Analytics and monitoring decision
Scouting Stats AI
Keep paying — the product rests on proprietary event data and calibrated modelling validated at scale (transfers and tens of thousands of matches), which are costly and time-consuming to recreate; a useful DIY subset is possible but won't match core value.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All Scouting Stats AI alternatives, with the arithmetic →
What a replacement has to do
- Tell it your club; it detects squad needs, ranks realistic targets by fit and market constraints, and lets you search/compare players and build scout reports.
What it still won’t have
- Proprietary event data coverage across 150+ leagues and the 40,000+ player database
- The vendor's proprietary, calibrated modelling pipeline and validation against 11,409 real transfers and 50,000+ matches
- Scout-video rendering pipeline (TikTok-ready, sub-minute render) and built-in export templates
- Pre-built recruitment rules (market corridors, budget filtering, GBE work-permit estimates) and product polish
What remains hard
- Proprietary models
Proprietary models, tested where it counts.
- Proprietary data
Pro-grade event data from 150+ leagues — every shot, pass, pressure, carry.
- Proprietary data
11,409 Real transfers validating our fit model
- Proprietary data
400+ features, and run through calibrated ML models tested on 50,000+ matches.
First-year cost
Keep paying
Paying is—cheaper in year one.
On cash alone, building overtakes the subscription at 11 seats.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal ScoutingStats replacement using: Postgres for relational storage, ClickHouse for analytics (optional), Python (FastAPI) backend, Airflow or cron for ETL, XGBoost/LightGBM for models, and React for a web UI. Scope: (1) ingest a provided sample match event feed and normalize to Postgres; (2) implement scheduled feature-engineering jobs to compute rolling form and league-normalised metrics (aim for a subset of ~50 core features); (3) train and serve a calibrated gradient-boosted model for match probability and a simple player attribute scorer; (4) implement backend endpoints for player search, similarity-by-features, predictions, and watchlists; (5) build a basic React UI with player profile page and a drag-and-drop report canvas that binds to live data and exports PNG/PDF. Out of scope: full 150+ league coverage, TikTok video rendering, automated work-permit/legal rules. Require: error handling for failed ETL/model jobs, unit tests for ETL and model training pipelines, CI configuration, and simple deployment scripts (Docker + Kubernetes or Docker Compose).
How we checked
How the score was reached
- Pay verdict base20
- An open-source build was found+5
- 4 cited sources+3
- Price verified on pricing page+3
- 3/3 assessment runs agreed+4
- Hard moats found in the evidence-6
- Evidence score29
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.
How scoring works →Cited sources · 4
Every page the run actually retrieved.
- official productScoutingStats — product
- official pricingScoutingStats — pricing
- open sourceapache/superset
- open sourcemetabase/metabase
Integrity checks
What held up, and what did not.
