSB in Indonesia.

Log in or sign up for Devpost to join the conversation.

Submission history