# Action Network Scraper — Betting Splits, Odds & Line Movement (`logiover/action-network-scraper`) Actor

Action Network scraper for public betting splits (tickets % and money % per side), sportsbook odds with best price, opening-to-current line movement, player props and expert picks. NFL, NBA, MLB, NHL, college, soccer, UFC, tennis. No login. One row per market side, export JSON, CSV, Excel.

- **URL**: https://apify.com/logiover/action-network-scraper.md
- **Developed by:** [Logiover](https://apify.com/logiover) (community)
- **Categories:** Developer tools, AI, Automation
- **Features:** Limited permissions
- **Stats:** 8 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks
- **User rating**: No ratings yet

## Pricing

from $0.30 / 1,000 results

This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events.
Since this Actor supports Apify Store discounts, the price gets lower the higher subscription plan you have.

Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event

## What's an Apify Actor?

An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes.
In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours,
and optionally produces a well-defined JSON output, datasets with results, or files in key-value store.
In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server.

Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md.

## How to integrate an Actor?

If asked about integration, you help developers integrate Actors into their projects.
You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready.

Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run.

For examples already wired to this Actor's own input schema, see the [API](#api) section below.

Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`).

# README

## Action Network Scraper — Betting Splits, Odds & Line Movement

> **Not affiliated with, endorsed by or connected to The Action Network, Inc.** This is an independent tool that reads the public JSON the actionnetwork.com website serves to every visitor. It does not log in, use anyone's account or place bets. "Action Network" is a trademark of its owner, used here only to say which public data this Actor reads.

**Action Network Scraper** turns the public betting data on actionnetwork.com into clean, flat rows: **public betting percentages** (the share of tickets and the share of money on each side), **sharp money** signals, consensus and **sportsbook odds with the best price per side**, **opening-to-current line movement**, **closing lines with graded results**, **player props** and **expert picks with their win-loss record and ROI**. One row is one side of one market of one game, so a night of NBA basketball is a table you can sort, filter, chart or load into a model without unpacking nested objects. It covers NFL, college football, NBA, college basketball, WNBA, MLB, NHL, every major soccer league, UFC, boxing, ATP and WTA tennis and PGA golf. No login, no API key, no browser — the Actor reads the same JSON feed the website itself uses and exports **JSON, CSV, Excel, XML or HTML** from the Apify dataset.

If you were looking for an **Action Network API**, a **betting splits scraper**, **public betting percentages** in a spreadsheet, a **sharp money tracker**, a **reverse line movement** finder, an **odds comparison and line shopping** feed, a **closing line value (CLV)** dataset for backtesting or a **player props API** with best-price detection, this is it. Every price is delivered as **American odds, decimal odds, fractional odds and implied probability** (plus a no-vig probability), every game carries its start time in ISO-8601 UTC, and every row has a stable `rowId` so repeated snapshots can be de-duplicated, joined and tracked over time. This is data for research, analytics and modelling — not betting advice and not a promise of profit.

### Contents

- [At a glance](#at-a-glance)
- [What you get](#what-you-get)
- [Who uses it](#who-uses-it)
- [Quick start](#quick-start)
- [Recipes](#recipes)
- [Input reference](#input-reference)
- [Output: the data dictionary](#output-the-data-dictionary)
- [Coverage: leagues, books, periods and how far back](#coverage-leagues-books-periods-and-how-far-back)
- [Scale and speed](#scale-and-speed)
- [What you pay](#what-you-pay)
- [Integrations](#integrations)
- [Performance and cost tips](#performance-and-cost-tips)
- [Troubleshooting and status messages](#troubleshooting-and-status-messages)
- [Honest limits](#honest-limits)
- [How it compares with the alternatives](#how-it-compares-with-the-alternatives)
- [FAQ](#faq)
- [Glossary](#glossary)
- [Legality and responsible use](#legality-and-responsible-use)
- [Changelog](#changelog)
- [Support](#support)

### At a glance

| | |
|---|---|
| **What it reads** | The public JSON of actionnetwork.com: scoreboards, sportsbook markets, player-prop markets, expert picks and per-market price history |
| **Login / API key** | None; everything is served to anonymous visitors |
| **Leagues** | 27 leagues: NFL, NCAA football, UFL, NBA, NCAA basketball, WNBA, MLB, NHL, every major soccer league, UFC, boxing, ATP and WTA tennis, PGA golf (full list in the coverage section) |
| **Rows** | One per side of a market. `recordType` is `market`, `prop`, `pick` or `bookPrice`; all share one `rowId` scheme |
| **Splits** | `ticketPercent`, `moneyPercent`, signed `sharpGap` wherever Action Network publishes them |
| **Odds** | American, decimal, fractional, implied and no-vig probability; consensus price, opening price, best price among the sportsbooks you pick, and every book's own price |
| **Movement** | Opening line and price vs current, `lineMovement`, `oddsMovement`; optional timestamped consensus history |
| **Results** | `win` / `loss` / `push` for finished full-game rows, with the final score |
| **History depth** | Prices and splits are served from the 2023-24 seasons onward (measured per league below); earlier days return results without prices |
| **Speed (measured 2026-10-29)** | 100 rows in 3.1 s, 1,000 rows in 2.5 s, 12,468 unique rows in 62 s from one run |
| **Price** | Pay per result: $0.50 per 1,000 results on the Free plan, down to $0.30 on Gold and above. One game is about 11 rows, so roughly half a cent on Free. You never pay for rows you did not ask for |

### What you get

#### The four record types

Every row has `recordType`. The types share the event columns (`eventId`, `eventName`, `league`, `startTime`, teams, scores, status), so you can keep them in one table or split them with a filter.

| `recordType` | One row is | Switch |
|---|---|---|
| `market` | one side of a game market (a team on the moneyline or spread, Over or Under on a total or team total) with consensus price, tickets % and money %, opening price, movement, best price and each selected book's price | on by default (`includeMarkets`) |
| `prop` | one side of a player prop (Over or Under a player's points, passing yards, strikeouts ..., or a yes/no market such as first basket scorer) | `includeProps` |
| `pick` | one expert pick with the handicapper's public record over the recent window (wins, losses, pushes, units, ROI) | `includePicks` |
| `bookPrice` | one sportsbook's price on one market or prop side, as its own row: the long format for line-shopping tables | `includeBookPrices` |

#### Market rows

The product most people come for. One NBA game is six to ten rows: both sides of the moneyline, the spread and the total, plus (unless you filter them out) Over and Under for each team's own total. Each row answers the usual questions:

- *Price and line now?* `line`, `oddsAmerican`, `oddsDecimal`, `oddsFractional`, `impliedProbability`, `noVigProbability`.
- *Public vs money?* `ticketPercent`, `moneyPercent`, `sharpGap`.
- *Where did it start, how far did it move?* `openLine`, `openOddsAmerican`, `lineMovement`, `oddsMovement`.
- *Best price?* `bestOddsAmerican`, `bestBook`, `bookCount` and the `bookOdds` list with every selected book's price and line.
- *How did it end?* `isFinal`, `awayScore`, `homeScore`, `result` (`win` / `loss` / `push`).

#### Prop rows

Prop rows carry the same price columns plus the player (`playerName`, `playerTeam`, `playerPosition`, `jerseyNumber`, `opponent`) and the stat (`statType`, `statName`). The default prop menu per league: NBA, WNBA and college basketball points, rebounds, assists, threes, combinations, steals, blocks, turnovers and first-basket markets; NFL and college football yardage, touchdown scorers and kicking; MLB hits, home runs, strikeouts, total bases; NHL goal scorers, points, assists, shots, saves. `"propTypes": ["all"]` adds every market the league's props page lists, including the "N or more" ladders.

#### Expert pick rows

Public expert picks: who (`expertName`, `expertUsername`, followers), what (`pickText`, `pickType`, `line`, `oddsAmerican`, `units`), how it ended (`pickResult`, `unitsNet`, `settledAt`) and the expert's record over `expertRecordWindow` (`expertWins`, `expertLosses`, `expertPushes`, `expertUnitsNet`, `expertRoiPercent`). Useful to follow, to fade, or to measure who really beats the closing line.

#### Odds in every format

American odds are what the source serves; everything else is computed from that one number, so it is always consistent: decimal odds (4 decimals), the exact reduced fraction (`-110` is `10/11`, `+150` is `3/2`) and the implied probability including the sportsbook's margin. `noVigProbability` removes the margin across the sides of the same market (two-way markets, three-way soccer moneylines); where the sides are not exclusive, as with "anytime scorer", it stays empty instead of inventing a number.

#### Betting splits, sharp gap and what they mean

`ticketPercent` is the share of bets, `moneyPercent` the share of the money. A side can hold 80 % of the tickets and 40 % of the money when many small bets sit on one team and a few large ones on the other. `sharpGap = moneyPercent - ticketPercent` puts that difference in one sortable column: strongly positive is the classic **sharp money** reading, strongly negative means the public is on it. Combine it with `lineMovement` to find **reverse line movement**, the number moving against the side most tickets are on. Splits are Action Network's own tracked sample (see `totalBets`) and exist where it publishes them: moneylines, spreads and totals of most games, usually not team totals or props, where the columns stay `null` rather than showing a false 0 / 0.

