Part 3
# I Scanned 22 Biotech Stocks This Week. 16 Are Dead. Here’s the Full Body Count.
The Catalyst Scanner is two weeks old. In that time, I’ve run 22 tickers through the framework, killed 16 of them, built positions in 2, and rebuilt the entire scanner from scratch after it missed a trade that would have made me 100%.
This is the post where I show you everything — the updated system (v3), the kills, the misses, and what I learned from getting it wrong.
If you read the launch post, you saw the original framework. That version is already dead. Here’s what replaced it, and why.
---
## What Broke in the Original Scanner
Three things happened in the first two weeks that exposed gaps I didn’t know existed.
**1. A stock I killed made someone else 100%.**
ACHV had clean Phase 3 data — two trials published in JAMA — a $354 million capital raise from top-tier funds, and an upcoming FDA decision date. The scanner killed it because the company pre-announced it expected a rejection due to a manufacturing inspection failure at a third-party factory. Correct call. The FDA decision was dead.
But the stock had crashed to $2.65. Cash per share was roughly $1.80. The institutional investors who just wrote $180 million in checks bought in at $3.64. The stock was sitting below their cost basis with the science completely intact — the drug works, the factory doesn’t, and they were already switching manufacturers.
Six weeks later, the stock hit $5.29. That’s a 100% return the scanner couldn’t see, because it was looking for a binary FDA catalyst that didn’t exist. The *trade* that existed — buy near cash value, sell when it recovers to the institutional entry price — was a completely different setup that the framework had no category for.
I added one.
**2. My highest conviction stock failed my own price filter.**
SPRO — first oral carbapenem antibiotic, backed by GSK, Phase 3 stopped early for efficacy, 38 days from an FDA approval decision — drifted below $3. The original scanner had a hard price floor at $3 to avoid penny stocks. So technically, the best trade on my entire board failed the scanner that was supposed to find it.
The problem was obvious: a $2.50 stock with a $145 million market cap and 400,000 shares of daily volume is not a penny stock. It’s a depressed small-cap with institutional sponsorship. The price filter was catching the wrong thing. What actually matters for avoiding penny stock traps is liquidity — market cap and volume — not the raw share price.
I replaced the price filter.
**3. I almost oversized my best position.**
After spending an hour confirming every angle of the SPRO thesis — the prior rejection was addressed, the SEC investigation was resolved, GSK is funding everything, the drug was stopped early for efficacy — I wanted to put 45% of my portfolio into it. The math was intoxicating: if it doubles on approval, that’s a $15,000 gain on a single trade.
The framework had position sizing guidelines but no hard mechanism to check the impulse. I needed a formal “are you sure?” gate that forces me to stare at the downside before I commit.
I added one of those too.
---
## The Scanner, Rebuilt
Version 3 isn’t a tweak. It’s a ground-up rebuild. Here’s what changed.
**The layer system is gone.** The old framework used numbered layers — Layer 1, Layer 1B, Layer 2, Layer 3 — and by the end of the first week I was spending more time debating which layer a stock belonged in than analyzing the stock itself. The new system uses descriptive tiers: Active, Pipeline, Watch, and Dead. A stock is either something I’m trading, something I’m preparing to trade, something I’m tracking, or something I’ve killed. No ambiguity.
**Active has three sub-tiers.** Binary Catalyst is the core — confirmed FDA decision date, all filters pass, position sized by the TAM × BIC quadrant. Recovery is the new one — stock crashed near cash value on fixable bad news with institutional investors above creating a price target. Override is for stocks that fail one non-critical filter but pass everything else with a decision date within 90 days.
**The price filter is replaced by a liquidity screen.** Market cap must be $20M–$500M. Average daily volume must be 200,000 shares or higher. There’s a soft price ceiling at $20 — above that gets flagged for reduced sizing, but it’s not an automatic kill. The hard $3 floor is gone.
**A manufacturing filter is now part of the primary screen.** If the manufacturer listed in the FDA application has an OAI classification — the most serious outcome of an FDA inspection — the stock is dead. No exceptions. A drug can have perfect clinical data and still get rejected because the factory that makes it failed inspection. I watched it happen in real time. The filter exists now.
**Position sizing has a formal over-sizing check.** Before putting more than 20% of my portfolio into any single name, I have to answer five questions. What’s my max dollar loss if this goes to zero? Can I survive that loss and still trade next month? Am I adding because the thesis improved or because the price dropped? Do I have enough cash for other opportunities? Would I be comfortable explaining this position size to subscribers? If any answer gives me pause, I don’t add.
**Exit rules exist for every quadrant.** The original scanner was all entry, no exit. Now every trade type has defined sell rules: when to take profits on approval, when to cut on a rejection, when to sell into a pre-decision run, and when to trim a losing position for pain management.
**A drawdown relief rule lets me trim without panic selling.** If a position hits -15% from my average cost with no negative news and the catalyst is more than 30 days away, I can sell up to 25% to reduce exposure and free capital. It’s not a stop loss — it’s a pressure valve. The remaining 75% holds for the catalyst.
**There’s a missed trade journal.** Every trade the scanner should have caught but didn’t gets documented: what was the setup, why didn’t the scanner see it, what rule would have caught it, and was the rule added. The feedback loop is the actual edge — not the framework at any single point in time, but the process that makes it better after every mistake.
