Resources

Users

User quant profiles describe how a Fomo author's past calls actually played out — lead time, outcome distribution, consistency and false positives — computed point in time (research-v1.0). Follower counts are never an input.

How a call is scored (outcome-v1.0)

Entry — the first market price at or after the thesis was posted, within 15 minutes; otherwise the call has no_market_data and is excluded.
Window — 72 hours. Returns are recorded at 1h, 6h, 24h and 3d, plus the maximum return, maximum drawdown and time to peak inside the window.
Major move — the price touches +25% within the window. Lead time = minutes from the thesis to that first touch.
False positive — no major move and a negative 3-day return.
Resolution — a call resolves 72h after it was posted. Only resolved calls count, and only from the moment they resolved.

Metrics

Metric	Definition
`hit_rate`	Share of resolved calls with a major move, with a 95% Wilson interval (`hit_rate_ci`).
`median_lead_time_min`	Median minutes from thesis to the first +25% touch, over calls that reached it.
`false_positive_rate`	Share of resolved calls that were false positives.
`median_drawdown`	Median of the maximum drawdown after each call.
`median_return_24h`	Median 24h return after each call.
`consistency`	100 − scaled dispersion of weekly hit rates (weeks with ≥ 2 calls, at least 2 weeks).
`outcome_volatility`	Standard deviation of 24h returns after calls.
`outcome_distribution`	The full distribution of 24h returns: histogram, box, percentiles, positive share.

Research score

An optional 0–100 composite: hit rate 35 pts, lead time 20, low false positives 20, consistency 15 and median 24h return 10. It is shrunk toward 50 with a prior of 20 calls, so a short lucky streak cannot produce an extreme score. Unknown components count as neutral.

Each component is linear between a “no credit” and a “full credit” anchor: hit rate 18% → 0, 60% → full; false positives 80% → 0, 30% → full; median lead time 36h → 0, 5h → full; median 24h return -10% → 0, +10% → full (0% earns half).

Resolved calls	Confidence
< 5	`INSUFFICIENT` — no research score (null)
5–14	`LOW`
15–49	`MEDIUM`
≥ 50	`HIGH`

Sector strength

Per narrative (AI Agents, Gaming, DePIN, …), computed only when the author has at least 8 resolved calls in that narrative, and shrunk with a prior of 10 calls. With fewer samples the sector is omitted rather than guessed — e.g. @rhea may show AI 81 · Gaming 72 and nothing for DeFi.

Point in time

Metrics are snapshotted (as_of) and history is never rewritten; the profile's history shows how the research score evolved.
Early Quality and backtests read author metrics as of the moment being evaluated — never today's.
Handles can change; profiles are keyed by a stable id (fp_…) and old handles still resolve.

Authors with strong records feed algorithmic cohorts and the Early Quality score.

Endpoints

Example responses use illustrative values ($XYZ, example_rhea, a placeholder address) — not real tokens or authors; every field name and shape is exactly what the API returns. Long sections are collapsed as { … }.

GET/v1/quant/users/{handle}

scope users:read

Point-in-time author metrics for a Fomo handle.

Hit rate with a Wilson interval, median lead time, false-positive rate, drawdown, consistency, outcome volatility, sector strength (only with ≥ 8 samples per sector) and a research score with confidence. Never follower-based.

Parameters

Name	Description
`handle`required path · string	Fomo handle with or without `@`.

handlepath · stringrequired
Fomo handle with or without @.

curl "https://fomoquant.app/v1/quant/users/example_rhea" \
  -H "Authorization: Bearer $FOMOQUANT_API_KEY"

FomoQuant provides analytics and historical/statistical context, not financial advice or guaranteed predictions.