#### Opening line, movement and closing line value

The `Open` book supplies the opening price and line. `lineMovement` is current line minus opening line of the selection (a favourite moving from -4.5 to -3.5 reads +1.0: the number got shorter). `oddsMovement` is the price change in American points on a continuous scale, so -105 to +100 is +5, not a jump of 205. For finished games the consensus price is the last number before the game, your closing line: compare it with the price you took for **closing line value (CLV)**, and use `result` for return.

#### Best price and line shopping

`bestOddsAmerican` is the highest price among your selected sportsbooks that quote the *consensus line*; books on another line appear in `bookOdds` with `atConsensusLine: false` but are never called best. `bookCount` says how many selected books quote the line. For a full line-shopping table switch on `includeBookPrices`: every book's price becomes a `bookPrice` row with `isBestPrice`.

### Who uses it

1. **Sharp-money watchers** — tonight's spreads with `onlyWithSplits`, sorted by `sharpGap` and `lineMovement`.
2. **Reverse-line-movement hunters** — tickets lean one way, `lineMovement` goes the other.
3. **Closing-line-value researchers** — opening and closing lines for a season: how often did the open beat the close?
4. **Model builders** — training sets of opening line, closing line, splits and final score with `result` already graded.
5. **Tipsters** — compare picks with the public percentages and check expert records before following anyone.
6. **Line shoppers** — a daily table of the best price per side and which book offers it.
7. **Arbitrage and +EV scanners** — `bookPrice` rows and `noVigProbability` as one source among several.
8. **Fantasy and DFS players** — player-prop lines with best price and movement as a cross-check.
9. **Newsletters and media** — a weekly "where the public is" table without copying numbers.
10. **Quants** — a consensus-price time series: schedule a run every few minutes and key on `rowId` + `scrapedAt`.
11. **Product teams** — an odds and splits feed for a dashboard, a Discord bot or an internal tool, without maintaining a scraper.

### Quick start

#### Run it in the Apify Console

1. Open the Actor and press **Try for free** (or **Start**).
2. Leave **Leagues** empty for every league, or type some: `nba`, `nfl`, `nhl`, `mlb`, `ncaaf`, `soccer`, `ufc`.
3. Choose the days: the default is today and the next two (US Eastern game days). For history set **From date** / **To date**; for NFL and college football use **Weeks** (and **Season** for past years).
4. Switch on **Player props**, **Expert picks**, **One row per sportsbook price** or **Line history** if you want them.
5. Set **Maximum rows** and, for a hard ceiling, **Maximum cost per run**.
6. Press **Start**. A one-day slate finishes in seconds. The **Output** tab has table views (splits, props, picks, sportsbook prices); export as JSON, CSV, Excel, XML or HTML.

#### Run it through the API

```bash
curl -s -X POST "https://api.apify.com/v2/acts/logiover~action-network-scraper/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"leagues":["nba","nhl"],"daysAhead":0,"marketTypes":["spread"],"onlyWithSplits":true,"maxItems":200}'
```

That call waits for the run and returns the rows. For longer runs start the Actor asynchronously, poll, then download `.../datasets/<id>/items?format=csv`.

### Recipes

Every block below is a complete input. Paste it into the Input tab (JSON view) or send it as the body of an API call. All dates are US Eastern game days, which is how Action Network groups a slate.

#### 1. Tonight's slate in every league

Everything with a game today, all leagues, all markets (the default window is today plus two days; `daysAhead: 0` keeps it to today).

```json
{
  "daysAhead": 0,
  "maxItems": 2000
}
```

#### 2. NFL week with betting splits

The current NFL week, moneyline, spread and total only, and only rows that carry splits.

```json
{
  "leagues": [
    "nfl"
  ],
  "marketTypes": [
    "moneyline",
    "spread",
    "total"
  ],
  "onlyWithSplits": true,
  "maxItems": 500
}
```

#### 3. Sharp-money watchlist for tonight's NBA

Upcoming NBA spreads with splits; sort the dataset by `sharpGap` or `lineMovement`.

```json
{
  "leagues": [
    "nba"
  ],
  "daysAhead": 0,
  "gameStatus": [
    "scheduled"
  ],
  "marketTypes": [
    "spread"
  ],
  "onlyWithSplits": true
}
```

#### 4. Closing lines and results for a month

Finished NBA and NHL games of a month: closing price, opening price, final score, graded `result`.

```json
{
  "leagues": [
    "nba",
    "nhl"
  ],
  "dateFrom": "2026-09-29",
  "dateTo": "2026-10-28",
  "gameStatus": [
    "final"
  ],
  "maxItems": 20000
}
```

#### 5. A whole NFL season for backtesting

Regular-season weeks 1-18 of 2025 (weeks, season and season type replace dates for the weekly leagues).

```json
{
  "leagues": [
    "nfl"
  ],
  "season": 2025,
  "seasonType": "reg",
  "weeks": "1-18",
  "maxItems": 30000
}
```

#### 6. Line-shopping table, one row per sportsbook price

Spreads and totals of the NFL week with every selected book's price as its own row.

```json
{
  "leagues": [
    "nfl"
  ],
  "marketTypes": [
    "spread",
    "total"
  ],
  "includeBookPrices": true,
  "books": [
    "draftkings",
    "fanduel",
    "betmgm",
    "caesars",
    "bet365"
  ]
}
```

#### 7. NBA player props with the best price

Points, rebounds, assists and threes for tonight, without the game markets.

```json
{
  "leagues": [
    "nba"
  ],
  "daysAhead": 0,
  "includeMarkets": false,
  "includeProps": true,
  "propTypes": [
    "points",
    "rebounds",
    "assists",
    "3fgm"
  ]
}
```

#### 8. NFL anytime touchdown scorer prices

One row per player with the best price and every book's price.

```json
{
  "leagues": [
    "nfl"
  ],
  "includeMarkets": false,
  "includeProps": true,
  "propTypes": [
    "anytime_touchdown_scorer"
  ]
}
```

#### 9. Expert picks with records and ROI

Today's public picks for the NBA and NHL.

```json
{
  "leagues": [
    "nba",
    "nhl"
  ],
  "daysAhead": 0,
  "includeMarkets": false,
  "includePicks": true
}
```

#### 10. Price history of tonight's moneylines

Adds `lineHistory`, the timestamped consensus price from the opener to the latest tick, to each moneyline row.

```json
{
  "leagues": [
    "nhl"
  ],
  "daysAhead": 0,
  "marketTypes": [
    "moneyline"
  ],
  "includeLineHistory": true
}
```

#### 11. UFC fight card

Fight moneylines with opening and current prices for the next week.

```json
{
  "leagues": [
    "ufc"
  ],
  "daysAhead": 7,
  "marketTypes": [
    "moneyline"
  ]
}
```

#### 12. Follow three teams

Every market of any game that involves these teams in the next six days.

```json
{
  "leagues": [
    "nba",
    "nfl",
    "nhl"
  ],
  "teams": [
    "lakers",
    "chiefs",
    "bruins"
  ],
  "daysAhead": 6
}
```

### Input reference

All fields are optional. With an empty input the Actor reads every league for today and the next two days and stops at 1,000 rows.

| Field | Type | Default | What it does |
|---|---|---|---|
| `leagues` | list | all | League slugs, listed in the coverage section. Names such as `premier-league` work too. `soccer` already contains the single soccer leagues, so naming both does not read them twice. Leagues without games in your dates return nothing and cost nothing. |
| `daysBack` / `daysAhead` | integer | 0 / 2 | Window around today in US Eastern game days. |
| `dateFrom` / `dateTo` | date | – | Exact window, `YYYY-MM-DD`. Overrides the two fields above. One date alone reads that single day. Windows are capped at 800 days per run. |
| `weeks` | text | – | For NFL, NCAA football and UFL (scheduled by week): `1-4,8`. Empty reads the weeks that touch your dates. |
| `season`, `seasonType` | number, select | current, `reg` | Season year and `reg` / `post` / `pre` for the weeks above. |
| `periods` | list | `["event"]` | `event` is the full game. Add `firsthalf`, `secondhalf`, `firstquarter` … `fourthquarter`, `firstperiod` … `thirdperiod`, `firstinning`, `firstfiveinnings`. Shortcuts `1h`, `2h`, `1q`, `1p`, `f5` work. |
| `includeMarkets` | boolean | true | The core market rows. Switch off for props-only or picks-only runs. |
| `marketTypes` | list | all | Keep `moneyline`, `spread`, `total`, `team_total`. |
| `gameStatus` | list | all | `scheduled`, `live`, `final`. |
| `teams` | list | – | Keep games where either side matches a name or abbreviation (partial, case-insensitive). |
| `onlyWithSplits` | boolean | false | Drop rows without tickets % / money %. |
| `books` | list | seven majors | Names (`draftkings`, `fanduel`, `betmgm`, `bet365`, `caesars`, `betrivers`, `unibet`, `novig`, `polymarket`, `prizepicks`) or Action Network book ids. The consensus and opening books are always added. |
| `includeBookPrices` | boolean | false | Add `bookPrice` rows. |
| `includeLineHistory` | boolean | false | Add `lineHistory` to market rows (one extra request per market). |
| `includeProps` | boolean | false | Add `prop` rows (NBA, WNBA, NCAAB, NFL, NCAAF, MLB, NHL). |
| `propTypes` | list | main props | `all`, `main`, a readable name (`points`, `passing_yards`, `strikeouts`, `hr`) or a full id (`27_points`, `core_bet_type_27_points`). |
| `includePicks` | boolean | false | Add `pick` rows. |
| `maxItems` | integer | 1000 | Total rows over all record types. `0` means no limit. |
| `proxyConfiguration` | proxy | none | The Actor reads Action Network directly and uses a small US residential rescue only if the direct route is refused. Your own proxy settings, if given, replace that rescue. |

Input is validated, not trusted: an unknown league is tried as a raw slug and skipped with a note if the site answers 400; an unreadable date, week or period is ignored with a note. Notes go to the log and to the `SUMMARY` record.