**Dead tickers have a 30-day cooling period before resurrection.** This prevents me from killing a stock on Monday and resurrecting it on Friday because it went up 15% and I got jealous. Dust settles in 30 days. If the kill reason is genuinely resolved after a month, the stock can be re-evaluated — but it enters Pipeline, never directly to Active.
---
## The Full Body Count
I scanned 22 tickers in the first two weeks. Here’s every single one and what happened to it.
**Active: Binary — 2 survivors**
Two stocks passed every filter and received full deep dives. Both are positioned with real capital. Both have FDA catalysts in the next 38 days. The details are for paid subscribers — but I can tell you one is a Grand Slam quadrant (first-in-class, large market, major pharma partner) and the other is an Orphan Blockbuster (only therapy of its kind for a rare disease, Breakthrough Therapy Designation, $370,000/year pricing potential). Combined, they represent 46% of my portfolio with the remaining 54% in cash.
**Pipeline — 4 names being monitored**
Four stocks are in the preparation zone. One has a December 2026 FDA decision approaching the scan window — it gets a full deep dive next month. One is waiting on Phase 3 data that could drop any week. Two others are waiting on regulatory milestones before they can enter the active tier. No capital deployed in any of them.
**Watch — 5 names being tracked**
Five stocks are too early for action but interesting enough to monitor. One is a potential Grand Slam for 2028 — same CRL recovery pattern as my top active position, but the new trial just started and data is 18 months away. Another has strong Phase 1 data but volume too thin for my framework. The rest are waiting on data readouts or enrollment milestones.
**Dead — 16 kills**
Here’s the graveyard. Every kill has a specific, documented reason.
One died because 80% of its shares were covered by warrants — even an FDA approval wouldn’t move the stock because the dilution would absorb the entire pop. One died because a third-party manufacturer failed an FDA inspection and the company pre-announced it expects a rejection. One died because it had $2.3 million in cash and was weeks from running out of money. One died because Phase 2 data showed only 25% of patients still responding at 12 months — the stock fell 74% in a single day.
Three died because their market caps were above my ceiling. Two died because they were Phase 1 or Phase 2 with no FDA decision anywhere on the horizon. One died because it was a commercial platform with eight marketed products — a fine company, but not a binary catalyst trade. One died because it had already run 500% and was a $6 billion momentum story, not an asymmetric setup.
One was a zombie pharmaceutical company with four employees and a failed Phase 3 history, repositioning itself as an investment holding company. That one died fastest.
The kill rate — 73% — is the scanner working exactly as designed. Out of 22 stocks that looked interesting enough to screen, only 2 earned real capital. The other 20 would have been dead money, losses, or distractions.
---
## The Quadrant That Didn’t Exist
The original framework had four quadrants for scoring a stock’s post-approval potential: Grand Slam, Orphan Blockbuster, Crowded Large, and Small Incremental. After scanning 22 stocks, I found a fifth pattern that didn’t fit any of them.
Some companies are already profitable. They have existing products generating revenue, a commercial sales force, and a real business — and then they also have an FDA binary catalyst sitting on top of that business. The downside is cushioned by the commercial floor. The upside is incremental, not transformational. That’s a different risk profile than a single-asset biotech where approval is everything and rejection is near-terminal.
The new quadrant is called Commercial + Catalyst. It gets moderate position sizing — bigger than a lottery ticket, smaller than a full position. The existing business means you’re not betting on a single binary outcome, but the catalyst still creates an asymmetric opportunity. One stock on my Pipeline list fits this profile, with an FDA decision in December 2026. It gets a deep dive next month when it enters the scan window.
---
## What’s Coming
The next 38 days are the most concentrated catalyst window I’ve scanned. Two FDA decisions in my active positions within days of each other, plus a third decision on a stock I’m watching but not positioned in. The next few posts will cover:
Weekly scans continue every Monday — free for all subscribers. The calendar sweep, new kills, and framework updates.
For paid subscribers: daily portfolio tracking starts next week as we enter the 30-day countdown. Real positions, real P/L, real-time reactions to news. Full pre-decision analysis with scenario planning and exit protocols. And the live coverage during decision week — what I’m doing, when I’m doing it, and why.
The scanner killed 73% of what I looked at. The 27% that survived are about to face the only test that matters: the FDA’s decision. Everything up to this point has been preparation. The next 38 days are execution.
---
## The Scanner Evolves
Version 1 had no manufacturing filter, no exit rules, and a price floor that would have killed my best trade. Version 2 added those, plus a recovery trade category and a catalyst clustering protocol. Version 3 rebuilt the entire architecture — new tier system, liquidity screen, flow analysis, over-sizing checks, drawdown relief, and a missed trade journal.
I publish the full framework for free because the system isn’t the moat. The execution is. Anyone can read these filters. The value is in the 2-3 hours of deep dive research per stock, the daily monitoring, the discipline to hold a losing position when the thesis is intact, and the discipline to kill a stock that looks exciting but fails the filters.
The scanner will keep evolving. Every missed trade, every blown stop, every lesson gets incorporated. That feedback loop — not the framework at any single point in time — is the edge.
If you want to watch it work in real time, the next 38 days are the proving ground.
---
*Not financial advice. I hold positions in stocks discussed and will always disclose them. Biotech trading involves substantial risk of total capital loss. The Catalyst Scanner documents my personal research process — nothing here is a recommendation to buy or sell any security. Do your own due diligence. Past results don’t predict future performance.*