FanDuel props and history exports, CSV or NDJSON

Load a slate or a season of line history straight into pandas, DuckDB or a spreadsheet.

Two formats

CSV with a header row, or NDJSON one object per line, both streamed.

Resumable

Pages by cursor; a body may end with one error line naming next_cursor, so a long export resumes where it stopped.

Training set

One row per prop: open line and price, closing line and price, and the moves before the start, ready to join to graded results.

What it returns now

Real rows, 5 min delay, as of 3:25 PM ET

prop_id,league,player,stat,period,line,over_american,under_american,changed_at
734.196074478-12103341,NFL,"Austin Hooper",receiving_yards,game,6.5,-114,-114,2026-11-06T03:52:00.113Z
734.195663780-41066843,NFL,"Bijan Robinson",receiving_yards,q1,0.5,100,-132,2026-11-06T16:13:12.303Z
734.195448624-41072627,NFL,"Chase Brown",receiving_yards,q1,2.5,-118,-114,2026-11-06T16:51:42.168Z
734.195673178-73434816,NFL,"Dohnte Meyers",receiving_yards,game,17.5,-114,-114,2026-11-06T03:52:00.113Z
734.195450822-41172406,NFL,"Drake London",receiving_yards,game,78.5,-114,-114,2026-11-06T03:52:00.113Z
734.196074482-36184724,NFL,"Drew Sample",receiving_yards,game,3.5,-114,-114,2026-11-06T03:52:00.113Z
734.195448685-40890769,NFL,"Ja'Marr Chase",receiving_yards,game,78.5,-114,-114,2026-11-06T03:52:00.113Z
734.195663786-41155779,NFL,"Jahan Dotson",receiving_yards,q1,0.5,168,-230,2026-11-06T16:51:42.168Z

The full export carries every column: cursor, prop_id, sport, league, event_id, event_name, starts_at, player, team, stat, period, line, over_american, under_american, over_decimal, under_decimal, market_status, changed_at, first_seen_at. History exports add one row per recorded state.

Endpoints and parameters

Code

curl "https://fanduelapi.com/api/v1/export?kind=history&format=csv&from=2026-11-05T00:00:00Z&to=2026-11-06T00:00:00Z" \
  -H "Authorization: Bearer $FANDUEL_API_KEY"

import os, io, requests, pandas as pd

r = requests.get(
    "https://fanduelapi.com/api/v1/export",
    params={"kind": "history", "format": "csv", "from": "2026-11-05T00:00:00Z", "to": "2026-11-06T00:00:00Z"},
    headers={"Authorization": f"Bearer {os.environ['FANDUEL_API_KEY']}"},
)
r.raise_for_status()
df = pd.read_csv(io.StringIO(r.text))
print(df.groupby("stat").size())

Keys look like fd_.... Keep one in the FANDUEL_API_KEY environment variable on your server, never in browser code. Every response's meta says which plan answered, how far behind it is and how fresh the collection is.

Which plans include it

• FreeWith a full archive purchase
• DeveloperIncluded
• ProductionIncluded
• ScaleIncluded

Read from the plans on the pricing page. Every request counts against the monthly and per-minute allowance; request quotas explains how.

Works well with

• Line historyReplay a prop or a game line from the first time it was seen to now, move by move./api/v1/props/{id}/history
• Players and resultsJoin each FanDuel line to what actually happened, and train on graded outcomes instead of guesses./api/v1/players
• CoverageKnow what the feed holds before you build on it, and show users exactly how fresh it is./api/v1/coverage