### Output: the data dictionary

Rows are flat JSON objects; missing values are `null`, never zero or an empty string. Numbers are numbers, times are ISO-8601 UTC, URLs are absolute. The **fill rate** column (percent of rows; **M** market, **P** prop, **K** pick, **B** bookPrice) is measured on real runs of 2026-10-29: market rows on the 12,468-row proof run (16 leagues requested, 11 with prices), prop, sportsbook-price and line-history rows on a 30,000-row run of the NBA, NHL and NFL boards, pick rows on 1,825 picks of NFL weeks 1-3 2025 and today's boards. A low fill rate is the data, not a bug: team totals carry no splits, a golf outright has no line, an unfinished game has no score, a finished tennis match has no prices.

#### Identity

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `rowId` | string | Stable unique id: m:<eventId>:<period>:<marketType>:<marketId>:<side>:<team> for market rows, p:... props, k:<pickId> picks, b:...:<bookId> sportsbook prices. The same game gives the same id on every run. | M 100 · P 100 · K 100 · B 100 | `m:295345:event:spr…` |
| `recordType` | string | market, prop, pick or bookPrice. | M 100 · P 100 · K 100 · B 100 | `market` |
| `scrapedAt` | string | ISO-8601 UTC time of the read; prices and splits are a snapshot of that moment. | M 100 · P 100 · K 100 · B 100 | `2026-10-29T08:47:1…` |

#### Event

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `eventId` | number | Action Network's game, fight or match id (end of the game-page URL). | M 100 · P 100 · K 100 · B 100 | `295345` |
| `eventType` | string | game (two teams) or competition (fight, match, tournament). | M 100 · P 100 · K 100 · B 100 | `game` |
| `eventName` | string | Away @ Home; A vs B for fights and matches; tournament title for golf. | M 100 · P 100 · K 100 · B 100 | `Cleveland Cavalier…` |
| `eventUrl` | string | Game page (team sports). | M 85.4 · P 100 · K 100 · B 100 | `https://www.action…` |
| `league` | string | League slug (the match's own for soccer). | M 100 · P 100 · K 100 · B 100 | `nba` |
| `leagueId` | number | Action Network's numeric league id. | M 100 · P 100 · K 100 · B 100 | `4` |
| `sport` | string | basketball, football, baseball, hockey, soccer, mma, boxing, tennis, golf. | M 100 · P 100 · K 100 · B 100 | `basketball` |
| `season` | number | Season year as Action Network labels it. | M 100 · P 100 · K 100 · B 100 | `2026` |
| `seasonType` | string | reg, post or pre. | M 85.4 · P 100 · K 100 · B 100 | `reg` |
| `week` | number | Week (NFL, NCAAF, UFL). | M 29 · P 51.7 · K 99.3 · B 58.3 | `8` |
| `startTime` | string | Scheduled start, ISO-8601 UTC. | M 100 · P 100 · K 100 · B 100 | `2026-10-30T00:00:0…` |
| `status` | string | scheduled, inprogress, complete, cancelled ... as served. | M 100 · P 100 · K 100 · B 100 | `scheduled` |
| `statusDetail` | string | Display text (Final - OT, FT). | M 90.9 · P 0 · K n/a · B 0 | `Final - OT` |
| `isLive` | boolean | In-play game or live price. | M 100 · P 100 · K 100 · B 100 | `false` |
| `isFinal` | boolean | Complete; consensus = closing line. | M 100 · P 100 · K 100 · B 100 | `false` |
| `awayTeam` | string | Away team or competitor. | M 88.3 · P 100 · K 100 · B 100 | `Cleveland Cavaliers` |
| `awayTeamId` | number | Away id. | M 88.3 · P 100 · B 100 | `167` |
| `awayTeamAbbr` | string | Away code. | M 88.3 · P 100 · K 100 · B 100 | `CLE` |
| `homeTeam` | string | Home team or competitor. | M 88.3 · P 100 · K 100 · B 100 | `Atlanta Hawks` |
| `homeTeamId` | number | Home id. | M 88.3 · P 100 · B 100 | `159` |
| `homeTeamAbbr` | string | Home code. | M 88.3 · P 100 · K 100 · B 100 | `ATL` |
| `awayScore` | number | Away score. | M 80.1 · P 0 · K 92.7 · B 0 | `125` |
| `homeScore` | number | Home score. | M 80.1 · P 0 · K 92.7 · B 0 | `127` |
| `totalBets` | number | Bets Action Network tracked on the game (its own sample). | M 85.4 · P 100 · B 100 | `1358` |

#### Market and selection

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `period` | string | event = full game; firsthalf, firstquarter, firstperiod ... for segments. | M 100 · P 100 · K 100 · B 100 | `event` |
| `marketType` | string | moneyline, spread, total or team_total; other markets keep Action Network's own id (matchup, core_bet_type_3_top_5 for golf top-5 finishes ...). | M 100 · B 100 | `spread` |
| `marketName` | string | Readable market name. | M 100 · B 100 | `Spread` |
| `marketId` | string | Market id, the same across sportsbooks. | M 100 · P 100 · K 94.1 · B 100 | `27785559` |
| `side` | string | away, home, over, under, draw, noside. | M 100 · P 100 · B 100 | `away` |
| `selection` | string | What you would bet on: team, fighter, Over/Under, Draw, player. | M 100 · P 100 · B 100 | `Cleveland Cavaliers` |
| `team` | string | Team the selection belongs to (team markets). | M 66.3 | `Cleveland Cavaliers` |
| `teamId` | number | Action Network team id of the selection. | M 66.3 | `167` |
| `competitorId` | number | Competitor id for fights and matches. | M 13.6 | `270042` |

#### Price, probability and splits

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `line` | number | Consensus handicap or total (null for moneylines); a spread's own handicap. | M 77.6 · P 48 · K 100 · B 43.4 | `1.5` |
| `oddsAmerican` | number | Consensus American price (bookPrice: that book's; pick: the expert's). | M 100 · P 100 · K 100 · B 100 | `-116` |
| `oddsDecimal` | number | Decimal odds, 4 decimals. | M 100 · P 100 · K 100 · B 100 | `1.8621` |
| `oddsFractional` | string | Exact reduced fraction of profit to stake. | M 100 · P 100 · B 100 | `25/29` |
| `impliedProbability` | number | Implied probability incl. margin, 0-1. | M 100 · P 100 · K 100 · B 100 | `0.537` |
| `noVigProbability` | number | Implied probability without margin (two-way markets, 3-way moneylines); null otherwise. | M 88.7 · P 47.7 | `0.5107` |
| `lineStatus` | string | normal, suspended, locked ... as served. | M 100 · P 100 · B 100 | `normal` |
| `ticketPercent` | number | Share of tickets on this side, 0-100; null when not published. | M 46.8 · B 1.3 | `52` |
| `moneyPercent` | number | Share of the money wagered on this side, 0-100. | M 46.8 · B 1.3 | `53` |
| `sharpGap` | number | moneyPercent - ticketPercent; positive = more money than tickets (sharp reading). | M 46.8 | `1` |

#### Opening line and movement

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `openLine` | number | Line at the open. | M 77.4 · P 48 | `1.5` |
| `openOddsAmerican` | number | American price at the open. | M 99.7 · P 100 | `-120` |
| `openOddsDecimal` | number | Decimal form of the opening price. | M 99.7 · P 100 | `1.8333` |
| `openImpliedProbability` | number | Implied probability of the opening price, 0-1. | M 99.7 · P 100 | `0.5455` |
| `lineMovement` | number | Current line minus opening line (0 = unchanged). | M 77.4 · P 48 | `0` |
| `oddsMovement` | number | Price change since the open in American points on a continuous scale (-105 to +100 = +5); positive = better for the bettor. | M 99.7 · P 100 | `4` |
| `lineHistory` | array | With includeLineHistory: \[{t, oddsAmerican, line, status}] from opener to latest tick. | M 100 | `[{"t":"2026-10-29T…` |

