> 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/investigate-the-business/themes-and-scenarios.md).

# Themes and scenarios

Use Matterhorn economic exposures and scenario tools to investigate concentration, theme overlap, and sensitivity without treating simulations as forecasts.

**Themes** organizes companies by economic demand drivers such as grid construction, AI capital expenditure, healthcare utilization, or defense budgets. This can reveal a shared exposure that sector labels miss.

## Choose the question you want to explore

The five tabs in **Themes** keep the exposure mapping and its scenario views together:

| Tab                     | Use it to                                                              |
| ----------------------- | ---------------------------------------------------------------------- |
| **Exposures & targets** | Inspect mapped companies and set the shared allocation budget          |
| **Stress scenarios**    | Examine the modeled portfolio and market under a particular shock      |
| **Drawdown comparison** | Compare the severity of modeled losses across scenarios                |
| **Across scenarios**    | Review the range of outcomes rather than relying on one favorable case |
| **Holdings**            | Inspect the companies and weights used in the modeled basket           |

These views share the same allocation assumptions. Switching tabs does not create a new portfolio.

![The five Themes tabs above the economic exposure explanation](https://981865776-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNOJXLd7UJBniObM136Nu%2Fuploads%2Fgit-blob-ceb9d55773c300efd6a2721638c208b3b6a423f9%2Fthemes-workspace.png?alt=media)

## Read the mapping before the allocation

Start with **Economic Theme Exposures**. Inspect the companies carrying a theme and its share of mapped exposure. The mapping is substantially driven by industry and SIC rules. It is not a measured revenue breakdown for every company, and some themes contain the same or heavily overlapping issuers.

Owning companies labeled “data center power” and “grid construction” may therefore provide less diversification than the two labels suggest. Open the constituent information and overlap warnings before drawing a conclusion.

## Find companies within a theme

Each compact row keeps the theme name, exposure, and ticker sample together. Hover the theme name or ticker preview, focus the preview with the keyboard, or tap **View scores** to open up to **25 highest-scored companies** belonging to that theme. Select a company to open its research. Press **Escape** to dismiss the list.

![Economic theme company preview showing the highest company opportunity scores](https://981865776-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNOJXLd7UJBniObM136Nu%2Fuploads%2Fgit-blob-b4ea2adccee428a4f1a8ec7aa5b56148d013aa78%2Ftheme-companies.png?alt=media)

In standard mode, this list sorts by the displayed company **opportunity score**. Standard **Top 10** uses the adjusted ordering described in [rankings and filters](/find-opportunities/rankings-and-filters.md), so the two lists can differ. With **ML enhanced** enabled, the theme preview sorts by current ML scores. Companies without a matching ML prediction are omitted; the preview does not substitute their standard scores. The preview draws from full theme membership, not just the few ticker names shown in the row.

## Explore an allocation deliberately

Hover a theme row or tab to its target to reveal the allocation field; **Edit targets** keeps the fields open. They share a 100% budget, shown in the allocation gauge. **Save targets** sits beside **Clear all** on the right of the budget strip and persists the target configuration. **Clear all** empties the draft; saving that cleared draft removes the saved targets. In the scenario construction, unallocated weight goes to the strongest standard-ranked candidates rather than automatically remaining cash.

A target percentage is your intended share of the modeled portfolio budget for that theme. Editing it immediately updates the allocated/unallocated budget, the remaining room for other targets, and the gap against the observed mix. The gap is **target minus mix**: a 20% target against a 31.2% mix is **−11.2 percentage points**, meaning your target is below the observed exposure share. Editing the target does not change that observed mix, which describes mapped exposure across the scored universe rather than your brokerage holdings.

The other Themes tabs use your current draft too: **Stress scenarios**, **Drawdown comparison**, **Across scenarios**, and **Holdings** recalculate from those targets when viewed, even before you save. **Save targets** persists the configuration; it is not required to preview a draft. Changes can alter the modeled basket, company weights, and scenario results, but do not change the main **Top 10**, **Companies** rankings, or stored ML scores.

Target allocation is not a score boost or tilt-strength percentage. A 20% target assigns a share of the modeled portfolio budget to that theme; it does not increase its companies’ scores by 20%.

These are modeled allocations. The application does not place brokerage orders or establish that the selected basket matches your actual holdings. Saved targets belong to your account; saving does not change another member’s targets.

## Read the stress model's assumptions

Open **Stress scenarios** and use its **?** explanation. Read how company weights, measured sensitivities, historical episodes, and forward assumptions produce the result. A simulation of today's basket through a past shock is not proof that those companies would have been selected or even investable then.

Historical analogies help frame risk. Forward shocks are scenarios, not forecasts. The model's confidence ranges do not encompass every possible business or market event.

### Read balance colors against the starting amount

A modeled ending balance is green when it is above the starting capital, red when it is below, and neutral at break-even. That color answers whether capital grew or shrank. Compare the separate market balance to judge outperformance: a green portfolio result can still trail the market.

![Stress scenario with a red ending balance below starting capital, despite finishing above the market balance](https://981865776-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNOJXLd7UJBniObM136Nu%2Fuploads%2Fgit-blob-7dd28995ed356a45b51c7407045cc7cc29352eb7%2Fstress-scenario.png?alt=media)

*This modeled example ends at $936,785 from $1,000,000, versus $889,021 for the market. It loses capital while outperforming the market. This is a stress simulation of the current constructed basket, not a historical selection backtest.*

![Across-scenario summary with a red worst-case balance and a green typical ending balance](https://981865776-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNOJXLd7UJBniObM136Nu%2Fuploads%2Fgit-blob-b36aea64335c5cf45b1bd717cc8b9c9b0e8324a1%2Fscenario-balances.png?alt=media)

A beta coefficient or scenario probability is shown neutrally. Its size alone is not a favorable or unfavorable investment result.

The current ML toggle does **not** switch theme allocation or stress calculations to the beta rank. Use the standard basis when interpreting those outputs.

> **Pro Tip — Diversify the reason for being right.** Three different tickers can depend on the same spending cycle. Ask which exposure would hurt them together before counting names.

Related: [comparison](/investigate-the-business/compare-companies.md), [model and backtest interpretation](/reference/methodology.md).