> For the complete documentation index, see [llms.txt](https://matterhorn-doc.mometic.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://matterhorn-doc.mometic.com/understand-the-scores/opportunity-shape.md).

# Eight-axis opportunity shape

Interpret Matterhorn ML growth, forward demand, quality, market, valuation, evidence, durability, and emergence axes, including missing measurements.

The radar is a profile of the opportunity. It helps you see whether a candidate is supported by several complementary strengths or dominated by one narrow feature.

![AMD eight-axis opportunity chart, with an explicitly unmeasured valuation axis](https://981865776-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNOJXLd7UJBniObM136Nu%2Fuploads%2Fgit-blob-8b8402eae56e2185ea94e18022534d3ace68bab9%2Fcompany-ml.png?alt=media)

*Interface captured September 9, 2026 (America/Chicago); the frozen ML data snapshot is dated September 5, 2026. AMD is an interface example, not a recommendation.*

| Axis                | What it describes                                                 | Useful follow-up                                                  |
| ------------------- | ----------------------------------------------------------------- | ----------------------------------------------------------------- |
| Growth              | Revenue and profit growth and acceleration                        | Is gross profit improving alongside revenue?                      |
| Forward demand      | Supported backlog, orders, recurring-business and demand signals  | Are definitions and comparison periods consistent?                |
| Quality             | Margins, cash generation, capital returns, leverage, and dilution | How much growth reaches existing shareholders?                    |
| Market confirmation | Price strength, drawdown, and volatility                          | Is market behavior confirming the business case?                  |
| Valuation           | Price relative to supported financial measures                    | What expectations appear embedded in the price?                   |
| Evidence            | Completeness, confidence, and freshness of inputs                 | Which part of the thesis has the least support?                   |
| Durability          | Separate four-quarter economics-persistence estimate              | Does the company's starting economics fit the model's definition? |
| ML emergence        | Learned rank percentile in the eligible pool                      | What historical combination might explain the relative score?     |

## How to read it well

Hover or focus a legend entry for its definition and input coverage. Compare the same axis across companies, then return to underlying values and source periods. A growth score of 76 is not 76% revenue growth, and an emergence score of 93 is not a 93% chance of outperformance.

The axes are not eight equally weighted, independent signals. Some share inputs, and the learned model uses several of the same families summarized by the financial components. A larger filled area is not an additional portfolio-ranking formula.

## A gap is informative

An entirely unsupported family is plotted as unmeasured. The fixed component formula retains a neutral midpoint for unavailable inputs instead of inventing favorable measurements. Partial coverage can therefore produce a middle-of-the-road score supported by relatively few observations.

In the example, **Valuation —** means the frozen ML family is unmeasured. It does not mean AMD is cheap, expensive, or that the live company checklist has no valuation data. The ML snapshot and ordinary research can have different coverage and dates.

> **Pro Tip — Read the smallest supported axis and the biggest gap before admiring the strongest axis.** They often identify the next worthwhile piece of diligence.

Next: [compare two or more companies](/investigate-the-business/compare-companies.md), [evidence quality](/understand-the-scores/evidence-quality.md).