#### Best price and sportsbooks

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `bestOddsAmerican` | number | Highest price among your books at the consensus line. | M 94.3 · P 94 | `-118` |
| `bestOddsDecimal` | number | Decimal form of the best price. | M 94.3 · P 94 | `1.8475` |
| `bestBook` | string | Sportsbook offering the best price. | M 94.3 · P 94 | `FanDuel` |
| `bestBookId` | number | Action Network id of that sportsbook. | M 94.3 · P 94 | `69` |
| `bookCount` | number | Selected books quoting the consensus line. | M 100 · P 100 | `4` |
| `bookOdds` | array | Each selected book's price: bookId, book, oddsAmerican, oddsDecimal, line, impliedProbability, atConsensusLine; best first. | M 99.9 · P 96.7 | `[{"bookId":69,"boo…` |
| `priceSource` | string | consensus, or the book that stood in when there was no consensus price. | M 100 · P 100 | `consensus` |
| `result` | string | win / loss / push for finished full-game rows, graded from the score; else null. | M 82.6 | `win` |

#### Player props

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `propType` | string | Action Network's market id of the prop. | P 100 · K 52.1 | `core_bet_type_21_3…` |
| `statType` | string | Short stat name as the site shows it. | P 100 · K 52.1 | `3pt M` |
| `statName` | string | Full stat name. | P 100 | `3pt Shots Made` |
| `playerId` | number | Action Network player id. | P 99.6 · K 50.4 · B 98 | `281398` |
| `playerName` | string | Player name. | P 99.6 · B 98 | `Jaylon Tyson` |
| `playerTeamId` | number | Action Network id of the player's team. | P 99.2 | `167` |
| `playerTeam` | string | Team name. | P 99.2 | `Cleveland Cavaliers` |
| `playerTeamAbbr` | string | Team abbreviation. | P 99.2 | `CLE` |
| `playerPosition` | string | Position as served. | P 99.6 | `SG` |
| `jerseyNumber` | string | Jersey number. | P 97.7 | `20` |
| `opponent` | string | The opposing team of the player's team in this game. | P 99.2 | `Atlanta Hawks` |

#### Expert picks

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `pickId` | number | Pick id. | K 100 | `564248006` |
| `pickText` | string | Pick as posted. | K 100 | `P.Watson u1.5 3pt M` |
| `pickType` | string | spread / total / moneyline / custom. | K 100 | `custom` |
| `units` | number | Units risked. | K 100 | `1.32` |
| `unitsNet` | number | Net units won (0 pending). | K 100 | `0` |
| `pickResult` | string | pending / win / loss / push. | K 100 | `pending` |
| `isVerifiedPick` | boolean | Verified at posting. | K 100 | `true` |
| `bookId` | number | Sportsbook id. | K 100 · B 100 | `68` |
| `book` | string | Sportsbook name. | K 98.6 · B 100 | `DraftKings` |
| `pickedAt` | string | Posted, ISO UTC. | K 100 | `2026-10-29T05:39:0…` |
| `settledAt` | string | Graded, ISO UTC. | K 91.8 | `2026-10-30T03:12:1…` |
| `expertId` | number | User id. | K 100 | `1074136` |
| `expertName` | string | Public display name. | K 100 | `Kyle Murray` |
| `expertUsername` | string | Username. | K 100 | `kylemurray03` |
| `expertIsVerified` | boolean | Profile verified flag. | K 100 | `false` |
| `expertIsInternal` | boolean | Staff handicapper. | K 100 | `false` |
| `expertFollowers` | number | Followers. | K 100 | `343229` |
| `expertActivePicks` | number | Open picks. | K 100 | `4` |
| `expertRecordWindow` | string | Record window. | K 100 | `last_month` |
| `expertWins` | number | Wins. | K 100 | `56` |
| `expertLosses` | number | Losses. | K 100 | `46` |
| `expertPushes` | number | Pushes. | K 100 | `1` |
| `expertUnitsNet` | number | Net units. | K 100 | `10.8138` |
| `expertRoiPercent` | number | ROI %. | K 100 | `9.98` |
| `expertStreakType` | string | hot / cold. | K 100 | `cold` |
| `expertStreak` | number | Streak length. | K 100 | `3` |

#### Sportsbook price rows

| Field | Type | Meaning | Fill rate | Example |
|---|---|---|---|---|
| `atConsensusLine` | boolean | Book quotes the consensus line. | B 100 | `true` |
| `isBestPrice` | boolean | Best price among books at the consensus line. | B 100 | `true` |
| `consensusLine` | number | Consensus line, for comparison. | B 43.4 | `119.5` |
| `consensusOddsAmerican` | number | Consensus price, for comparison. | B 100 | `-115` |

#### A real market row

An NBA spread side from a finished game, as the Actor wrote it (from the proof run of 2026-10-29; the list of books is cut to the first two):

```json
{
  "rowId": "m:295140:event:spread:23924054:away:182",
  "recordType": "market",
  "eventId": 295140,
  "eventType": "game",
  "eventName": "Golden State Warriors @ Los Angeles Lakers",
  "eventUrl": "https://www.actionnetwork.com/nba-game/warriors-lakers-score-odds/295140",
  "league": "nba",
  "leagueId": 4,
  "sport": "basketball",
  "season": 2026,
  "seasonType": "reg",
  "week": null,
  "startTime": "2026-10-22T02:00:00.000Z",
  "status": "complete",
  "statusDetail": "Final",
  "isLive": false,
  "isFinal": true,
  "awayTeam": "Golden State Warriors",
  "awayTeamId": 182,
  "awayTeamAbbr": "GSW",
  "homeTeam": "Los Angeles Lakers",
  "homeTeamId": 186,
  "homeTeamAbbr": "LAL",
  "awayScore": 105,
  "homeScore": 127,
  "totalBets": 40414,
  "period": "event",
  "marketType": "spread",
  "marketName": "Spread",
  "marketId": "23924054",
  "side": "away",
  "selection": "Golden State Warriors",
  "team": "Golden State Warriors",
  "teamId": 182,
  "competitorId": null,
  "line": 6.5,
  "oddsAmerican": -108,
  "oddsDecimal": 1.9259,
  "oddsFractional": "25/27",
  "impliedProbability": 0.5192,
  "noVigProbability": 0.4957,
  "lineStatus": "normal",
  "ticketPercent": 70,
  "moneyPercent": 60,
  "sharpGap": -10,
  "openLine": 5.5,
  "openOddsAmerican": -115,
  "openOddsDecimal": 1.8696,
  "openImpliedProbability": 0.5349,
  "lineMovement": 1,
  "oddsMovement": 7,
  "bestOddsAmerican": -104,
  "bestOddsDecimal": 1.9615,
  "bestBook": "FanDuel",
  "bestBookId": 69,
  "bookCount": 5,
  "bookOdds": [
    {"bookId": 69, "book": "FanDuel", "oddsAmerican": -104, "oddsDecimal": 1.9615, "line": 6.5, "impliedProbability": 0.5098, "atConsensusLine": true},
    {"bookId": 68, "book": "DraftKings", "oddsAmerican": -108, "oddsDecimal": 1.9259, "line": 6.5, "impliedProbability": 0.5192, "atConsensusLine": true}
  ],
  "priceSource": "consensus",
  "result": "loss",
  "scrapedAt": "2026-10-29T08:56:05.688Z"
}
```

#### A real prop row

```json
{
  "rowId": "p:295345:event:core_bet_type_24_steals:27804922:over:63444",
  "recordType": "prop",
  "eventId": 295345,
  "eventType": "game",
  "eventName": "Cleveland Cavaliers @ Atlanta Hawks",
  "eventUrl": "https://www.actionnetwork.com/nba-game/cavaliers-hawks-score-odds/295345",
  "league": "nba",
  "leagueId": 4,
  "sport": "basketball",
  "season": 2026,
  "seasonType": "reg",
  "week": null,
  "startTime": "2026-10-30T00:00:00.000Z",
  "status": "scheduled",
  "statusDetail": null,
  "isLive": false,
  "isFinal": false,
  "awayTeam": "Cleveland Cavaliers",
  "awayTeamId": 167,
  "awayTeamAbbr": "CLE",
  "homeTeam": "Atlanta Hawks",
  "homeTeamId": 159,
  "homeTeamAbbr": "ATL",
  "awayScore": null,
  "homeScore": null,
  "totalBets": 1519,
  "period": "event",
  "propType": "core_bet_type_24_steals",
  "statType": "Stl",
  "statName": "Steals",
  "marketId": "27804922",
  "side": "over",
  "selection": "Donovan Mitchell Over",
  "playerId": 63444,
  "playerName": "Donovan Mitchell",
  "playerTeamId": 167,
  "playerTeam": "Cleveland Cavaliers",
  "playerTeamAbbr": "CLE",
  "playerPosition": "SG",
  "jerseyNumber": "45",
  "opponent": "Atlanta Hawks",
  "line": 1.5,
  "oddsAmerican": 112,
  "oddsDecimal": 2.12,
  "oddsFractional": "28/25",
  "impliedProbability": 0.4717,
  "noVigProbability": 0.4415,
  "lineStatus": "normal",
  "openLine": 1.5,
  "openOddsAmerican": 130,
  "openOddsDecimal": 2.3,
  "openImpliedProbability": 0.4348,
  "lineMovement": 0,
  "oddsMovement": -18,
  "bestOddsAmerican": 130,
  "bestOddsDecimal": 2.3,
  "bestBook": "bet365",
  "bestBookId": 79,
  "bookCount": 3,
  "bookOdds": [
    {"bookId": 79, "book": "bet365", "oddsAmerican": 130, "oddsDecimal": 2.3, "line": 1.5, "impliedProbability": 0.4348, "atConsensusLine": true},
    {"bookId": 69, "book": "FanDuel", "oddsAmerican": 112, "oddsDecimal": 2.12, "line": 1.5, "impliedProbability": 0.4717, "atConsensusLine": true}
  ],
  "priceSource": "consensus",
  "scrapedAt": "2026-10-29T09:24:55.238Z"
}
```

