Crowd tape
Reddit crowd sentimentTracked across the finance subredditsSee the dashboard for live mention data
The Data / Downloads
The Data · Downloads

Free data downloads

All seven files can be downloaded now, free and with no key.

The files

Australian short positions

Short positions on every ASX stock, as the regulator publishes them each business day, a few days behind. Shares sold short and shares on issue, per stock and report date.

From Australian Securities and Investments Commission.
Dated 29 Sep 2026

French disclosed short positions

Net short positions that holders have disclosed to the French regulator, added up per company. This is what was disclosed, not total short interest, and it is dated the day we read the register.

From Autorité des marchés financiers.
Dated 2 Oct 2026

House stock trades

Stock trades that members of the U.S. House have reported, one row per reported trade, with the date of the trade and the date it was filed. Filings that could not be read are left out, and no amounts are included.

From U.S. House of Representatives, Office of the Clerk.
Dated 1 Oct 2026

Insider buying clusters

Companies where several insiders bought shares on the open market within the last 30 days, counted from SEC Form 4 filings. A rolling window, not a history.

From SEC EDGAR.
Dated 2 Oct 2026

Fails to deliver

Shares the SEC lists as failed to deliver, per stock and settlement day. It is an outstanding balance, not a count of new failures, and it runs two to four weeks behind.

From U.S. Securities and Exchange Commission.
Dated 14 Sep 2026

Economic release calendar

The dates the big U.S. economic numbers and central bank decisions are due, from the agencies that publish them. A date marked unconfirmed can still move.

From the Bureau of Labor Statistics, the Bureau of Economic Analysis and the Federal Reserve Board.
Dated 5 Oct 2026

Fear and greed, market legs

The five market-based legs of our fear and greed gauge, one row a day. The headline score is left out because it also folds in Reddit, and nothing derived from Reddit is offered for download.

From The Distributed.
Dated 5 Oct 2026

Use a file

Every file is a plain link. No key, no sign-up. The JSON feed carries the same rows as the CSV, in the same order.

In Google Sheets

=IMPORTDATA("https://thedistributed.co/data/asic-shorts.csv")

In Python

import pandas as pd
df = pd.read_csv("https://thedistributed.co/data/asic-shorts.csv", comment="#")

The first lines of a CSV start with a # and hold the cite line and the date. Tools that read a CSV skip them with a comment setting; Sheets shows them as the first rows.

Cite a file

Every file opens with its own cite line and the date of its newest data. Copy the cite line as it is. Each row also names its source.

Limits

One address may make 30 requests a minute. Past that the answer is a plain 429 that says how many seconds to wait. A file is read 5,000 rows at a time: add ?page=2 for the next 5,000. The X-Pages header and the JSON feed say how many pages there are, and a page is never cut short.

What is not here

Anything we derive from Reddit is never offered as a download or a feed. Price series are never offered either. Politician files carry no amounts, and every member gets the same columns.