Smart Money Buzz: Insider Buying Plus Reddit and HN Chatter

For each ticker, this pipeline joins two independent signals: insiders buying their own stock on the open market (SEC Form 4) and the crowd starting to talk about it on Reddit and Hacker News. Either alone is weak. Insider buying is informed but quiet; social buzz is loud but often uninformed. The edge is the overlap — smart money accumulating right as retail attention turns.

Every data source is plain HTTP or a public JSON API (SEC EDGAR, Reddit RSS, HN Algolia), with no browser, no proxy, and no API keys, so runs are fast and cheap.

What it does

1. Insider buying. Pulls Form 4 open-market buys and sells (transaction codes P and S) for your tickers over the lookback window.
2. Social chatter. Searches Reddit and Hacker News for each ticker by cashtag ($TICKER) and, when you supply it, by company name. A keyword filter keeps matches precise so generic-word tickers do not pull noise.
3. Join and score. Per ticker, combines the insider buying, the mention counts, and how fresh both signals are into a 0 to 100 convergence score, then tiers each ticker.

Scoring

The convergence score (0 to 100) is the sum of:

• Insider buying (up to 45). Open-market buy dollar value and the number of distinct buys, with a bonus when buyers outnumber sellers and net buying is positive.
• Buzz (up to 40). Combined Reddit plus Hacker News mention count, weighted up by Hacker News points and comment volume.
• Velocity (up to 15). How recent the freshest signal is. Both legs being recent is the most tradable setup.

Tiers:

• converging — insiders buying on the open market AND social chatter above the threshold, at the same time. The premium signal.
• insider_only — insider buying with little or no social chatter.
• buzz_only — active Reddit/HN chatter with no insider buying.
• watch — weak on both legs.

Output

One row per ticker:

{
  "ticker": "GME",
  "company": "GameStop",
  "tier": "converging",
  "convergenceScore": 78,
  "scoreBreakdown": { "insider": 38, "buzz": 25, "velocity": 15 },
  "insider": {
    "lookbackDays": 30,
    "buyCount": 2,
    "sellCount": 0,
    "buyValueUsd": 1400000,
    "netValueUsd": 1400000,
    "netBuying": true,
    "lastBuyDate": "2026-06-05",
    "topBuys": [{ "insider": "Ryan Cohen", "role": "Chairman", "shares": 100000, "price": 14.0, "valueUsd": 1400000, "date": "2026-06-05" }]
  },
  "buzz": {
    "windowHours": 168,
    "mentions": 11,
    "redditCount": 8,
    "hnCount": 3,
    "hnPoints": 64,
    "hnComments": 22,
    "lastBuzzDate": "2026-06-07",
    "topPosts": [{ "source": "reddit", "title": "GME insider just bought 100k shares", "url": "https://reddit.com/...", "points": null, "date": "2026-06-07" }]
  },
  "scoredAt": "2026-06-07T10:00:00.000Z"
}

Input

• tickers — the companies to scan (uses the $TICKER cashtag for social search).
• companyNames — optional map of ticker to company name, e.g. {"GME": "GameStop"}, to catch posts that never use the cashtag.
• insiderLookbackDays — how far back to pull Form 4 buys/sells (default 30).
• buzzMaxAgeHours — how recent a post must be to count as buzz (default 168 = 7 days).
• minMentions — combined posts needed to count as having buzz (default 2).
• includeInsider, includeBuzz, maxPostsPerSource, minScore — toggles and caps.

Pricing and combined cost

This actor charges per scored ticker: watch $0.03, single-signal $0.07 (insider_only or buzz_only), converging $0.12. The first converging ticker per run is free so you can validate output.

This is a pipeline: it runs three child actors, and each child also bills you for its own per-item usage (sec-form4-insider-tracker per transaction, reddit-lead-monitor per post, hn-lead-monitor per item). Your total for a run is the tier charges above plus those child charges. Because all three children are plain HTTP / public-API calls, the per-run compute is small.

Notes

• Open-market buys (code P) are the conviction signal. Grants, tax withholding, and option exercises are deliberately excluded because they are not discretionary buys.
• Reddit posts pulled via RSS carry no vote or comment counts, so Reddit contributes mention count only; Hacker News supplies the engagement weight.
• Supplying companyNames materially improves buzz coverage for tickers people rarely write as a cashtag.