EarningsWatcher
HomeWikiMCP Trading OS
Case study · AI connectors

How a Member Turned EW MCP Into a Personal Trading OS

Not a “ask ChatGPT for picks” story. Lifetime member littlebeard wired Earnings Watcher into his AI workflow, turned the channel into searchable heuristics, and ran a real IV Rush paper trade end to end — research, automation, journal.

EarningsWatcher July 2026 Featuring littlebeard · lifetime member

Most people try a new trading tool like this: open the app, poke around, maybe paper one trade, then forget it.

littlebeard did the opposite — and he didn’t need a new dashboard to do it. He did it in chat.

What using our MCP looks like

No API keys to babysit. No CSV exports. You connect Earnings Watcher once, then you talk to it like a coworker who already knows the platform:

Chat interface connecting to Earnings Watcher MCP with a simple whoami check
Real session: “connect to earnings-watcher mcp please” → logged in → ready to pull data, analyze, or walk a strategy. That’s the whole UX.
As simple as a chat Same window where you already think out loud. Ask for tomorrow’s IV Rush names, a playbook walkthrough, a paper trade, or a journal note — and the connector hits live EW tools, not a generic options FAQ scraped from the internet.

That’s the product surface. The rest of this case study is what littlebeard built on top of it in about a week: a personal trading OS, a YouTube → heuristics pipeline, and a closed NFLX IV Rush paper trade with a written postmortem. All without treating the AI like a guru.

1 prompt
To connect & verify
~1 wk
Connect → working OS
105
Videos → notes / heuristics
+8.6%
EW Live Trades (paper)

2. Building the vault — a personal trading OS

Chat is the front door. What littlebeard did next is the interesting part: he turned MCP answers into a durable knowledge system he could reopen on trade day.

Their map of content isn’t a bookmark list — it’s an architecture:

Map of content with four playbooks and YouTube ingest
Four playbooks, methodology, onboarding, YouTube ingest, studies, and trade logs (littlebeard · lifetime).

What they pulled in first (via MCP learning tools + their own notes):

Why that matters: once the playbooks live in their vault, every later chat starts from shared context. They’re not re-explaining IV Rush every morning. They’re asking, “Is this name analysis-ready today?”

Tools to build your own edge — not a guru’s picks to follow.

That line sat at the top of their vault. It’s also the point of MCP: the connector should make you better at using EW, not replace your judgment.

3. Turning YouTube into a trade-day system

Watching videos is easy. Keeping the lessons is hard. This is where the MCP workflow got interesting.

On trade week they ran a full YouTube channel ingest and converted it into three working docs:

ArtifactWhat it becameWhy it matters
EW Video Library Index of extraction notes across ~105 videos Searchable memory instead of “I think Amin said something about noon exits”
EW Trade Heuristics ~21 KB evidence-tagged rules (their longest reference doc) Decision checklist, not vibes — when a setup is valid / invalid
Next Earnings Play Runbook Step-by-step trade-day checklist The “open this on the morning of” doc

That trio is the difference between content consumption and an operating system:

  1. Library — what was taught
  2. Heuristics — what to do / not do
  3. Runbook — what to do today, in order
Why MCP helped here They weren’t just summarizing YouTube in a vacuum. The AI could cross-check channel lessons against live EW tools and methodology — so “exit before the print” isn’t a slogan, it’s tied to IV Rush Radar signals and crush math.

They also kept process logs — an ideation log (idea → locked design) and a session log for trade day. That’s journaling culture, not screenshot culture.

4. Deep dive: NFLX IV Rush paper trade (2026-07-16)

This is the proof the OS wasn’t theoretical. On Netflix’s AMC print day littlebeard ran the IV Rush playbook as a closed-loop experiment — research in chat, paper fills in more than one place, then a written postmortem.

NFLX IV Rush paper trade write-up showing setup, signals, and exit-before-print logic
His NFLX postmortem: signal, setup, why exit-before-print, mechanical P&L decomposition (broker account details redacted).

