An open, community-maintained database of trading rules across 50+ forex proprietary trading firms. Daily drawdown, max drawdown, profit target, news rule, copy-trade rule, platforms, and more — all in one CSV.
Live spreadsheet: Open in Google Sheets Article + context: hftforexcopier.com/prop-firm-rules-database/
Every row is one prop firm. Every column is one rule. Every cell links back to the firm's official source page. Free, open, machine-readable.
The motivation: every trader who funds a prop account spends hours re-reading rule books. Prop firms also change rules quietly. This repo aims to be the single source of truth — community-maintained, source-linked, version-controlled.
22 columns per firm:
| Column | Description |
|---|---|
firm_name |
Trading name and common alternatives |
founded |
Year the firm was established |
jurisdiction |
Country of registration |
eval_type |
1-step, 2-step, instant funding, scaling plan |
account_sizes_usd |
Available account sizes (min and max) |
profit_target_pct |
Percentage gain required per phase |
daily_dd |
Daily drawdown rule (%, $, static/trailing, reset time) |
max_dd |
Maximum drawdown rule |
min_trading_days |
Minimum active days per phase |
max_period_days |
Maximum days to complete each phase |
news_rule |
News-trading restriction (window in minutes) |
weekend_hold |
Weekend position holding allowed? |
ea_allowed |
Automated trading (EAs) permitted? |
copy_trading_allowed |
Copy trading / signal following permitted? |
hedging_allowed |
Hedging permitted? |
consistency_rule |
Single-day-profit cap (% of total) |
profit_split_pct |
Trader's share of profit |
platforms |
MT4, MT5, cTrader, DXtrade, MatchTrader, etc. |
payout_policy |
Frequency, minimum, method |
notes |
Edge cases, recent changes, asset-specific rules |
source_url |
Direct link to the firm's official rule book |
last_verified |
Date the row was last cross-checked |
data/prop-firms.csv— canonical CSV (UTF-8, comma-separated, RFC 4180)data/prop-firms.json— JSON mirror (regenerated from CSV on every merge)data/prop-firms.md— markdown table (human-readable, regenerated)docs/changelog.md— significant rule changes by firm
Easiest path: open the Google Sheet. Use the filter view to narrow by news rule, copy-trading rule, profit split, platform, etc.
curl -O https://raw.githubusercontent.com/HFTArbitrage/prop-firm-rules-database/main/data/prop-firms.csvOr in Python:
import pandas as pd
url = "https://raw.githubusercontent.com/HFTArbitrage/prop-firm-rules-database/main/data/prop-firms.csv"
df = pd.read_csv(url)
# Filter: firms that allow copy trading and have profit split >= 80%
df[(df["copy_trading_allowed"] == "yes") & (df["profit_split_pct"] >= 80)]The CSV is stable in column order and naming. Wire it into Looker Studio, Grafana, Metabase, or anything that consumes CSV.
Two paths — pick whichever fits your workflow.
- Fork the repo
- Edit
data/prop-firms.csvdirectly, or open the row file infirms/<firm-slug>.yamlif present - Update
last_verifiedto today's date - Open a PR with a clear description and the source URL for every changed cell
- Open the Sheet
- Use Suggesting mode to propose changes
- We review weekly and merge approved suggestions back into the canonical CSV
A new firm needs:
- All 22 columns filled (use
unclearif the rule is not documented) - A
source_urlfor at least: profit target, drawdown rules, news rule, platforms supported, profit split - The firm must be operating publicly (not private trading desks)
- The firm must offer forex (firms that are pure futures or stocks are out of scope for v1)
- Update only the cells that changed
- Update
last_verifiedto today - Reference the source URL in the PR / suggestion
Every cell follows three rules:
- Source-first. Each row links to the firm's own published rule book or terms page. We do not paraphrase from third-party reviews or affiliate sites.
- Verbatim where possible. Numbers, percentages, and time windows are copied as the firm publishes them.
- Honest gaps. Rules that are not documented unambiguously get the value
unclearwith a link to the relevant page.
The data is reviewed weekly. Source URLs are revisited monthly to catch silent rule changes. Significant changes are noted in docs/changelog.md.
This database is a research tool and a starting point. Prop firms change rule books without notice. Before opening a position whose outcome depends on a specific rule, always verify with the firm directly.
The maintainers and contributors are not liable for trading decisions made on the basis of this data. The last_verified date and source_url columns exist precisely so users can confirm freshness before relying on a cell.
The dataset is licensed under Creative Commons Attribution 4.0 International (CC BY 4.0). You may copy, redistribute, and adapt the data — including for commercial purposes — provided you give appropriate attribution and link back to this repository.
The scripts and tooling in this repo (in scripts/, tools/) are licensed under MIT.
The project was initiated and is maintained by the HFT Forex Copier team. Day-to-day updates are community-driven through pull requests.
For questions: open a GitHub issue, or email support@hftforexcopier.com.
Current scope (v1):
- 50+ forex-focused prop firms
- 22 rule columns
- CSV + JSON + markdown formats
- CC BY 4.0 license
Planned:
- Historical data — track how rule books change over time
- Per-firm pages with full rule-book quotes
- Compatibility matrix: prop firm ↔ trade copier ↔ platform
- API endpoint for programmatic access
- Pure-futures and pure-stock prop firms (separate dataset)
Not in scope:
- Rankings, "best of" lists, or recommendations
- Affiliate links or sponsored entries
- Subjective reviews of firms