Skip to content
NTDevLab

NTDevLab / CSP

CSP.GG — CS2 matches and predictions

Outcome probabilities, team form, maps and lineups for Counter-Strike 2. Open statistics of the CSP model results and a live mode during matches.

  • Python
  • FastAPI
  • ML
  • PostgreSQL
  • Active project
  • Interface languages: RU / EN

The problem it solves

Following CS2 teams and maps manually takes hours. CSP collects match data and turns it into outcome probabilities and readable team form statistics.

Data and ML

The model uses match results, map statistics and team form data. The ML pipeline and result validation are maintained by NTDevLab.

The NTDevLab approach

The prediction is published before the event and is never rewritten afterwards: the open history keeps both hits and misses, so model accuracy can be checked independently.

What you can see there

Match calendar with probabilities

Team form and map statistics

Live mode during matches