#### A real expert-pick row

```json
{
  "rowId": "k:564248006",
  "recordType": "pick",
  "scrapedAt": "2026-10-29T09:24:55.238Z",
  "eventId": 295345,
  "eventType": "game",
  "eventName": "Cleveland Cavaliers @ Atlanta Hawks",
  "eventUrl": "https://www.actionnetwork.com/nba-game/cavaliers-hawks-score-odds/295345",
  "league": "nba",
  "leagueId": 4,
  "sport": "basketball",
  "season": 2026,
  "seasonType": "reg",
  "week": null,
  "startTime": "2026-10-30T00:00:00.000Z",
  "status": "scheduled",
  "isLive": false,
  "isFinal": false,
  "awayTeam": "Cleveland Cavaliers",
  "awayTeamAbbr": "CLE",
  "homeTeam": "Atlanta Hawks",
  "homeTeamAbbr": "ATL",
  "awayScore": null,
  "homeScore": null,
  "pickId": 564248006,
  "pickText": "P.Watson u1.5 3pt M",
  "pickType": "custom",
  "propType": "core_bet_type_21_3fgm",
  "statType": "3pt M",
  "period": "game",
  "line": 1.5,
  "oddsAmerican": -132,
  "oddsDecimal": 1.7576,
  "impliedProbability": 0.569,
  "units": 1.32,
  "unitsNet": 0,
  "pickResult": "pending",
  "isVerifiedPick": true,
  "bookId": 68,
  "book": "DraftKings",
  "playerId": 198211,
  "marketId": "27804957",
  "pickedAt": "2026-10-29T05:39:06.198Z",
  "settledAt": null,
  "expertId": 1074136,
  "expertName": "Kyle Murray",
  "expertUsername": "kylemurray03",
  "expertIsVerified": false,
  "expertIsInternal": false,
  "expertFollowers": 343235,
  "expertActivePicks": 4,
  "expertRecordWindow": "last_month",
  "expertWins": 56,
  "expertLosses": 46,
  "expertPushes": 1,
  "expertUnitsNet": 10.8138,
  "expertRoiPercent": 9.98,
  "expertStreakType": "cold",
  "expertStreak": 3
}
```

#### CSV columns

Exported as CSV, a market-only run has these columns, in this order: `rowId`, `recordType`, `eventId`, `eventType`, `eventName`, `eventUrl`, `league`, `leagueId`, `sport`, `season`, `seasonType`, `week`, `startTime`, `status`, `statusDetail`, `isLive`, `isFinal`, `awayTeam`, `awayTeamId`, `awayTeamAbbr`, `homeTeam`, `homeTeamId`, `homeTeamAbbr`, `awayScore`, `homeScore`, `totalBets`, `period`, `marketType`, `marketName`, `marketId`, `side`, `selection`, `team`, `teamId`, `competitorId`, `line`, `oddsAmerican`, `oddsDecimal`, `oddsFractional`, `impliedProbability`, `noVigProbability`, `lineStatus`, `ticketPercent`, `moneyPercent`, `sharpGap`, `openLine`, `openOddsAmerican`, `openOddsDecimal`, `openImpliedProbability`, `lineMovement`, `oddsMovement`, `bestOddsAmerican`, `bestOddsDecimal`, `bestBook`, `bestBookId`, `bookCount`, `bookOdds`, `priceSource`, `result`, `scrapedAt`. The `bookOdds` and `lineHistory` columns hold JSON text in CSV and are real arrays in JSON. Runs with props, picks or sportsbook prices add the columns of those record types.

### Coverage: leagues, books, periods and how far back

#### Leagues

| Slug | League | Scheduled by | Notes |
|---|---|---|---|
| `nfl`, `ncaaf`, `ufl` | NFL, NCAA football, UFL | week | quarters, halves; props for NFL, NCAAF |
| `nba`, `wnba`, `ncaab`, `ncaaw` | basketball | day | quarters, halves; props for NBA, WNBA, NCAAB |
| `mlb`, `world_baseball_classic` | baseball | day | first inning, first five; props for MLB |
| `nhl` | NHL | day | periods; props |
| `soccer` | every soccer league | day | three-way moneyline with draw; `league` is the match's own league |
| `epl`, `laliga`, `bundesliga`, `seriea`, `ligue1`, `mls`, `champions`, `europa`, `worldcup`, `club_worldcup`, `euro` | single competitions | day | the same rows as inside `soccer` |
| `ufc`, `boxing` | MMA, boxing | day | one event per fight, the fighters in `awayTeam` / `homeTeam` |
| `atp`, `wta` | tennis | day | one event per match; prices only before and during a match |
| `pga` | PGA Tour golf | day | one event per tournament: outright, top 5 / 10 / 20, matchups; no `awayTeam` / `homeTeam` |

An out-of-season league returns an empty board; the Actor treats that as an answer, not an error. Unknown slugs are tried as typed.

#### Sportsbooks

Action Network lists 463 book ids (state sites of one brand are separate ids). The default is seven major books: DraftKings (68), FanDuel (69), BetMGM (75), bet365 (79), Caesars (123), BetRivers (71) and Unibet (247). `books` accepts those names, a few more (`novig`, `polymarket`, `prizepicks`) or any numeric id. Two ids are always read in addition, because the splits and the movement live there: **15 = Consensus** (Action Network's own price and the betting percentages) and **30 = Open** (the opening line).

#### Periods

`periods` defaults to `["event"]`, the full game (fights are read automatically). Other segments appear only when the league and game offer them: halves and quarters, hockey periods, baseball innings. A game that does not price a segment yields no rows for it.

#### How far back (measured 2026-10-29)

| League | Prices and splits | Notes |
|---|---|---|
| NBA | splits measured from 1 December 2023 (1 November 2023 has consensus prices with every split at zero, which the Actor reports as `null`) | opening line present for the dates measured |
| NFL | splits measured from week 10 of the 2023 season (week 1 of 2023 has none); 2024 and 2025 weeks measured with splits | weeks 1 and 10 of 2025 came back without the opening book, so the opening columns are `null` there |
| MLB | June 2024 and later measured | September 2023 returned results without prices |
| NHL | 2025 and 2026 measured | November 2023 and February 2024 returned no prices |
| NCAA football | 2024 weeks carry splits; 2023 week 5 did not | |
| Player props | past days are served too (NBA March 2024, January 2026; MLB August 2025 measured) | split columns on props are not published, so they are omitted |
| Line history | upcoming games and games of roughly the last two days | older games return the prices without the tick series |

Whatever Action Network serves for a day is what you get; what it does not serve is `null`.

### Scale and speed

Measured on the Apify platform on 2026-10-29, 1 GB of memory, HTTP only (no browser):

| Run | Result | Time |
|---|---|---|
| Prefill (7 leagues, today and tomorrow, 100 rows; the daily check) | 100 rows, 0 duplicates | 3.1 s, first rows within a second or two |
| 1,000 rows (all leagues, 9 days of history) | 1,000 rows, 0 duplicates | 2.5 s |
| **Proof: 1 October to 28 October 2026, all leagues** | **12,468 unique rows, 0 duplicate `rowId`s** from 1,105 games, fights and matches with prices in 11 leagues (1,807 events read; finished tennis matches and the like carry no prices), 376 requests, all HTTP 200 on the direct route | **62 s**, 90 MB of memory |
| Everything switched on (market + props + picks + sportsbook prices + line history) | 30,000 rows in one run, 0 duplicates | 12.7 s |

The proof rows split into soccer 3,301, NCAA football 2,895, NHL 1,926, golf 1,453, NBA 1,353, NFL 721, UFC 333, MLB 320, WNBA 130 and tennis 36, across the four core market types plus golf's top-finish and matchup markets. At about 11 rows per game, fight or match, roughly 900 events make 10,000 rows; the proof month had 1,105.

The Actor makes one request per league and day (per week for the weekly leagues), plus four props requests, one picks request and one history request per market when those are on. Requests start at most every 60 ms and run six at a time; the proof run read 375 scoreboards in about a minute. If the platform's own address is ever refused, requests move to a small, capped US residential rescue route and the run still finishes.

