Scores

Wallet scores, trade scores, and market scores measure different things.

Choose a score

Choose the field for the question you want to answer:

Your question	Fields to read
How strong is this wallet’s realized-profit record?	`grade` and `score`.
How much evidence supports its forecasting score?	`forecast_score` and `forecast_evidence`.
How is it performing recently?	`streak_tier`.
How do its performance and copyability compare?	`quant_metrics.smart_score` and `quant_metrics.copy_score`.
Which large trade should I review first?	`review_score`.
Which outcome has more net large-trade exposure?	`sharp_money.net_flow_usd` and `sharp_money.direction`.
Which markets should I browse first?	The explore endpoint’s `sort` parameter.
Which upcoming game has more graded money concentrated on one side?	`directional_rank_score`.

Limits

These fields describe observed activity or calculated rankings. None guarantees a market result.

Wallet scores

Trader can return these fields:

Field	Meaning
`grade`	A letter from `S` to `F`, dominated by realized profit. See Wallet grades.
`score`	The returned wallet ranking value from 0 to 100.
`forecast_score`	An independent forecasting percentile, from 0 to 100.
`forecast_evidence`	The degree of support from the wallet’s observed record, from 0 to 1. Higher values indicate more evidence.
`streak_tier`	A comparison of the wallet’s last 7 days with other graded wallets: `hot`, `rising`, `neutral`, `cooling`, or `cold`.
`rank`	The wallet’s leaderboard position, when available.

Missing values are omitted. streak_tier is absent without recent activity; the forecasting fields are absent when their signal is unavailable. A leaderboard row has no rank field; its place in the returned list gives its position. An A wallet can be cooling, and a B wallet can be hot, because grade and recent form are separate measures.

Large-trade review scores

review_score is a review-priority score from 0 to 1. Read it alongside the trade’s observed size, price, timestamp, and wallet record. review_score is current when you read it. recorded_review_score keeps the value stored when the trade was first recorded; older trades can have null. The deprecated names signal_score and recorded_signal_score remain available with identical values. Prefer the review names in new code.

Market flow

Market flow and Sharp money flows return sharp_money. Their smart_money compatibility field contains the same data.

Field	Meaning
`net_flow_usd`	Net exposure: `BUY YES` and `SELL NO` add dollars; `BUY NO` and `SELL YES` subtract dollars.
`direction`	`NO` for negative unrounded net exposure and `YES` otherwise. Exactly zero uses `YES` as a tie-break.
`token_id`	The CLOB token ID for that outcome, or `null` if unavailable.
`buy_volume_usd`, `sell_volume_usd`	Unsigned gross volume across both outcomes in the selected timeframe.
`whale_trade_count`	The number of large trades included.
`top_positions`	Up to 5 open-position holders with grades `S`, `A`, or `B`, largest first. Available on market flow; each wallet appears once on its larger side.

Market flow includes tracked large trades without a grade filter. Sharp money flows applies min_grade, which defaults to B and excludes ungraded wallets. Both use large trades only. Fills below the large-trade size threshold do not contribute, regardless of wallet grade.

curl -H "Authorization: Bearer $OXINSIDER_API_KEY" \
  "https://api.0xinsider.com/api/v1/markets/sharp-money-flows?timeframe=24h&min_grade=B&limit=10"

Explore scores

Explore markets returns discover_score and score_components. discover_score supplies the ranking for sort=hot. Explore also returns smart_score, smart_count, and smart_label, each of which can be null. Its smart_score is separate from both wallet quant metrics and the pre-game score below.

Pre-game scores

Pre-game sides returns upcoming markets with graded-wallet backing. Use the returned order and measured positions for context.

Field	Meaning
`smart_score`	The signed share of graded-wallet money, from -1 to 1. Positive values favor `YES`.
`conviction_score`	A calculated backing score. Higher values indicate stronger measured backing.
`directional_rank_score`	The returned ranking score for this market. Use the server’s order when presenting these markets.

s_count, a_count, and b_count count wallets on the heavier side; sharp_pct is that side’s share of graded dollars. The endpoint accepts min_grade values S, A, and B; other grade values return 400. For deeper wallet measures, request quant metrics. For how old a source is, request trust metadata.

Getting started

Integrations

Guides

Concepts

Choose a score

Limits

Wallet scores

Large-trade review scores

Market flow

Explore scores

Pre-game scores

​Choose a score

​Limits

​Wallet scores

​Large-trade review scores

​Market flow

​Explore scores

​Pre-game scores

Choose a score

Limits

Wallet scores

Large-trade review scores

Market flow

Explore scores

Pre-game scores