PrizePicks & Sleeper Props API: Goblins, Demons, Line Moves

PrizePicks & Sleeper Props API: the data feed behind your pick'em tool

Building a Discord bot, a prop optimizer, a picks newsletter or your own model? This Actor gives you the data layer: every PrizePicks player prop (standard, goblin and demon ladders, promos) and every Sleeper Picks line (over/under payout multipliers, pick counts, last-10 results and hit rate) as clean JSON, in one schema, with IDs that survive line moves.

You don't have to deal with DataDome, US-only geo-blocking, residential proxies, JSON:API joins or Sleeper's ID lookups. Call it from your code, schedule it every morning or every hour, and read the dataset. No PrizePicks or Sleeper account needed.

Built for people who ship prop tools

• Discord and Telegram bots: post the day's goblins, the biggest line moves or free squares as soon as they appear.
• Pick'em optimizers and +EV finders: full goblin/demon ladders grouped by marketKey, next to Sleeper multipliers for the same player and stat.
• Models and dashboards: a timestamped snapshot every run (scrapedAt) and a stable projectionId, so your database gets a line history neither app publishes.
• Picks newsletters and sheets: send each run to Google Sheets, Airtable or a webhook with no code.

Why this feed

Use it as an API

Copy to your AI assistant

Paste this into ChatGPT, Claude, Cursor or any coding assistant to start building on the feed:

epicscrapers/prizepicks-props-scraper on Apify: PrizePicks and Sleeper Picks player props in one schema.
Call: ApifyClient("TOKEN").actor("epicscrapers/prizepicks-props-scraper").call(run_input={...}), then read
client.dataset(run["defaultDatasetId"]).iterate_items(). Inputs: sources (["prizepicks","sleeper"]),
leagues (string[], PrizePicks names: NFL, NBA, MLB, NHL, CFB, WNBA, TENNIS, SOCCER, UCL, CS2...; empty = all),
includePeriodBoards (bool), maxResultsPerLeague (int, 0 = full board, default 500), playerNames (string[],
substring, accent-insensitive), statTypes (string[], exact statType or Sleeper statKey), lineTypes
(["standard","goblin","demon"]). One row per line: source, league, playerName, team, opponent, statType,
lineScore, lineType, overMultiplier, underMultiplier, pickCountOver, pickCountUnder, recentResults,
recentHitRate, isPromo, flashSaleLineScore, gameStartTime, status, projectionId (stable across line moves),
marketKey (groups a goblin/demon ladder), scrapedAt (same for the whole run). Fields one app lacks are null.
Pricing: $0.03 per run + $0.05 per 1,000 lines. Token: https://console.apify.com/account/integrations

Call it from code

Run the Actor and get the rows in one HTTP call (good for boards up to a few thousand lines):

curl -X POST "https://api.apify.com/v2/acts/epicscrapers~prizepicks-props-scraper/run-sync-get-dataset-items?token=<YOUR_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"sources": ["prizepicks", "sleeper"], "leagues": ["NBA"], "maxResultsPerLeague": 0}'

Python:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_TOKEN>")
run = client.actor("epicscrapers/prizepicks-props-scraper").call(run_input={
    "leagues": ["NFL", "NHL"],
    "lineTypes": ["goblin"],
    "maxResultsPerLeague": 0,
})
for line in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(line["playerName"], line["statType"], line["lineScore"], line["lineType"])

JavaScript:

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<YOUR_TOKEN>' });
const run = await client
    .actor('epicscrapers/prizepicks-props-scraper')
    .call({ leagues: ['NBA', 'NFL'], maxResultsPerLeague: 0 });
const { items } = await client.dataset(run.defaultDatasetId).listItems();

Recipes

Track line moves

Schedule the Actor (every morning, or hourly on game days) and store each run. Every row has scrapedAt (the same for the whole run) and a projectionId that stays the same when the line moves. Join two runs on projectionId and compare lineScore. A projectionId that disappears was pulled from the board.

Rebuild goblin and demon ladders

Goblins and demons come in ladders: the same player, stat and game at several lines. Group by marketKey and sort by lineScore to get each ladder with its standard line.

Compare PrizePicks with Sleeper

Both sources share one schema and the same league names. Match on league, playerName and statType to see where the two apps disagree. The apps name some stats differently (PrizePicks Pass Yards, Sleeper Passing Yards), so map those in your code.

Flag hit-rate outliers

Every Sleeper row has recentResults (last ≤10 games for that stat), recentAverage and recentHitRate (share of those games strictly over today's line). Filter to recentHitRate ≥ 0.7 or ≤ 0.3 and show them with overMultiplier, underMultiplier and the pick counts. The Hit rates (Sleeper) view in the Output tab does this in the Console.