### What you pay

Pay per result, one price for every record type. A result is one row in the dataset.

| Plan | Per 1,000 results | 1,000 rows | 10,000 rows | 100,000 rows |
|---|---|---|---|---|
| Free | $0.50 | $0.50 | $5.00 | $50.00 |
| Bronze | $0.433 | $0.43 | $4.33 | $43.33 |
| Silver | $0.367 | $0.37 | $3.67 | $36.67 |
| Gold and above | $0.30 | $0.30 | $3.00 | $30.00 |

**What a game costs.** Every market side is its own row — both sides of the moneyline, the spread and the total, plus Over and Under for each team's total — so one game is several rows, not one. In the proof run (1-28 October 2026) 1,105 games, fights and matches produced 12,468 rows: **about 11 rows each**. A team-sport game with team totals is about 10 rows (6 without them), a soccer match 11, a UFC fight about 6, and a golf tournament several hundred because every player is a selection. At 11 rows a game costs about $0.0055 on Free and $0.0033 on Gold and above: a ten-game NBA night (about 100 rows) is roughly $0.05 on Free, a 14-game NFL week (about 140 rows) $0.07, and the whole proof month of every league (12,468 rows) $6.23 on Free or $3.74 on Gold and above.

Filters are applied before anything is written, so you pay only for rows you keep: `marketTypes`, `teams`, `gameStatus`, `onlyWithSplits`, `maxItems` and the record-type switches reduce the bill directly. With moneyline, spread and total only, an NBA game is six rows (about $0.003 on Free).

**Maximum cost per run.** Apify lets you cap what a run may spend. The Actor reads that cap and stops cleanly when the next row would exceed it — the run ends **SUCCEEDED** with the rows saved so far and a status message that says so, never aborted halfway. Set it in the run options or with `maxTotalChargeUsd` in the API call. `maxItems` is the other lever: a run never writes more rows than either limit allows.

### Integrations

```bash
curl -s -X POST "https://api.apify.com/v2/acts/logiover~action-network-scraper/run-sync-get-dataset-items?token=$APIFY_TOKEN&format=json" \
  -H "Content-Type: application/json" -d '{"leagues":["nfl"],"onlyWithSplits":true,"maxItems":300}' > nfl_splits.json
```

```javascript
import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('logiover/action-network-scraper').call({ leagues: ['nba', 'nhl'], daysAhead: 0, marketTypes: ['spread'], onlyWithSplits: true });
const { items } = await client.dataset(run.defaultDatasetId).listItems();
const sharp = items.filter((r) => r.sharpGap >= 15);
```

```python
from apify_client import ApifyClient
client = ApifyClient("APIFY_TOKEN")
run = client.actor("logiover/action-network-scraper").call(run_input={"leagues": ["nba"], "daysAhead": 0, "includeBookPrices": True, "marketTypes": ["spread", "total"]})
best = [r for r in client.dataset(run["defaultDatasetId"]).iterate_items() if r["recordType"] == "bookPrice" and r["isBestPrice"]]
```

**Schedules and time series.** One run is one snapshot, and splits and lines are a time series. Schedule the Actor hourly on game day, once more shortly before the start to capture the closing line, and key your table on `rowId` plus `scrapedAt`: the `rowId` of a market side is the same from first snapshot to last. A **webhook** on success can post the dataset URL to your own service. **Google Sheets, Excel, Airtable and Notion** read the CSV or Excel export, or the Google Sheets integration on a schedule. **Make, Zapier and n8n** run the Actor and fetch items, then branch on a column: a Slack, Telegram or Discord message when `sharpGap` crosses a threshold, a database row for every `bookPrice` with `isBestPrice`. **AI agents** can call the Actor through the Apify MCP server and answer "which NBA spread has the biggest gap between money and tickets tonight?" from one dataset read. **Warehouses and notebooks** (BigQuery, Snowflake, DuckDB, pandas) need no unnesting except the optional `bookOdds` and `lineHistory` arrays; `rowId` is the key.

### Performance and cost tips

- **Narrow before you scale.** `marketTypes: ["moneyline","spread","total"]` removes the four team-total rows per game; `onlyWithSplits` removes rows without percentages; `gameStatus: ["scheduled"]` keeps upcoming games.
- **Name the leagues.** Empty days cost almost nothing, but 400 days over every league is thousands of requests. Use `soccer` instead of eleven single soccer leagues.
- **`includeBookPrices` multiplies rows by the number of books;** the `bookOdds` array on every market row already carries the same prices.
- **Props are the heaviest request;** name the prop types instead of `all`. **Line history costs one request per market;** switch it on for the games you track.
- **Cap the run** with `maxItems` and Apify's maximum cost per run; both stop it cleanly with the rows already saved.

### Troubleshooting and status messages

The run status is **SUCCEEDED** for a normal finish, a partial finish and an empty answer; it is **FAILED** only when the source could not be read or the input cannot work. The final status message tells you which, and the `SUMMARY` record in the default key-value store has the counts (requests, routes, status codes, rows by type and league, notes).

| Outcome | Status message (shape) | What it means and what to do |
|---|---|---|
| COMPLETED | `Saved N rows from G games/matches across L leagues for <window>.` | Normal finish. |
| COMPLETED (row limit) | `Saved N rows — the limit of M was reached; K further day/league requests were not needed.` | `maxItems` stopped the run. Raise it for more. |
| COMPLETED (cost limit) | `Saved N rows — this run's maximum cost ($X) covers M results at the current price. Raise "Maximum cost per run" to collect more.` | The run's cost cap was reached; nothing was written beyond it. |
| PARTIAL | `Saved N rows, but K request(s) did not finish [before the run deadline] (<last error>).` | Some days or leagues could not be read. Run again for the missing part; `SUMMARY.counts` shows what failed. |
| EMPTY_RESULT | `No games with odds for <leagues> on <window>. Out-of-season leagues and days without a slate return nothing — widen the dates or add leagues.` | The source answered; there is nothing to return. Succeeds with an empty dataset. |
| BLOCKED / FETCH_FAILED | `No data could be fetched from Action Network (<error>) after N attempt(s). This is a source-access failure, not an empty result.` or `No rows saved and N request(s) could not be completed, so an empty result cannot be confirmed (<error>).` | The run fails on purpose, because silence would look like "no games". Try again later; `SUMMARY.statusCounts` shows 403 / 429 / timeouts. |
| INVALID_INPUT | `No league to read. Leave "leagues" empty for every league, or name some (nba, nfl, nhl, mlb, soccer, ufc ...).` | The input named nothing usable. |

Warnings in the log and in `SUMMARY.notes`: an unknown league slug ("trying it as the Action Network slug", then "does not serve a league called ... (HTTP 400) — skipped"), an unreadable date, week, period, sportsbook or prop type ("Ignored ..."), a window cut to 800 days, swapped dates, a soccer league dropped because `soccer` contains it, and a built-in book-name fallback when the sportsbook list could not be read.

**Few rows?** Check the dates (US Eastern game days), the league's season, and that `onlyWithSplits`, `gameStatus` and `teams` are not removing everything. **No splits?** Team totals, props and far-out games carry none. **`result` empty?** Only finished full-game rows are graded.

### Honest limits

- **A snapshot, not a tick archive.** Each run shows the prices at that moment. The optional `lineHistory` returns the consensus tick series only for upcoming games and games of roughly the last two days (measured 2026-10-29); for longer history, schedule runs.
- **Splits are Action Network's sample.** `totalBets` and the percentages describe the bets Action Network tracks, not the whole market. They are not published for every market (team totals, props) and not for every day in the past (see the coverage table).
- **Opening line coverage varies.** The opening book is missing for some seasons (measured: NFL 2025 weeks 1 and 10). Where it is missing the opening and movement columns are `null`.
- **One consensus line.** A book quoting another line is listed with `atConsensusLine: false` and never counts as the best price.
- **Not every sportsbook quotes every market.** `bookCount` can be zero and `bestOddsAmerican` then empty. State sites of one brand are separate ids.
- **No futures, injuries, standings, box scores or play-by-play.** The Actor is built for splits, prices, movement, props and picks.
- **Not betting advice.** Odds change between a snapshot and a bet, and the source can contain errors.
- **Anonymous data only.** Anything behind an Action Network account (PRO tools) is neither read nor offered.
- **The source can change** without notice; the Actor reports failures instead of hiding them.

### How it compares with the alternatives

| Option | What you get | The catch |
|---|---|---|
| Copying numbers from the website | exactly what you see, for the games you open | hours per week, no history, easy to mistype |
| A generic odds API | prices from many books, usually without public betting percentages | monthly plans; the splits and sharp-money view are what it lacks |
| A browser scraper | anything visible on a page | slow, heavy, breaks on layout changes |
| **This Actor** | splits, consensus, opening and best prices, movement, results, props and picks as flat rows, from the site's own JSON | pay per row, no subscription |

Action Network has no documented public API; this Actor adds a stable schema, scheduling, retries, exports and a cost cap on top of its public feed.

### FAQ

#### Is there an Action Network API?

Not a public, documented one. The website runs on a JSON feed anyone can read; this Actor turns it into a stable schema with scheduling and exports.

