Skip to main content
Glama
DanielTomaro13

sportsdata-mcp

puntersedge_racing_closing_lines

Read-onlyIdempotent

Access archived racing closing lines with opening prices and finishing positions to calculate CLV or backtest betting models.

Instructions

The permanent closing-line archive: one row per race/runner/bookmaker with the last price before the jump, the first price after the market opened, and the finishing position. The CLV and backtesting surface. Costs 5 credits (JSON). Needs Standard or higher — Free and Hobby get a 403. Standard reads a rolling 90 days, Plus and above the whole archive; window_days on the response says which.

Returns: {archive_from, limit, offset, total_rows, rows_returned, resulted_rows, rows:[{race_id, venue, venue_id, venue_site, race_number, category, country, start_time, meeting_date_aet, bookmaker_key, runner_name, runner_number, runner_key, runner_ref, open_win_price, close_win_price, open_secs_to_jump, close_secs_to_jump, points_observed, is_closing_line, open_is_baseline, finish_position, result_status, venue_split_suspect, name_fragment_suspect}]} — an ENVELOPE with rows, not a bare array. Read the flags before modelling: only 85.5% of series have their final observation within 300s of the jump.

NOTE: this shape is from the vendor's documentation and has NOT been verified against a live response (we hold no key for this provider). Treat it as approximate — inspect the actual payload before relying on a field name.

Example: Yesterday's AU thoroughbred closing lines with results attached {"date": "", "category": "horse", "country": "AU", "resulted_only": true, "limit": 500}

Auth: needs your own key in PUNTERSEDGE_API_KEY.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
dateNoSingle-day alias, YYYY-MM-DD — exactly from=<date>&to=<date>. Mutually exclusive with from/to.
limitNoRows to return.
venueNoVenue name, case-insensitive, exact match.
formatNocsv streams a flat table with a stable column order, and costs 20 credits instead of 5.json
offsetNoRows to skip, for paging.
countryNoTwo-letter country code, e.g. AU or NZ.
date_toNoISO date or datetime. Defaults to now. Mutually exclusive with `date`. Sent upstream as `to`.
race_idNoA single race, joinable to the live price-history store while that race is still inside its 45-day snapshot window.
categoryNoOne racing code. Singular and single-valued here, unlike `categories` on the other racing tools. One of: horse, harness, greyhound.
date_fromNoISO date or datetime on the race start. Defaults to the archive floor. Mutually exclusive with `date`. Sent upstream as `from`.
bookmakersNoComma-separated bookmaker keys, case-insensitive. An unrecognised key returns a free 422 naming the valid keys.
closing_onlyNoKeep only rows whose last observation was within 300s of the jump. TRUE by default because 14.5% of archived series are not closing lines, and silently mixing them in is how a CLV study goes wrong.
resulted_onlyNoKeep only rows from races that have a result — set true when you need finish_position on every row.
include_flaggedNoInclude rows flagged venue_split_suspect or name_fragment_suspect. False by default; set true only if you intend to handle the contaminated rows.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv0.33.0

TDQS

A4.4/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations cover read-only/idempotent, and the description adds substantial context beyond them: credit cost, tier gating, the 90-day rolling vs full-archive window, the envelope return shape, and a candid caveat that the vendor-documented shape is unverified. The 85.5% flag warning is exactly the kind of behavioral nuance an agent needs before modelling.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the identity and cost/access facts before the return shape and caveats, with a worked example. Dense but nearly every clause carries distinct information; the full return-field list is long though arguably necessary given no output schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With no output schema, the description fully carries the return contract (envelope with `rows`, all fields listed) plus auth, tiering, cost, and a reliability caveat. An agent has everything needed to call it and interpret the payload.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so each of the 14 parameters is already documented upstream; the description adds only the note that `window_days` appears on the response. Baseline 3 is appropriate when the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a specific verb+resource ('the permanent closing-line archive: one row per race/runner/bookmaker') and immediately states its role as 'the CLV and backtesting surface'. An agent can distinguish it from puntersedge_racing_results or movers without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete operational context: cost (5 credits), required tier (Standard+, Free/Hobby 403), and the rolling-window behavior per tier. However it does not explicitly compare against sibling tools like puntersedge_racing_events or puntersedge_racing_results to say when to choose this over those.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools