Fair odds, closing line value and line moves

Prices with the bookmaker margin removed, the closing line and how your price compared with it, and the lines that moved across bookmakers.

Updated

These routes need Pro or above. Other keys get 403 Plan restriction ("This endpoint requires Pro access."). A fair price is a model of the market, not a promise: check the price at the bookmaker before acting.

The routes

RouteWhat it returns
GET /v1/fair-odds?league=Fair prices for a league's events, market by market.
GET /v1/events/{event_id}/fair-oddsFair prices for one event.
GET /v1/events/{event_id}/closingEach bookmaker's opening and closing price, the fair closing price and closing line value, and optionally your own bet's.
GET /v1/moves?league=Markets whose line moved at one or more bookmakers in a recent window, with steam flagged.

Fair odds

A fair price is a market's price with the bookmaker margin taken out. For each market we build it from one of two bases, and fair.source says which:

  • sharp_book: Pinnacle's own market with its margin removed, when Pinnacle prices every outcome of it. fair.bookmaker is pinnacle and fair.books is 1.
  • consensus: otherwise, the median of every bookmaker that prices the whole market, each with its own margin removed, when at least three do. fair.bookmaker is null and fair.books says how many bookmakers it was built from.

A market with neither base is left out, as are markets whose runners are players or scores (first goal scorer, correct score and the like). fair.overround is the implied probabilities of the base added up: the sharp bookmaker's own, or the median across the consensus bookmakers. 1.03 is a 3% margin.

devig chooses how the margin is removed:

  • proportional (the default) divides each implied probability by their sum. It is the same de-vig /v1/opportunities/ev uses, so the two routes agree.
  • power raises each implied probability to the one exponent that makes them sum to 1, which takes more margin off long prices. Ask for it per call with devig=power; long shots on exchanges and prediction markets often look less generous under it.

Each outcome carries its fair_probability and fair_decimal_odds, and best_price: the best current price among the bookmakers your key may read, with ev_pct, its edge against the fair price in percent (negative when the best price is shorter than fair). best_price is null when no bookmaker has a current price on the outcome.

Two rules keep best_price honest:

  • A lone price far over fair gives way. A price more than 25% above the fair price is chosen only when a second bookmaker also prices the outcome more than 25% above it. Otherwise the next best price is shown. One stale or mislabelled price cannot make a +100% edge on its own.
  • Once an event has started, only in-play prices count. Its fair price and best_price are built only from prices observed since the event's start, so a pre-match price left over from before kick-off never sits beside in-play prices.

GET /v1/fair-odds needs league and takes markets, bookmakers, alternates, devig, starts_after and starts_before. With no start window it lists events that started at most 12 hours ago and upcoming ones, in start order, 25 a page at most. Main lines only unless alternates=true. GET /v1/events/{event_id}/fair-odds takes the same filters for one event; there, bookmakers limits both the bookmakers a consensus is built from and the bookmakers best_price is chosen among. Each call costs 5 credits; an empty answer costs nothing.

Example response for GET /v1/events/ev_7k2m9qx4hd3c/fair-odds?markets=winner (sample prices, not real odds; the teams are invented):

{
  "data": {
    "id": "ev_7k2m9qx4hd3c",
    "name": "Harbour City Comets v Riverside Owls",
    "league": { "key": "nba", "name": "NBA" },
    "scheduled_start": "2026-11-19T00:30:00.000Z",
    "status": "scheduled",
    "participants": [
      { "role": "home", "identity": { "key": "harbour_city_comets", "name": "Harbour City Comets" } },
      { "role": "away", "identity": { "key": "riverside_owls", "name": "Riverside Owls" } }
    ],
    "markets": [
      {
        "id": "mk_8d2k5r9w3h7c4n6x",
        "key": "winner",
        "name": "Winner",
        "period": "full",
        "subject": null,
        "alternate": false,
        "fair": { "source": "sharp_book", "bookmaker": "pinnacle", "devig": "proportional", "books": 1, "overround": 1.03159 },
        "outcomes": [
          {
            "id": "oc_5m2x8k4r7c9w3h6t",
            "name": "Harbour City Comets",
            "side": "home",
            "line": null,
            "fair_probability": 0.5875,
            "fair_decimal_odds": 1.702,
            "best_price": { "bookmaker": "sportsbet", "decimal_odds": 1.7, "ev_pct": -0.125, "observed_at": "2026-11-18T03:12:10.000Z" }
          },
          {
            "id": "oc_7r3c9h5k2w8m4x6d",
            "name": "Riverside Owls",
            "side": "away",
            "line": null,
            "fair_probability": 0.4125,
            "fair_decimal_odds": 2.424,
            "best_price": { "bookmaker": "tab", "decimal_odds": 2.45, "ev_pct": 1.063, "observed_at": "2026-11-18T03:11:58.000Z" }
          }
        ]
      }
    ]
  },
  "meta": { "count": 1, "cursor": null },
  "links": { "next": null }
}

Here Pinnacle's 1.65 and 2.35 carry a 3.16% margin. Taking it out gives fair prices of 1.702 and 2.424, so TAB's 2.45 on the away side is about 1.06% over fair, and Sportsbet's 1.70 on the home side is just under it.

Closing line and closing line value

GET /v1/events/{event_id}/closing returns, for each market and outcome:

  • prices: each bookmaker's opening and closing price (as snapshots=opening,closing on the odds routes returns them; see History and snapshots), and opening_clv_pct: the closing line value of that bookmaker's opening price against the fair closing price, in percent. Positive beat the close.
  • closing_fair_probability and closing_fair_decimal_odds: the fair closing price, built from the closing prices the same way /fair-odds builds a fair price. closing_fair on the market says how (source, bookmaker, devig, books, overround).

Give outcome_id and odds together to get the closing line value of your own bet in bet: clv_pct (your odds against the fair closing price, in percent) and beat_close. Without markets the main markets are read (winner, result, handicap and the game totals); name up to 20 market keys for others. The route also takes bookmakers, alternates and devig. Each call costs 10 credits.

Closing prices, closing_fair and the closing line value are null until the event has started. closing_fair is also null when no bookmaker closed the whole market.

Example response for GET /v1/events/ev_7k2m9qx4hd3c/closing?markets=winner&bookmakers=pinnacle,sportsbet&outcome_id=oc_5m2x8k4r7c9w3h6t&odds=1.8, after the game started (sample prices, not real odds; the teams are invented):

{
  "data": {
    "id": "ev_7k2m9qx4hd3c",
    "name": "Harbour City Comets v Riverside Owls",
    "league": { "key": "nba", "name": "NBA" },
    "scheduled_start": "2026-11-19T00:30:00.000Z",
    "status": "live",
    "participants": [
      { "role": "home", "identity": { "key": "harbour_city_comets", "name": "Harbour City Comets" } },
      { "role": "away", "identity": { "key": "riverside_owls", "name": "Riverside Owls" } }
    ],
    "markets": [
      {
        "id": "mk_8d2k5r9w3h7c4n6x",
        "key": "winner",
        "name": "Winner",
        "period": "full",
        "subject": null,
        "alternate": false,
        "closing_fair": { "source": "sharp_book", "bookmaker": "pinnacle", "devig": "proportional", "books": 1, "overround": 1.03395 },
        "outcomes": [
          {
            "id": "oc_5m2x8k4r7c9w3h6t",
            "name": "Harbour City Comets",
            "side": "home",
            "line": null,
            "closing_fair_probability": 0.597015,
            "closing_fair_decimal_odds": 1.675,
            "prices": [
              {
                "bookmaker": "pinnacle",
                "opening": { "decimal_odds": 1.75, "implied_probability": 0.571429, "observed_at": "2026-11-16T22:04:00.000Z" },
                "closing": { "decimal_odds": 1.62, "implied_probability": 0.617284, "observed_at": "2026-11-19T00:29:41.000Z" },
                "opening_clv_pct": 4.478
              },
              {
                "bookmaker": "sportsbet",
                "opening": { "decimal_odds": 1.78, "implied_probability": 0.561798, "observed_at": "2026-11-17T01:15:00.000Z" },
                "closing": { "decimal_odds": 1.6, "implied_probability": 0.625, "observed_at": "2026-11-19T00:29:30.000Z" },
                "opening_clv_pct": 6.269
              }
            ]
          },
          {
            "id": "oc_7r3c9h5k2w8m4x6d",
            "name": "Riverside Owls",
            "side": "away",
            "line": null,
            "closing_fair_probability": 0.402985,
            "closing_fair_decimal_odds": 2.481,
            "prices": [
              {
                "bookmaker": "pinnacle",
                "opening": { "decimal_odds": 2.15, "implied_probability": 0.465116, "observed_at": "2026-11-16T22:04:00.000Z" },
                "closing": { "decimal_odds": 2.4, "implied_probability": 0.416667, "observed_at": "2026-11-19T00:29:41.000Z" },
                "opening_clv_pct": -13.358
              },
              {
                "bookmaker": "sportsbet",
                "opening": { "decimal_odds": 2.1, "implied_probability": 0.47619, "observed_at": "2026-11-17T01:15:00.000Z" },
                "closing": { "decimal_odds": 2.38, "implied_probability": 0.420168, "observed_at": "2026-11-19T00:29:30.000Z" },
                "opening_clv_pct": -15.373
              }
            ]
          }
        ]
      }
    ],
    "bet": {
      "outcome_id": "oc_5m2x8k4r7c9w3h6t",
      "decimal_odds": 1.8,
      "closing_fair_decimal_odds": 1.675,
      "clv_pct": 7.463,
      "beat_close": true
    }
  },
  "meta": { "count": 1, "cursor": null },
  "links": { "next": null }
}