#### How do I scrape Action Network public betting percentages?

Run the Actor for your leagues and read `ticketPercent` and `moneyPercent` on the market rows. Set `onlyWithSplits` to drop rows without them.

#### What is the difference between tickets % and money %?

Tickets % counts bets, money % weighs them by stake. Many small bets on one side and a few big ones on the other give the first side a high ticket share and the second a high money share.

#### What does sharp money mean, and how do I find it here?

Money from large, informed bettors. A common proxy is a side whose money share is far above its ticket share. `sharpGap` is `moneyPercent - ticketPercent`; sort by it and check `lineMovement`.

#### How do I find reverse line movement?

Look for rows where `ticketPercent` is above 50 while `lineMovement` shows the number moving against that side. The Actor supplies the columns; it does not label the pattern for you.

#### Can I get closing line value (CLV)?

Yes. For a finished game the consensus price is the closing line; `openOddsAmerican` and `openLine` are the opener. Compare your price's `impliedProbability` with the closing one. `result` gives the return.

#### How far back does the data go?

Prices and splits are served from the 2023-24 seasons; see the coverage table for each league. Earlier days return results without prices.

#### Does it need a login or an API key?

No. It reads what the website shows anonymous visitors.

#### Is it legal to scrape Action Network?

The Actor reads publicly served data without logging in. Laws and terms differ by country and use, and you are responsible for yours; see the legality section.

#### Which sports and sportsbooks are supported?

The leagues in the coverage table; by default DraftKings, FanDuel, BetMGM, bet365, Caesars, BetRivers and Unibet plus the consensus and opening lines. Others, including Novig, Polymarket US and PrizePicks, or any of the 463 book ids, can be named.

#### What are book 15 and book 30?

15 is Action Network's consensus line, which carries the betting percentages; 30 is the opening line. They are added to every request.

#### What does `bestBook` mean?

The sportsbook with the highest price among your selected books that quotes the consensus line. Empty when nobody quotes it.

#### How do I get decimal or fractional odds?

Every row has `oddsDecimal` and `oddsFractional` next to `oddsAmerican`, plus `impliedProbability`.

#### What is `noVigProbability`?

The implied probability with the margin removed across the sides of the market. Empty where the sides are not exclusive.

#### Can I get player props and expert picks?

Yes: `includeProps` and `includePicks`. Props come with the player, team, position, opponent, line, prices and best price; picks with the handicapper's record, units and ROI.

#### Can I get line movement history?

The opening line and the current price are on every market row. The timestamped consensus series (`lineHistory`) is available for upcoming games and games of about the last two days.

#### How do I track a line over time?

Schedule the Actor and key your table on `rowId` and `scrapedAt`; the `rowId` of a market side is stable.

#### How do I get one row per sportsbook?

Switch on `includeBookPrices`; each book's price on each side becomes a `bookPrice` row with `isBestPrice`.

#### Why are some `ticketPercent` and `moneyPercent` values empty?

Action Network does not publish splits for every market (team totals, props) or for games far in the future. The Actor writes `null` instead of a false 0 / 0.

#### What does `result` mean?

`win`, `loss` or `push` for the selection of a finished full-game moneyline, spread, total or team total, computed from the final score and the consensus line. Anything else is `null`.

#### Why does my NFL run ignore the date?

NFL, college football and UFL are scheduled by week. Use `weeks`, `season` and `seasonType`, or leave them empty to read the weeks that touch your dates.

#### Why does a golf row have no `awayTeam`?

A golf event is a tournament with a field of players, not a game between two sides; the players are in `selection`.

#### How fresh is the data, and how many rows will I get?

It is read when the run starts. About ten rows per NBA, NFL, NHL or MLB game with team totals, six without, eleven for a soccer match (about 11 across the proof month); a full day across all leagues is a few hundred rows, a month about 12,000.

#### How much does it cost?

$0.50 per 1,000 results on the Free plan, down to $0.30 on Gold and above. Every market side is its own row, so one game is about 11 rows (measured: 1,105 games and fights, 12,468 rows) — about half a cent on Free. `maxItems` and Apify's maximum cost per run cap a run.

#### What happens if the website blocks the Actor?

It retries, then falls back to a small, capped US residential route. If nothing works the run fails with a clear message instead of returning an empty dataset that looks like "no games".

#### Can I use my own proxy?

Yes, in the proxy field; it replaces the built-in rescue route. The normal route is direct and needs no proxy.

#### Can I export to Excel, CSV or Google Sheets?

Yes: JSON, CSV, Excel, XML, HTML and RSS from the dataset, or the Google Sheets integration on a schedule.

#### Does the Actor place bets or touch accounts?

Never. It reads public pages and writes a dataset.

#### Can I find arbitrage or +EV bets with it?

It supplies the ingredients: many books' prices on one line, implied and no-vig probabilities and the best price per side. Scanning, staking and deciding to bet are yours.

#### Does it work for games months away?

Lines for far-out games exist for some markets and are read with the window you give. Splits usually appear in the last day or two before a game.

#### Can I read only the NFL, only one team or only upcoming games?

Yes: `leagues`, `teams` and `gameStatus` filter before anything is written, and you pay only for the rows you keep.

#### Who do I ask if a column looks wrong?

Open an issue with the run id and input. We reply within 24 hours.

### Glossary

- **Betting splits** — the share of tickets and of money on each side. **Tickets** — individual bets; **handle** — money wagered.
- **Sharp money / public money** — large informed wagers / many small recreational ones. **Sharp gap** — money % minus tickets %.
- **Reverse line movement** — the line moving against the side most tickets are on.
- **Opening / closing line** — the first and last number before the game; **CLV** compares your price with the close.
- **Moneyline, spread (point spread, run line, puck line), total (over/under), team total, prop, ladder** — the bet types; a ladder is a family of "N or more" props.
- **American, decimal, fractional odds** — three ways to write one price; **implied probability** includes the margin, **no-vig probability** removes it.
- **Juice / vig / hold** — the sportsbook's margin. **Line shopping** — taking the best price among books. **Push** — a tie with the line.
- **Consensus / Open** — Action Network's published and opening price. **Unit / ROI** — a handicapper's stake unit; return on units risked.

### 🤖 AI Agents, MCP & RAG Integration

**Action Network Scraper** works natively as an on-demand data tool for AI assistants and autonomous workflows.

Connect this Actor directly to **Apify AI**, **Claude Desktop**, **Cursor**, **VS Code**, **Windsurf**, **LangChain**, or **LlamaIndex** via the official **Apify MCP Server** — no custom wrapper code needed:

```json
{
  "mcpServers": {
    "apify": {
      "type": "http",
      "url": "https://mcp.apify.com?tools=logiover/action-network-scraper"
    }
  }
}
```

- **Schema-typed inputs & outputs:** Clean JSON records with explicit `.actor/input_schema.json` and `.actor/output_schema.json` definitions so LLM agents infer parameters and parse fields accurately without HTML noise.
- **Bounded & predictable runs:** Built-in volume caps and pay-per-result budget guards prevent runaway agent tool loops.
- **RAG & automation ready:** Stream run datasets straight into vector stores, n8n, Make, Zapier, or webhook pipelines.

### Legality and responsible use

This Actor reads public data that Action Network serves to anonymous visitors: odds, aggregate betting percentages, the public profiles and picks of experts, team and player names on prop markets. It does not log in, bypass any access control or touch private data. Web scraping of public data is generally lawful in the United States and the European Union, but the rules depend on the country, on the site's terms, on what you collect and on what you do with it. You are responsible for your use. If you combine the data with personal data, data-protection rules such as GDPR apply to you as the controller. Expert names and usernames are public profile data; treat them accordingly.

Sports betting is regulated in many places and illegal in some. Nothing here is betting advice or a promise of profit; bets can lose, and if gambling stops being fun, seek help in your country.

### Changelog

#### 2026-11-14

- Verified live production health run (2026-11-14): extracted **35** structured records in **2.5s** with valid schema output.
- Hardened internal proxy stall fallback and pay-per-event dataset instance budget protection.

#### 2026-11-11

- Added reliable terminal run status telemetry and structured Apify AI / MCP Server (`mcp.apify.com`) tool configuration.
- Added direct Issues & Reviews support links and refreshed live maintenance verification.

#### 2026-11-01

- Monthly health check (2026-11-01): a live run on the current build completed successfully and returned data. Input options, output fields and pricing are unchanged.

#### 1.0 (2026-10-29)

First release. Market, prop, expert-pick and sportsbook-price rows from Action Network's public JSON for 27 leagues; consensus, opening and best prices in four odds formats plus no-vig probability; betting splits with `sharpGap`; movement; graded results; optional consensus price history; date and week windows; filters; outcome statuses with a `SUMMARY` record; clean stop at the maximum cost per run. Price set to $0.50 per 1,000 results (Free) down to $0.30 (Gold and above). Proof on build 1.0.6 (later builds changed only README, metadata and an internal test switch; extraction unchanged): 12,468 unique rows in one run (62 s, 0 duplicates).

