> 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/reference/columns.md).

# Column dictionary

Look up Matterhorn rank, opportunity, completeness, confidence, emergence, compounding rate, ML evidence quality, and comparison column meanings.

Use this page when the same company appears to carry several different scores. The mode and panel determine the meaning.

## Companies and company details

| Label                  | Standard view                                                   | ML enhanced view                                                   |
| ---------------------- | --------------------------------------------------------------- | ------------------------------------------------------------------ |
| **# / Rank**           | Position in the admitted, adjusted standard ordering            | Position in the frozen beta bundle                                 |
| **Ticker / Name**      | Stored company identity                                         | Same company; beta matching also checks the frozen issuer identity |
| **Sector**             | Industry/checklist bucket derived from SEC classification       | Retains its standard basis                                         |
| **Opp / ML score**     | Standard opportunity value; ranking can use further adjustments | 70% component / 30% learned-pattern blend                          |
| **Compl**              | Weighted standard evidence completeness                         | Retains standard completeness; not ML family coverage              |
| **Conf**               | Confidence of evaluated standard evidence                       | Beta evidence-quality value for a matched prediction               |
| **Components**         | Five standard component bars                                    | Eight ML opportunity-shape axes                                    |
| **Emerg**              | Standard rate/durability/headroom emergence score               | Retains the standard emergence score                               |
| **Rate**               | Standard calculated compounding rate                            | Retains the standard calculated rate                               |
| **Tier / Rankability** | Configured evidence-readiness classification                    | Retains the standard classification                                |

In ML mode, the table labels these fields **ML score**, **ML quality**, **ML shape**, **Std compl**, **Std emerg**, **Std rate**, and **Std tier**. Hover the header help control and check the scoring mode before copying values into a comparison.

## ML company breakdown

| Field                                                     | Meaning                                                                              |
| --------------------------------------------------------- | ------------------------------------------------------------------------------------ |
| ML enhanced score                                         | Blended research score from 0 to 100                                                 |
| Financial components                                      | Aggregate of the fixed financial/evidence component families and durability estimate |
| Learned patterns                                          | Cross-sectional percentile of the frozen emergence model output, scaled to 100       |
| Evidence quality                                          | Input support, not a likelihood of a successful investment                           |
| Data as of                                                | Observation date of the published prediction bundle                                  |
| Growth / demand / quality / market / valuation / evidence | Family summaries; inspect hover coverage                                             |
| Durability                                                | Separate economics-persistence estimate, subject to applicability                    |
| ML emergence                                              | Learned relative rank, not an expected return                                        |

The backend also retains model and feature versions, raw predictions, out-of-distribution diagnostics, feature values, contribution explanations, and source lineage. These are audit outputs; the compact beta UI does not display every field as a separate panel or offer a general CSV export button.

## Common units

**%** is a rate or share; **pp** is a difference between percentage rates or percentile readings; **×** is a multiple; **B/M/T** abbreviate billion/million/trillion. An internal fraction of 0.10 is displayed as 10%. A backlog/revenue reading of 1.5× means one and a half times the specified revenue denominator, not 150% growth.

The [glossary](/reference/glossary.md) defines the financial terms. [Scores and ranks](/understand-the-scores/understanding-scores.md) explains why a component value, composite, and rank need not move together.