The play (in his words)

The setup he logged

FieldWhat he wrote
Spot~\$73.68 (post-split)
Implied move~8.86%
ATM straddle (07-17 K=74)~\$6.60 (call ~3.15 / put ~3.45)
Front ATM IV~146–152%
Liquidity score~96.7
Term structureFront week elevated vs later weeks (~146% → ~75% → ~49%)

Why NFLX — the MCP signal

He didn’t pick NFLX because it was “famous.” He used EW IV Rush signals:

The timing lesson (this is the money) Six days earlier (2026-07-10) the same name’s rush read was not favorable (~−11% in his log). It flipped in the −1 day window. That matches the playbook: early projections are noisy — get serious on the last trading day, often watching into midday, then decide.

Why “exit before print” can still make sense

His note decomposes the straddle into base vol vs earnings-jump premium, then shows the mechanical bleed of exiting flat before the print (~1–2% time-value) versus the crush/theta hit if you hold through (~tens of percent on his table). The edge he’s hunting is the last-day IV demand, not the overnight gap.

Paper trade twice — follow it everywhere

Here’s the MCP detail that makes this feel like a real product, not a blog experiment:

littlebeard didn’t only automate a fill at his broker paper account. He also asked the connector to open the same IV Rush paper trade on Earnings Watcher — so the position showed up in the app’s Live Trades feed, with the same story, while his external paper fill ran in parallel.

His trader note on the platform says it plainly:

IV-rush exit-before-print; mirrors Alpaca paper fill 14:35 ET.

That’s the point of “follow it everywhere”: chat for research and actions, EW Live Trades for the social/platform trail, broker paper for his own automation — one thesis, multiple mirrors, one journal.

Earnings Watcher Live Trades feed showing Littlebeard NFLX IV-RUSH paper trade at +8.60%
On-platform: Littlebeard · NFLX · IV-RUSH · closed the same afternoon · +8.60% on the EW paper ticket (note mirrors the Alpaca fill).
What to ask the MCP After you’ve validated an IV Rush setup in chat, you can say something like: “Paper this ATM straddle on my EW account too, exit-before-print, and journal why.” That’s how the trade becomes visible in Live Trades / journal without leaving the conversation.

How far he took automation

This wasn’t “type a prompt and hope.” He combined:

Closed result (paper only) On Earnings Watcher Live Trades the NFLX IV-RUSH ticket closed around +8.60% (cost \$645 → last \$700.50 on that card). His vault note also tracks the mirrored Alpaca paper (~+6.2% / +\$40 on that side). Same idea, two mirrors — illustrative only, not a live track record, not “MCP prints money.”
What we are not saying We’re not recommending NFLX straddles, IV Rush on every name, or copying his sizing. IV Rush lives on the radar with its own rules. Other EW tools (moves analyzer, simulator likelihood, scanner) belong to other plays — he kept those lanes clean in the vault, which is exactly how we teach it.

5. The four-step MCP loop (what to steal)

StepWhat they didWhat you can do in chat
1. ConnectLinked EW MCP to their AI clientAdd the MCP URL in ChatGPT or Cursor
2. LearnIngested playbooks + 105 videos → heuristics / runbook“Teach me IV Rush step by step” / “Build my trade-day checklist”
3. ResearchUsed IV Rush signals + live chain context on NFLX“Is NFLX analysis-ready on IV Rush Radar?”
4. Act + journalPapered on EW and broker, exited before print, wrote the note“Paper this on my EW account too” → journal

They also ran studies that didn’t “win” as deployable edges (IV Rush / Big Movers / Runners notes marked closed). That’s a feature of a real OS: it can say no.

6. How to try the same setup

If the screenshot above felt familiar — good. That’s the bar. You should be productive in the first message.

  1. Open Earnings Watcher → Settings → AI Connectors (MCP) (or the in-app MCP page).
  2. Copy the MCP URL into ChatGPT (custom connector) or Cursor (Settings → MCP).
  3. Sign in with your EW account.
  4. Sanity check: “connect to earnings-watcher mcp” / “whoami” — you should see your account.
  5. Start learning: “Onboard me as options-familiar” or “Walk the IV Rush playbook.”
  6. On a real name: “Pull IV Rush signals for [ticker] reporting tomorrow — are we analysis-ready?”
  7. If you take a paper trade, ask MCP to place it on EW too so it shows in Live Trades — then force a journal note the same day.

Build your own MCP workflow

Connect Earnings Watcher in ChatGPT or Cursor — learn, research, paper trade, and journal in one place. Start from pricing if you’re not a member yet.

See pricing →