### 💬 Support, Issues & Reviews

- **Bug reports & target-site changes:** Hit an unexpected empty result or selector change? [Open a ticket in the Issues tab](https://apify.com/logiover/action-network-scraper/issues) with your **Run ID** or sample input — we actively monitor issues and ship fixes fast.
- **Feature & field requests:** Need an additional data attribute or filter? [Request it in the Issues tab](https://apify.com/logiover/action-network-scraper/issues).
- **Enjoying this Actor?** If it saved you engineering time, please consider leaving an honest [review on Apify Store](https://apify.com/logiover/action-network-scraper/reviews) or bookmarking [logiover/action-network-scraper](https://apify.com/logiover/action-network-scraper) — organic developer feedback helps others discover reliable tools and keeps maintenance prioritized.

# Changelog

This Actor's version history is a separate document: https://apify.com/logiover/action-network-scraper/changelog.md

# Actor input Schema

## `leagues` (type: `array`):

Which leagues to read. Leave EMPTY for every league (NFL, NCAA football, NBA, NCAA basketball, WNBA, MLB, NHL, all soccer, UFC, boxing, ATP/WTA tennis, PGA golf ...); a league with no games in your dates simply returns nothing. Use Action Network's slugs: nfl, ncaaf, ufl, nba, ncaab, ncaaw, wnba, mlb, nhl, soccer (every soccer league at once), epl, laliga, bundesliga, seriea, ligue1, mls, champions, europa, worldcup, ufc, boxing, atp, wta, pga. Common names such as premier-league or champions-league are understood. 'soccer' already contains the individual soccer leagues.

## `daysBack` (type: `integer`):

How many days before today (US Eastern) to include. 0 = start today. Finished games carry their closing line, final score and a graded result.

## `daysAhead` (type: `integer`):

How many days after today to include. Lines for games further out exist but splits usually appear only in the last day or two before the game.

## `dateFrom` (type: `string`):

Exact start day (YYYY-MM-DD, US Eastern game day). Overrides Days back / Days ahead. Prices and splits are served from the 2023-24 seasons onward; earlier days return results without prices.

## `dateTo` (type: `string`):

Exact end day (YYYY-MM-DD). Leave empty to read only the From date.

## `weeks` (type: `string`):

These leagues are scheduled by week, not by day. Leave empty to read the weeks that touch your dates, or name weeks like 1-4,8. Combine with Season and Season type for past seasons.

## `season` (type: `integer`):

Season year for the weeks above, for example 2025. Empty = the current season.

## `seasonType` (type: `string`):

Regular season, postseason or preseason.

## `periods` (type: `array`):

Full game is 'event'. Add segments to read their lines too: firsthalf, secondhalf, firstquarter ... fourthquarter (football, basketball), firstperiod ... thirdperiod (hockey), firstinning, firstfiveinnings (baseball). 1h, 2h, 1q, 1p and f5 work as shortcuts. Not every game prices every segment.

## `includeMarkets` (type: `boolean`):

The core rows: moneyline, spread, total and team totals with betting splits, opening line, movement and best price. Switch off to read only props or picks.

## `marketTypes` (type: `array`):

Keep only these market types. Empty = all.

## `gameStatus` (type: `array`):

scheduled = not started, live = in progress, final = finished. Empty = all. Useful to keep only upcoming games.

## `teams` (type: `array`):

Keep only games where one side matches (name or abbreviation, case-insensitive, partial match): lakers, KC, djokovic ...

## `onlyWithSplits` (type: `boolean`):

Drop market sides for which Action Network publishes no tickets % / money % (team totals, far-out games, props).

## `books` (type: `array`):

Which sportsbooks to price against. Names (draftkings, fanduel, betmgm, bet365, caesars, betrivers, unibet, novig, polymarket, prizepicks) or Action Network book ids (463 exist; the list is at /web/v1/books). The consensus and opening lines are always included, so splits and movement never go missing. Empty = the seven major books.

## `includeBookPrices` (type: `boolean`):

Add recordType 'bookPrice' rows: every sportsbook's price on every market side as its own row, ready for line-shopping tables. Multiplies the row count by roughly the number of books.

## `includeLineHistory` (type: `boolean`):

Add the timestamped consensus price series (opener to latest tick) to each market row as lineHistory. One extra request per market.

## `includeProps` (type: `boolean`):

Add recordType 'prop' rows: player props (points, rebounds, passing yards, strikeouts, anytime scorers ...) with consensus line, opening line, best price and every book's price. NBA, WNBA, NCAAB, NFL, NCAAF, MLB, NHL.

## `propTypes` (type: `array`):

Which props. Empty = the main player props of each league. 'all' adds every market the league's props page lists, including the 'N or more' ladders. Or name them: points, rebounds, assists, 3fgm, passing_yards, strikeouts, hr, 27_points ...

## `includePicks` (type: `boolean`):

Add recordType 'pick' rows: public expert picks with the handicapper's win-loss record, units and ROI over the recent window.

## `maxItems` (type: `integer`):

Stop after this many rows in total (all record types). 0 = no limit.

## `proxyConfiguration` (type: `object`):

The scraper reads Action Network directly and falls back to a small US residential rescue only if the direct route is refused. Set your own proxy here to use it for that rescue instead.

## Actor input object example

```json
{
  "leagues": [
    "nba",
    "nhl",
    "nfl",
    "mlb",
    "ncaaf",
    "ncaab",
    "soccer"
  ],
  "daysBack": 0,
  "daysAhead": 1,
  "seasonType": "reg",
  "periods": [
    "event"
  ],
  "includeMarkets": true,
  "onlyWithSplits": false,
  "books": [
    "draftkings",
    "fanduel",
    "betmgm",
    "bet365",
    "caesars",
    "betrivers",
    "unibet"
  ],
  "includeBookPrices": false,
  "includeLineHistory": false,
  "includeProps": false,
  "includePicks": false,
  "maxItems": 100,
  "proxyConfiguration": {
    "useApifyProxy": false
  }
}
```

# Actor output Schema

## `results` (type: `string`):

All rows collected in this run.

# API

You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup.

## JavaScript example

```javascript
import { ApifyClient } from 'apify-client';

// Initialize the ApifyClient with your Apify API token
// Replace the '<YOUR_API_TOKEN>' with your token
const client = new ApifyClient({
    token: '<YOUR_API_TOKEN>',
});

// Prepare Actor input
const input = {
    "leagues": [
        "nba",
        "nhl",
        "nfl",
        "mlb",
        "ncaaf",
        "ncaab",
        "soccer"
    ],
    "daysBack": 0,
    "daysAhead": 1,
    "periods": [
        "event"
    ],
    "books": [
        "draftkings",
        "fanduel",
        "betmgm",
        "bet365",
        "caesars",
        "betrivers",
        "unibet"
    ],
    "maxItems": 100
};

// Run the Actor and wait for it to finish
const run = await client.actor("logiover/action-network-scraper").call(input);

// Fetch and print Actor results from the run's dataset (if any)
console.log('Results from dataset');
console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`);
const { items } = await client.dataset(run.defaultDatasetId).listItems();
items.forEach((item) => {
    console.dir(item);
});

// 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs

```

## Python example

```python
from apify_client import ApifyClient

# Initialize the ApifyClient with your Apify API token
# Replace '<YOUR_API_TOKEN>' with your token.
client = ApifyClient("<YOUR_API_TOKEN>")

# Prepare the Actor input
run_input = {
    "leagues": [
        "nba",
        "nhl",
        "nfl",
        "mlb",
        "ncaaf",
        "ncaab",
        "soccer",
    ],
    "daysBack": 0,
    "daysAhead": 1,
    "periods": ["event"],
    "books": [
        "draftkings",
        "fanduel",
        "betmgm",
        "bet365",
        "caesars",
        "betrivers",
        "unibet",
    ],
    "maxItems": 100,
}

# Run the Actor and wait for it to finish
run = client.actor("logiover/action-network-scraper").call(run_input=run_input)

# Fetch and print Actor results from the run's dataset (if there are any)
print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}")
for item in client.dataset(run.default_dataset_id).iterate_items():
    print(item)

# 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start

```

## CLI example

```bash
echo '{
  "leagues": [
    "nba",
    "nhl",
    "nfl",
    "mlb",
    "ncaaf",
    "ncaab",
    "soccer"
  ],
  "daysBack": 0,
  "daysAhead": 1,
  "periods": [
    "event"
  ],
  "books": [
    "draftkings",
    "fanduel",
    "betmgm",
    "bet365",
    "caesars",
    "betrivers",
    "unibet"
  ],
  "maxItems": 100
}' |
apify call logiover/action-network-scraper --silent --output-dataset

```

## MCP server setup

```json
{
    "mcpServers": {
        "apify": {
            "type": "http",
            "url": "https://mcp.apify.com/?tools=fetch-actor-details,logiover/action-network-scraper"
        }
    }
}
```

The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer <APIFY_API_TOKEN>` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations).

## OpenAPI specification

Download the OpenAPI definition: https://api.apify.com/v2/actors/FF61LJvcVZa9IG7Ru/builds/7Fa1Rpv2h7SxyQ8Hw/openapi.json
