get_prop_backtest

25 creditsBacktestsPOST /v1/get_prop_backtest

How often a kind of player prop hit against what the market implied, by your filters, since May 2023

Use it when you want to test an idea about props before building anything: how often a kind of prop hit against what the market implied, by season, with the sample size and interval.

Parameters

marketenum: player_points | player_rebounds | player_assists | player_threes | player_points_rebounds_assists | player_blocks | player_steals | player_points_rebounds | player_points_assists | player_rebounds_assists | player_double_double | player_triple_doublerequired: The market, e.g. player_points.
sideenum: over | underoptional: over (default) or under. Yes/no markets are graded on the yes price: over only.
from_datestringoptional: First game date, YYYY-MM-DD. Default all held.
to_datestringoptional: Last game date, YYYY-MM-DD. Default all held.
player_idstringoptional: One player's id (e.g. 3945274, from search_players).
teamstringoptional: The player's team, e.g. BOS.
opponentstringoptional: The team he played.
home_awayenum: home | awayoptional: Whether his team was at home.
team_roleenum: favorite | underdogoptional: Whether his team was favored, by the books' closing moneylines.
line_minnumberoptional: Lowest median closing line to include.
line_maxnumberoptional: Highest median closing line to include.
fair_minnumberoptional: Lowest fair probability of this side to include, 0 to 1.
fair_maxnumberoptional: Highest fair probability of this side to include, 0 to 1.
leagueenum: nba | wnba | ncaamoptional: League. Defaults to nba. One of: nba, wnba, ncaam.

Notes

Over MCP, ask for it in plain English and the client calls get_prop_backtest itself. Point any MCP client at https://mcp.dunkapi.com/mcp?key=YOUR_API_KEY.
Each call costs 25 credits, taken before it runs and refunded if the server errors.
Not the call you need? Find the right call by what you are trying to answer.

Request

curl -X POST https://mcp.dunkapi.com/v1/get_prop_backtest \
  -H "Content-Type: application/json" \
  -H "x-api-key: YOUR_API_KEY" \
  -d '{
  "market": "player_points",
  "side": "over",
  "home_away": "away",
  "team_role": "underdog",
  "from_date": "2024-10-01"
}'

Response

json

{
  "market": "player_points",
  "label": "points",
  "side": "over",
  "filters": {
    "from_date": "2024-10-01",
    "home_away": "away",
    "team_role": "underdog"
  },
  "results": 12372,
  "hits": 6017,
  "misses": 6355,
  "pushes": 0,
  "hit_rate": 0.486,
  "avg_implied_prob": 0.531,
  "avg_fair_prob": 0.498,
  "no_result": 337,
  "no_price": 0,
  "games": 1651,
  "players": 529,
  "avg_books": 5,
  "avg_line": 12.6,
  "avg_actual": 12.8,
  "hit_rate_minus_implied": -0.045,
  "hit_rate_minus_fair": -0.012,
  "hit_rate_interval_95": [
    0.478,
    0.495
  ],
  "market_inside_interval": false,
  "by_season": [
    {
      "season": 2025,
      "results": 5913,
      "hits": 2895,
      "misses": 3018,
      "pushes": 0,
      "hit_rate": 0.49,
      "avg_implied_prob": 0.531,
      "avg_fair_prob": 0.498
    },
    {
      "season": 2026,
      "results": 5880,
      "hits": 2840,
      "misses": 3040,
      "pushes": 0,
      "hit_rate": 0.483,
      "avg_implied_prob": 0.531,
      "avg_fair_prob": 0.498
    },
    {
      "season": 2027,
      "results": 579,
      "hits": 282,
      "misses": 297,
      "pushes": 0,
      "hit_rate": 0.487,
      "avg_implied_prob": 0.531,
      "avg_fair_prob": 0.499
    }
  ],
  "method": "One result per player per game: the closing line is the median across books, the market's probability comes from the books at that line, and a push counts in neither the hits nor the rate. no_result counts players with no result for the game (not played, or no stat line of ours), no_price those with no price on this side. team_role is by the books' closing moneylines. avg_implied_prob averages each book's implied probability, margin included. fair_prob takes the margin out of each book's two prices for the same line, proportionally, then averages. Only identical lines are combined.",
  "caution": "A past result is not a forecast. Trying many filters will turn up some that beat the market by luck, so check by_season and the interval. Not a recommendation."
}