DunkAPI

Prop backtest

get_prop_backtest

Read-onlyIdempotent

Use to test an idea about player props without pulling every row: how often a kind of prop went over or under against what the market implied, across every finished game we hold prices for that fits your filters. Returns the hit rate, its 95% interval, the average implied and fair probabilities, and each season on its own. For example, points overs for players whose team was the road underdog. One result per player per game at the median closing line. It describes the past and recommends nothing. For one player game by game use get_prop_history. Costs 25 credits.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "market": "<market>"
    },
    "name": "get_prop_backtest"
  }
}

Input Schema

Name	Required	Description
`side`	No	over (default) or under. Yes/no markets are graded on the yes price: over only.
`team`	No	The player's team, e.g. BOS.
`league`	No	League. Defaults to nba. One of: nba, wnba, ncaam.
`market`	Yes	The market, e.g. player_points.
`to_date`	No	Last game date, YYYY-MM-DD. Default all held.
`fair_max`	No	Highest fair probability of this side to include, 0 to 1.
`fair_min`	No	Lowest fair probability of this side to include, 0 to 1.
`line_max`	No	Highest median closing line to include.
`line_min`	No	Lowest median closing line to include.
`opponent`	No	The team he played.
`from_date`	No	First game date, YYYY-MM-DD. Default all held.
`home_away`	No	Whether his team was at home.
`player_id`	No	One player's id (e.g. 3945274, from search_players).
`team_role`	No	Whether his team was favored, by the books' closing moneylines.

Output Schema

Name	Required	Description	Default
`hits`	Yes
`note`	No
`side`	Yes
`tool`	No
`games`	Yes
`label`	Yes
`market`	Yes
`method`	Yes
`misses`	Yes
`pushes`	Yes
`caution`	Yes
`credits`	No
`filters`	Yes
`players`	Yes
`results`	Yes
`avg_line`	Yes
`hit_rate`	Yes
`no_price`	Yes
`avg_books`	Yes
`by_season`	Yes
`no_result`	Yes
`avg_actual`	Yes
`avg_fair_prob`	Yes
`estimate_only`	No
`avg_implied_prob`	Yes
`hit_rate_minus_fair`	Yes
`hit_rate_interval_95`	Yes
`hit_rate_minus_implied`	Yes
`market_inside_interval`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

AddedNov 7, 2026

TDQS

A4.3/5.0

Behavior4/5

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

Annotations already declare readOnly, idempotent, closed-world and non-destructive, so the safety profile is covered. The description adds genuinely non-obvious behavior: a 25-credit cost, aggregation semantics ("one result per player per game at the median closing line"), and the interpretive caveat that it describes the past and recommends nothing. The enumeration of return fields is partly redundant with the existing output schema.

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?

The purpose is front-loaded in the first clause and nearly every sentence carries information an agent needs. It is a dense single block, though, and the sibling routing and cost note are buried mid-paragraph rather than surfaced where an agent scanning for alternatives would find them.

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

Completeness4/5

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

For a 14-parameter aggregate tool with a full output schema and complete annotations, the description supplies the missing conceptual layer: what a 'result' means, the closing-line basis, and the credit cost. Minor gaps remain around behavior when filters match nothing, but nothing essential to correct invocation is absent.

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

Parameters4/5

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

Schema description coverage is 100%, so the baseline is 3, but the description adds meaning beyond the schema: the worked example ("points overs for players whose team was the road underdog") shows how filters compose, and the median-closing-line statement clarifies which line the line_min/line_max filters apply to.

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?

The description names a specific verb and resource: compute hit rate of a prop kind versus market-implied probability across finished games matching filters. It also distinguishes itself from the nearest sibling by saying "For one player game by game use get_prop_history," so an agent can separate the two without opening a 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?

It gives a clear triggering condition ("test an idea about player props without pulling every row") plus a concrete example filter set, and it routes the single-player case to get_prop_history. It does not cover when-not-to-use relative to the other aggregation siblings such as get_prop_probabilities or get_prop_movement, so it falls short of full alternative mapping.

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

All DunkAPI tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.