Post free squares and flash sales

PrizePicks promo rows have isPromo: true, flashSaleLineScore (the discounted line), discountName (for example free_square) and discountPercentage. Filter on isPromo and post them to your channel.

Schedule it and connect the output

1. Save your input as a task (Save as new task at the top of the input page).
2. Open the task, then Schedule: for example 9:00 every day, or every hour on game days.
3. Connect the output: a webhook to your bot or backend on ACTOR.RUN.SUCCEEDED, the Google Sheets integration, Zapier, Make, or read the dataset with the API.

Input

Example inputs:

{ "sources": ["prizepicks", "sleeper"], "leagues": ["NBA", "NFL"], "maxResultsPerLeague": 0 }

{
    "sources": ["prizepicks"],
    "leagues": ["NFL"],
    "playerNames": ["josh allen", "puka nacua"],
    "lineTypes": ["goblin", "demon"]
}

{ "leagues": ["NFL", "NHL"], "includePeriodBoards": true, "maxResultsPerLeague": 0 }

Output

One row per line. Real example from a PrizePicks NBA board (a promo line):

{
    "source": "prizepicks",
    "league": "NBA",
    "playerName": "Anthony Edwards",
    "team": "MIN",
    "teamName": "Timberwolves",
    "opponent": "MIA",
    "position": "G",
    "statType": "Points",
    "lineScore": 27.5,
    "lineType": "standard",
    "flashSaleLineScore": 0.5,
    "isPromo": true,
    "discountName": "free_square",
    "discountPercentage": 98,
    "status": "pre_game",
    "gameStartTime": "2026-10-21T23:40:00.000Z",
    "updatedAt": "2026-10-08T18:05:51.382Z",
    "homeTeam": "MIA",
    "awayTeam": "MIN",
    "period": "Full",
    "trendingCount": 70627,
    "projectionId": "15575749",
    "marketKey": "prizepicks:188017:19:11:NBA_game_ZHHvKkKdvBJF6u8gKONUoWLw",
    "playerImageUrl": "https://static.prizepicks.com/images/players/NBA/Anthony_Edwards_NBA_player_rhlEP3GCAg5LDJD8uChjFU81.webp",
    "scrapedAt": "2026-10-12T22:24:46.985Z"
}

And a Sleeper line with multipliers and hit rate:

{
    "source": "sleeper",
    "league": "NHL",
    "playerName": "Shea Theodore",
    "team": "VGK",
    "opponent": "MIN",
    "statType": "Assists",
    "lineScore": 0.5,
    "lineType": "standard",
    "overMultiplier": 2.7,
    "underMultiplier": 1.29,
    "recentAverage": 0.33,
    "recentHitRate": 0.17,
    "gameStartTime": "2026-10-13T00:00:00.000Z",
    "projectionId": "20261012_MIN_VGK:assists_over_under_player_1031"
}

Pricing

$0.03 per run plus $0.05 per 1,000 lines saved, on every plan. Platform usage and the US residential proxy PrizePicks needs are included. The run fee is charged per GB of memory; the default 1 GB is plenty.

What real runs cost (Tuesday morning, October 13, 2026, full boards):

Monthly cost of common schedules:

About two thirds of PrizePicks lines are demons. If your tool doesn't need them,

lineTypes: ["standard", "goblin"]

FAQ

Is there an official PrizePicks API? No public one. This Actor reads the same JSON the PrizePicks board loads and gives it to you as an API.

How fresh is the data? Each run fetches the live boards. A run of every league takes about a minute; a single league takes a few seconds. Schedule it as often as your tool needs.

Can I get historical lines? Neither app publishes past boards. Schedule the Actor and keep the datasets (or push them to your database); scrapedAt and projectionId turn them into a line history.

Do I need a PrizePicks or Sleeper account, or a VPN? No. PrizePicks only answers US connections, so the Actor uses US residential IPs for it. You can run it from anywhere.

Why are some fields empty? Each app returns different data. Multipliers, pick counts and hit rates are Sleeper only; promos, ladders and trending counts are PrizePicks only. Combos have no single opponent.

Why did a league return nothing? The board is empty when a league has no games soon (for example NBA in the off-season). The log lists leagues with no board.

Underdog, Chalkboard or sportsbooks? Not yet. Tell us in the Issues tab which one your tool needs.

Is it legal? The Actor collects public line data shown to any visitor of the apps, without logging in. Check the apps' terms and your local laws on betting before you use or republish the data.