A bet on the home side at 1.80 beat the fair closing price of 1.675 by about 7.5%. Closing line value says how good the price was, not whether the bet won: graded outcomes are in Results.

Line moves and steam

GET /v1/moves?league=nba lists markets whose main line changed at one or more bookmakers in the last window minutes, upcoming events only. Lined markets only: handicaps, totals, and player and team over/unders. Head to head markets have no line, so they have no line moves.

  • window is 5 to 360 minutes in steps of 5 (default 60).
  • Each bookmaker counts once, by its net move in the window: a line that moved and came back is not a move. moves lists each bookmaker that moved, with its from_line, to_line and observed_at.
  • direction is up, down or mixed: which way most bookmakers moved. books_moved, books_up and books_down count them.
  • steam is true when at least min_books bookmakers (default 3) moved the same way and they are at least two thirds of the bookmakers that moved. steam=true lists only those.
  • sharp is true when Pinnacle is among the bookmakers that moved with the majority.
  • from_line and to_line are the median line of the majority before and after. For a handicap the line is the home side's (competitor 1 in tennis).

The route also takes markets, bookmakers and the usual paging. Steam comes first, then the markets the most bookmakers moved. Lines are sampled every few minutes, so a move is timed to the sample that first saw it. Each call costs 10 credits; an empty answer costs nothing.

One entry from GET /v1/moves?league=nba&window=60&steam=true (sample lines, not real odds; the teams are invented):

{
  "id": "mv_5e0b7d2c9a4f1863",
  "event": {
    "id": "ev_7k2m9qx4hd3c",
    "name": "Harbour City Comets v Riverside Owls",
    "league": "nba",
    "scheduled_start": "2026-11-19T00:30:00.000Z"
  },
  "market": { "key": "handicap", "name": "Handicap", "period": "full", "subject": null },
  "direction": "down",
  "steam": true,
  "sharp": true,
  "books_moved": 4,
  "books_up": 0,
  "books_down": 4,
  "from_line": -4.5,
  "to_line": -6.5,
  "first_move_at": "2026-11-18T02:41:00.000Z",
  "last_move_at": "2026-11-18T02:56:00.000Z",
  "moves": [
    { "bookmaker": "pinnacle", "from_line": -4.5, "to_line": -6.5, "observed_at": "2026-11-18T02:41:00.000Z" },
    { "bookmaker": "sportsbet", "from_line": -4.5, "to_line": -6.5, "observed_at": "2026-11-18T02:46:00.000Z" },
    { "bookmaker": "tab", "from_line": -4.5, "to_line": -6, "observed_at": "2026-11-18T02:51:00.000Z" },
    { "bookmaker": "neds", "from_line": -5, "to_line": -6.5, "observed_at": "2026-11-18T02:56:00.000Z" }
  ]
}

Four bookmakers moved the home side's line down inside 15 minutes, Pinnacle first: the median line went from -4.5 to -6.5.

Limits

  • Fair odds cover markets with fixed sides (home and away, over and under and the like). Markets whose runners are players or scores have no fair odds.
  • Closing prices and closing line value are null until the event has started.
  • A best price more than 25% over fair needs a second bookmaker that far over it too, and a started event's fair and best prices use only in-play prices.
  • Head to head markets have no line moves, and moves are only as fine as the few-minute samples.
  • A fair price is an estimate built from bookmakers' own prices, and prices move. Check the price at the bookmaker before acting.
  • Fair odds cost 5 credits a call; closing line and line moves cost 10. An empty answer costs nothing. See Plans and limits.

For ready-made arbitrage, middles and +EV lists, see Arbitrage, middles and +EV. For every recorded price change, see History and snapshots. Every route and field is in the API reference.

Need help?FAQ · Email support

Search across the SockOdds documentation and catalogue.