The sports betting
data API

curl "https://api.sportwizzard.com/api/v1/events?league=mlb" \
  -H "X-Api-Key: sw_live_your_api_key_here"

{
  "success": true,
  "data": [ /* … */ ],
  "nextCursor": "opaque-cursor-string-or-null",
  "meta": { "count": 50, "updated": "2026-07-09T16:41:27Z" }
}

Cache with ETags

Live-odds responses include an ETag header. Send it back as If-None-Match to get a free 304 when nothing changed. To stay in sync, poll /api/v1/odds/updated?since={timestamp} for the rows that changed.

Events are the spine

Everything keys off an event: one game or match. Odds, box scores, and context all carry an eventId. List events, take an id, then pull that event's board with /v1/events/{id}/odds.

Odds are one flat table

Each row is one price at one book. A row repeats everything you need to interpret it, so you can filter, join, or load rows into a dataframe without walking a tree. The id is a stable composite key ({eventId}:{book}:{subtype}:{selectionId}) that survives price moves. Upsert on it.

{
  "id": "31cd6744-…-c4a7d1fce547:ballybet:SPREAD:019f495d-8558-7324-…",
  "sportsbook": "ballybet",
  "league": "mlb",
  "eventId": "31cd6744-9444-4fbd-b25c-c4a7d1fce547",
  "market": "SPREAD",
  "marketSubtype": "SPREAD",
  "period": "FULL",
  "selection": "KC Royals",
  "side": "AWAY",
  "teamSide": "AWAY",
  "teamName": "KC Royals",
  "line": -1.5,
  "priceAmerican": 200,
  "priceDecimal": 3.00,
  "suspended": false,
  "eventStartTime": "2026-07-10T23:05",
  "updated": "2026-07-10T14:53:35Z"
}

Field glossary

Choose a read pattern

A typical pipeline: snapshot to bootstrap, then /odds/updated to stay in sync, /odds for targeted reads, include_history=true for line movement, and /historical/odds to replay the past.

Price history in the snapshot

Add ?include_history=true to /v1/snapshot and each selection's priceHistory is populated with its movement entries: fs/ls are first-seen and last-seen epoch seconds, pa/pd the American and decimal price, dm the DFS multiplier, and su: true marks a suspended span. Entries append on every price change, up to 50 per selection. The schema is identical either way: without the flag, priceHistory is null, so the same client code reads both. 1 credit without, 5 with. The two variants carry different ETags.

# board only (1 credit)            # board + line movement (5 credits), same schema
curl --compressed "https://api.sportwizzard.com/api/v1/snapshot" \
  -H "X-Api-Key: sw_live_…"        # …/api/v1/snapshot?include_history=true

Scope /v1/odds

Unscoped, /v1/odds spans every league and book. Always scope it, or use /v1/events/{id}/odds, /v1/players/{id}/odds, or /v1/teams/{id}/odds.

Three fields, three questions

A game line only has to say which team, in which game. A future has to answer three independent questions, so it uses three independent fields. Read them together and any row is unambiguous.

The axes are orthogonal: a conference winner is SEASON + FULL + AFC, a division winner is SEASON + REG_SEASON + AFC_EAST, and a season passing-yards total is SEASON + REG_SEASON with no segment at all. Season player props reuse the same PLAYER_TOTAL_* subtypes as game props, so marketScope is the only thing separating a 3,750.5-yard season line from a game line. Check it before you model on it.

Ask for them: ?scope=

/v1/odds takes a scope filter. Matching is exact: scope=season does not include tournament outrights, and vice versa. Omit it and you get every scope, which is the unchanged default. This parameter is on /v1/odds only — the delta feed and the player/team odds endpoints do not accept it today.

# season-long futures only
curl "https://api.sportwizzard.com/api/v1/odds?scope=season&league=nfl&limit=200" \
  -H "X-Api-Key: sw_live_…"

# scope=event  → game markets     scope=tournament → tournament outrights
# omitted      → all three (default, unchanged)

# a typo fails loudly instead of looking like "no futures exist"
curl "https://api.sportwizzard.com/api/v1/odds?scope=futures"
# 400 {"detail":"Invalid scope 'futures'. Use scope=event (game markets),
#       scope=season (season-long futures), scope=tournament (tournament
#       outrights), or omit the parameter for all scopes.",
#      "validValues":["event","season","tournament"]}

Why segment is load-bearing

Every NFL future for a season hangs off one synthetic season event, so all 32 DIVISION_WINNER selections share a single eventId. Group by eventId + marketSubtype alone and they read as one 32-team race whose implied probabilities sum to about 8. Group by segment as well and you get the eight real four-team markets. Note that segment is a response field, not a query filter — pull the subtype and split on it client-side.

{
  "id": "b2e737f7-…-852ce6b8d255:draftkings:DIVISION_WINNER:01a01c2f-a38f-76ec-…",
  "sportsbook": "draftkings",
  "league": "nfl",
  "eventId": "b2e737f7-dfb6-4535-bdc4-852ce6b8d255",
  "market": "CATEGORICAL",
  "marketSubtype": "DIVISION_WINNER",
  "period": "REG_SEASON",
  "selection": "BUF Bills",
  "priceAmerican": -136,
  "priceDecimal": 1.7400,
  "side": "YES",
  "teamSide": "NONE",
  "teamName": "BUF Bills",
  "suspended": false,
  "eventStartTime": "2027-02-16T23:59",
  "updated": "2026-08-20T16:44:17Z",
  "marketScope": "SEASON",
  "segment": "AFC_EAST"
}

Winner and award markets are categorical: one side: "YES" selection per competitor and teamSide: "NONE", because most books never post a NO side. Do not expect a two-sided market here.

One season event, every book on it

Futures from different books converge on the same season anchor, so a single eventId carries every book's price for the same outcome. Cross-book comparison on futures is a group-by, not a matching problem — which is the whole reason this data is worth having, since retail books and exchanges disagree most on the long-dated stuff.

curl "https://api.sportwizzard.com/api/v1/odds?scope=season&league=nfl&market=CATEGORICAL&limit=1000" \
  -H "X-Api-Key: sw_live_…" \
  | jq -r '.data[]
      | select(.marketSubtype=="CHAMPIONSHIP_WINNER" and (.teamName|test("Bills")))
      | [.sportsbook, .priceAmerican, .eventId, .eventStartTime] | @tsv' | sort

ballybet         1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
betonline        1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
betparx          1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
betrivers-kambi  1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
betway           1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
draftkings       1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
fanduel          1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
hardrock         1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
novig             953  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
pinnacle          988  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
polymarket       1011  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
prophetx          960  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
rebet            1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59
thescore         1000  b2e737f7-dfb6-4535-bdc4-852ce6b8d255  2027-02-16T23:59

Fourteen books, one eventId, one anchor date, captured live. The retail books sit on a round +1000 while the exchanges price it themselves. Note the team label differs by book (BUF Bills vs Buffalo Bills) — match on teamId or the canonical /v1/teams catalog, never on the display string.

Snapshot pulls are flat: 1 credit, or 5 with include_history=true. Your usage comes back in the X-Credits-Cost and X-Credits-Remaining response headers. Annual billing gets 2 months free. Full details on the pricing page.