Investment Methodology

How DYOR HQ scores, rates, and evaluates investment opportunities

What is a Conviction Score?

The conviction score is a composite 0-100 signal measuring the strength of the risk-adjusted return case over a 12-month investment horizon. It is not a probability of success, nor a target return — it is the overall quality and completeness of the investment case.

The score synthesises three distinct input dimensions:

The Three Pillars

Fundamentals

  • Revenue and earnings trajectory
  • Competitive position and moat
  • Balance sheet quality
  • Management quality and capital allocation
  • Cash generation and margin structure

Catalysts

  • Proximity of upcoming events
  • Magnitude of potential re-rating
  • Binary vs gradual outcomes
  • Regulatory or clinical milestones
  • Partnership or M&A potential

Sentiment

  • AI-powered news and social media analysis
  • Analyst consensus and price targets
  • Price momentum and positioning
  • Insider activity signals
  • Sector and macro tailwinds

Recommendation Tiers

Each report carries two labels: a conviction score (0-100) and a recommendation tier. The tiers translate the raw score into actionable guidance, accounting for stock type and risk profile.

Score Recommendation Position sizing Description
80 — 100 BUY Full position Exceptional case. Multiple clear catalysts, strong fundamentals, sentiment turning positive. The highest-conviction opportunities.
65 — 79 BUY (strong) Full position Strong case. Fundamentals are sound, at least one meaningful near-term catalyst. Full position sizing appropriate.
50 — 64 OPPORTUNISTIC BUY Half position Valid case with unresolved risks. Catalysts are either distant or binary. Requires active monitoring and a specific trigger to upgrade.
30 — 49 SPECULATIVE BUY Trail-size only A specific binary catalyst is required to justify the position. For clinical-stage biotech, pre-revenue tech, or turnaround situations where the bull case is low-probability but high-impact.
0 — 29 Avoid No position The investment thesis is broken, the risk/reward is genuinely negative, or the company faces an existential challenge without a credible recovery path.

In Practice

Examples from current coverage

CRWD 73 BUY (strong) First GAAP profit, Anthropic partnership, analyst PT $500-520 vs $420. Exceptional execution.
COP 68 BUY (strong) Iran peace deal pulled crude from $131 to $119. Jefferies PT $160. Q1 earnings April 30.
PXEN.L 40 SPECULATIVE BUY H&P broker 21p target vs 3.38p price. Gas price dependent. Binary: San Licence awarded April 2026.
NVDA 65 BUY (strong) Analyst PT $267-275 vs $199. Ising quantum AI extends TAM. 40x forward P/E — valuation is the constraint.
FARN 44 SPECULATIVE BUY Clinical-stage oncology. BEXMAB Phase I/II data read in 2026. Binary outcome. Trail-size only.
ONDO 58 OPPORTUNISTIC BUY $3bn TVL, Franklin Templeton partnership. Fee-switch vote H2 2026 is the key binary catalyst.

How the Score is Calculated

Each report uses a bull/base/bear scenario framework:

  • Bull case (positive outcome): Assigned a score of 80-120 depending on magnitude
  • Base case (central outcome): Assigned a score of 50-70
  • Bear case (adverse outcome): Assigned a score of 10-40

Each scenario is weighted (e.g. 30% bull, 55% base, 15% bear) and the weighted sum produces a raw conviction figure, which is normalised to the 0-100 scale. The final score also incorporates the AI sentiment dimension as a market-read adjustment.

The grok sentiment score (0-100) is derived from AI-powered analysis of news and social media. It is treated as one input among three — not a definitive signal on its own. A very low grok score against strong fundamentals (as with ServiceNow in April 2026) may indicate market fear rather than fundamental deterioration, and the score is adjusted accordingly.

Important Caveats

  • Not financial advice: All research is for informational purposes only. DYOR HQ research does not constitute financial advice. Always conduct your own due diligence.
  • Past performance: Historical conviction scores and recommendations do not predict future outcomes. Markets are forward-looking and conditions change.
  • Binary events: Clinical-stage biotech, early-stage tech, and turnaround situations carry binary risk that no scoring framework fully captures. Position sizing should reflect the true probability-weighted outcome, not just the conviction score.
  • Sentiment limitations: AI-powered sentiment analysis reflects current market mood, not fundamental value. During market dislocations, sentiment can detach from fundamentals for extended periods.
  • Currency: All prices and market cap figures use the native currency of the primary exchange. GBX (pence) stocks are denoted with .L ticker suffix. USD stocks use the plain ticker.

Disclaimer: DYOR HQ research is provided for informational and educational purposes only. It does not constitute financial, investment, legal, or tax advice. All analysis reflects the views of the author at the time of publication and is subject to change without notice. Investment markets involve risk. Past performance is not indicative of future results. Always consult a qualified financial adviser before making investment decisions. All data is sourced from live market feeds and company disclosures; DYOR HQ makes no warranty as to the accuracy or completeness of any data cited.