Every player prop, graded against
the actual box score.

What you actually get back

Mariners @ Cardinals, Julio Rodriguez Total Bases prop. He hit a double + a single — 3 total bases. Over 1.5 won; Under 1.5 lost. The shape is the standard bookmakers[].markets[].outcomes[] the-odds-api consumers already parse — with resolution and actual_value appended.

{
  "id": "11049",
  "home_team": "St. Louis Cardinals",
  "away_team": "Seattle Mariners",
  "completed": true,
  "scores": [
    {"name": "St. Louis Cardinals", "score": 4},
    {"name": "Seattle Mariners", "score": 7}
  ],
  "bookmakers": [{
    "key": "draftkings",
    "title": "DraftKings",
    "markets": [{
      "key": "batter_total_bases",
      "outcomes": [
        {
          "name": "Over",
          "description": "Julio Rodriguez",
          "point": 1.5,
          "price": -125,
          "resolution": "won",       // ← graded against real stat
          "actual_value": 3          // ← the actual TB he recorded
        },
        {
          "name": "Under",
          "description": "Julio Rodriguez",
          "point": 1.5,
          "price": 105,
          "resolution": "lost",
          "actual_value": 3
        }
      ]
    }]
  }]
}

The graded view is mirrored across every book that had the line. DraftKings, FanDuel, Pinnacle, Bovada, Underdog, PrizePicks — same stat, same verdict, different prices. That's the data shape you need to compute closing line value and book-by-book hit rate.

Get graded props in 2 lines

Same auth as every other PropLine endpoint — ?apiKey= or X-API-Key header. Free tier returns the market structure with resolution nulled and redacted: true set. Pro tier returns the full graded view.

# pip install propline
from propline import PropLine

client = PropLine(api_key="YOUR_KEY")

results = client.get_event_results(
    sport="baseball_mlb",
    event_id="11049",
)

for bm in results["bookmakers"]:
    for market in bm["markets"]:
        for o in market["outcomes"]:
            if o.get("resolution") == "won":
                print(
                    f'{o["description"]} {market["key"]} '
                    f'{o["name"]} {o["point"]} hit '
                    f'(actual: {o["actual_value"]})'
                )

curl "https://api.prop-line.com/v1/sports/\
baseball_mlb/events/11049/results\
?apiKey=YOUR_KEY"

// npm install propline
import { PropLine } from "propline";

const client = new PropLine({
  apiKey: process.env.PROPLINE_API_KEY!,
});

const results = await client.getEventResults({
  sport: "baseball_mlb",
  eventId: "11049",
});

How it works

Four moving parts. None of them require you to do anything different than a normal odds API call.

1. 1. Scraper records the prop. Every 90 seconds we poll every active sportsbook (Bovada, DraftKings, FanDuel, Pinnacle, BetMGM, BetRivers, Unibet, Underdog, PrizePicks, Kalshi, Polymarket, Matchbook, Smarkets) and write every Over/Under with the line and the price to odds_snapshots.
2. 2. Game ends. We poll the league's stats API every 5 minutes. When the game flips to completed=true, the resolver fetches the box score.
3. 3. Resolver grades every prop. Match player names (with team-prefix fallback for soccer / MLB, where books share roster aliases) against the box score, compare the actual stat to the line, and write resolution + actual_value back to every outcome row.
4. 4. You read it. GET /v1/sports/{sport}/events/{id}/results returns the joined view across every book. Or subscribe to a resolution webhook and get pushed every graded prop within seconds.