Support Board
Date/Time: Sun, 29 Mar 2026 14:38:43 +0000
Post From: Looking for cheap historical MNQ tick data with aggressor side classification
| [2026-03-29 03:39:16] |
| User920211 - Posts: 1 |
|
Building a backtesting platform for MNQ futures in Python. I need historical tick data going back to at least 2021, ideally 2019. The critical requirement is aggressor side classification per trade — meaning each trade record must identify whether it was executed on the bid or the ask. I need this to compute Cumulative Volume Delta (CVD) accurately. This is NOT the same as just having bid/ask quote prices attached to a trade. I already have Databento for 3 months of MNQ trades schema data. The schema is exactly what I need — aggressor side per tick, millisecond timestamps, clean Parquet format. But extending back to 2021+ costs around $2000 which is out of range for a personal project. Things I have already tried and ruled out: - Databento — perfect data, too expensive for full history - Sierra Chart Package 10 with SC Data — could not get more than a few months of intraday tick history despite 2200 day setting configured - Sierra Chart Denali feed — requires live funded trading account for real-time, overkill for historical only - FirstRateData — OHLCV bar data only, no tick with aggressor side - TickTradingData — unclear if they carry true aggressor side classification or just bid/ask quote prices - Portara/CQG — tick data available but no confirmed aggressor side classification, CSV only, expensive What I need: - MNQ futures tick data from 2019 or 2021 minimum to present - True aggressor side per trade (bid hit vs ask lift) — not simulated, not inferred - CSV or Parquet format so it slots into an existing Python/Polars pipeline - Affordable for a personal/independent trader project What are people actually using for this? |