Worst 1% Basket: 10 Key Oddsmaker Signals
Signal | Why It Matters | Extreme Examples |
|---|---|---|
1. Overall Score | Master signal: valuation + quality + balance sheet + timing | MRAM -660, POET -647, WOLF -505, NVTS -461 |
2. Percentile | All top 10 rank in bottom ~0.0–0.3% of universe | MRAM/POET/WOLF at 0.0% |
3. Super Multiple Predictor | Model flags multiple-compression risk | MRAM -795, POET -787, NVTS -601 |
4. EV/Sales Forward | Bubble valuation vs future revenue | POET 135.7x, NVTS 99.3x, RKLB 72.4x |
5. EV/EBITDA Forward | No earnings support / negative EBITDA | RKLB -1,581x, NVTS -117x, FCEL -28x |
6. FCF/EV Forward | Weak or negative cash yield | WOLF -11.1%, FCEL -10.1%, NVTS -0.9% |
7. FCF Margin Forward | Revenue does not convert to cash | NVTS -111%, FCEL -57%, WOLF -53% |
8. ROE / ROIC | Poor capital productivity | AGL ROE -110%, WOLF ROE -84%, SLNH ROE -82% |
9. Stock-Based Comp % Revenue | Dilution hidden in operating model | POET 568%, NVTS 80%, SLNH 36% |
10. Price vs 200D SMA / Timing Raw | Momentum extended vs fundamentals | RXT 4.9x 200D, AGL 3.9x, WOLF 3.1x |
CIO Read
The story is clear:
The model is not just saying these stocks are “expensive.” It is saying they combine extreme valuation, weak cash economics, poor capital returns, high dilution, and stretched momentum.
That is the exact profile that tends to break when liquidity, sentiment, or earnings expectations turn.
NVDA Contrast
NVDA is different. Its model data shows a quality monster with valuation risk:
NVDA Signal | Model Data |
|---|---|
Overall Score | +0.90 |
Percentile | 17.6% |
Growth Grade | A |
Quality Grade | A+ |
Balance Sheet | A- |
ROE | 101% |
ROIC | 92% |
Forward FCF Margin | 85% |
Forward EV/EBITDA | 21.3x |
Forward EV/Sales | 14.6x |
NVDA reports fiscal Q1 2027 earnings on May 20, 2026, at 2 p.m. PT / 5 p.m. ET. NVIDIA’s own investor site confirms the date and time.
Bottom line:
Worst 1% names = bad economics + inflated narratives.
NVDA = elite economics + high expectations.