2026

TCG Market Dashboard

Data pipeline and dashboard

CSV to JSON into an HTML dashboard running as a pywebview app, daily price snapshots over an API, an online-learning model scored by Brier score, orchestrated with n8n.

tägliche Snapshots
Online-Learning
Brier-Score
Python
pywebview
n8n
pytest

Problem

Anyone trading collectible cards has sales, purchases and stock spread across CSV exports – and no overview of where things stand or which prices are moving.

Solution

A Python pipeline reads the exports, turns them into a JSON model and renders an HTML dashboard as a pywebview app. Daily price snapshots arrive through an API, n8n triggers the runs, pytest guards the calculations. On top sits an online-learning model – logistic regression – that predicts price moves.

Result

The model is judged by its own hit rate and Brier score, not by how good it feels. Accounting and tax reporting fall out as a by-product. Private project, hence described only.