Operational Structural Mispricing Risk

How OSMR turns operating evidence into a structural map.

OSMR compares the market value assigned to a company with the depth and consistency of the operating evidence beneath it. The result is a transparent classification of current structural conditions—not a price target or prediction.

2.0.0-rc1Methodology version
28Fiscal quarters
7Operating anchors
5 × 5Mainland topology

Most valuation work asks whether a company looks cheap or expensive. OSMR asks a prior question: what has the business actually demonstrated strongly enough to support that valuation?

The framework then asks whether seven different operating lenses tell a coherent story. A company belongs on the mainland map only when both dimensions can be measured responsibly.

01

What has been demonstrated?

The cascade finds the deepest positive operating anchor supported by a complete seven-year record.

02

How stretched is valuation?

Enterprise value is compared with that cumulative operating anchor and ranked across the eligible universe.

03

Do the anchors agree?

Seven valuation ranks are compared. Wide dispersion signals that the company looks very different through different operating lenses.

01 · Launch universe

Eligibility is determined before any score is calculated.

The launch system is intentionally bounded. Records that cannot be compared consistently are retained as exclusions rather than forced into the ranked population.

Acquisition universeU.S.-exchange listings

Test issues, funds, rights, warrants, units, debt and other non-common instruments are removed.

Company scopeU.S.-domiciled operating companies

Common stocks only. ADRs, foreign issuers and unresolved domicile records are outside the launch universe.

Currency ruleUSD reporting

Non-USD and unresolved statement-currency mismatches fail closed.

Tradability floors$50M market cap · $5M ADV

Average daily dollar volume is measured over the trailing 20 trading days.

Fail-closed rule

Missing market capitalization, liquidity, domicile, currency, or required enterprise-value inputs never receive a convenient assumption. The record remains visible with an explicit reason.

02 · Enterprise value

The numerator is constructed consistently.

Vendor enterprise value is retained for quality assurance only. It is not the production numerator.

Constructed enterprise value

Market cap+Total debt+Preferred stock+Minority interestCash & equivalents

Required · never imputed

Market cap, total debt, cash

If any required component is missing, enterprise value is invalid and the company cannot enter the mainland scoring population.

Zero-imputable

Preferred stock, minority interest

Missing values are treated as zero because data vendors commonly omit these line items when the economic value is zero.

A valid enterprise value of zero or less is not treated as a data error. It enters the Negative-EV special cohort.

03 · Operating anchors

Seven measures of operating reality.

Each anchor is the cumulative sum of 28 complete, point-in-time-eligible fiscal quarters. Partial windows are not used.

01

Revenue

Can the company sell?

Cascade + disagreement
02

Gross Profit

Do sales leave economic room after direct costs?

Cascade + disagreement
03

EBITDA

What do operating earnings look like before capital consumption?

Disagreement only
04

EBIT

What remains after charging for long-lived assets?

Cascade + disagreement
05

Net Income

What remains after financing costs and taxes?

Disagreement only
06

Operating Cash Flow

Is accounting performance becoming cash?

Disagreement only
07

Free Cash Flow

Is cash left after reinvestment?

Cascade + disagreement
EBITDA is derived, not borrowed: EBITDA = EBIT + depreciation and amortization.

04 · Axis one

Cascade valuation: how much valuation rests on how much proof?

The company is assigned to the first complete cascade rung with a positive 28-quarter cumulative value.

01

Free Cash Flow

Deepest production rung and the penalty baseline.

0.000log penalty
02

EBIT

Positive cumulative operating income after depreciation.

+0.282log penalty
03

Gross Profit

Sales exceed direct costs, but deeper operating proof is absent.

+1.339log penalty
04

Revenue

The shallowest eligible operating proof in the cascade.

+2.340log penalty

Cascade score

logEnterprise valueCumulative assigned anchor+ empirical penalty

The shallower the company’s demonstrated operating foundation, the larger the empirical adjustment required to compare it with companies resting on deeper anchors.

Lower valuation stretchHigher valuation stretch
Very Low
Low
Moderate
High
Very High

Scores are percentile-ranked across cascade-eligible companies on the scoring date and divided into five cross-sectional buckets.

05 · Axis two

Anchor disagreement: does the operating evidence tell one story?

Every valid anchor produces its own valuation percentile. OSMR measures how widely those percentiles are dispersed within the company.

RevenueGross ProfitEBITDAEBITNet IncomeOperating Cash FlowFree Cash Flow
Sample standard deviationof valid anchor percentile ranksMinimum: 4 valid anchors · ddof = 1
Tightly Agreeing
Mildly Agreeing
Moderate
Mildly Disagreeing
Strongly Disagreeing

Option B uses the 20th, 40th, 60th and 80th percentiles of valid disagreement scores as the current cross-sectional boundaries. Higher disagreement means valuation looks increasingly different depending on which operating anchor is used.

06 · Mainland topology

Two coordinates become a distance—not a blended score.

The reference region is Moderate Cascade × Tightly Agreeing. Every mainland company is classified by its Manhattan distance from that cell.

Distance| Cascade position − 3 | + ( Disagreement position − 1 )
Anchor disagreement

Cascade valuation

Distance 0–1

Tier 1 — Structurally Attractive

Near the reference region: moderate valuation and tight operating agreement.

Distance 2–4

Tier 2 — Mid-Range

One or both structural dimensions are meaningfully displaced.

Distance 5–6

Tier 3 — Structurally Fragile

Far from the reference region because valuation, disagreement, or both are extreme.

There is no continuous composite in production.

OSMR v2 does not average the two axes, smooth the map, or add discretionary weights. “Structurally fragile” describes distance from the reference region; it is not an automatic sell signal or a general judgment about business quality.

07 · Exceptions

Companies that cannot be placed honestly remain off the mainland.

An exception is information. OSMR keeps analytically different conditions visible rather than forcing every company into the same geometry.

Invalid EV

One or more required enterprise-value inputs are missing or fail data-quality checks. The record is not treated as an economic signal.

Negative-EV

Constructed enterprise value is valid and less than or equal to zero. This is a distinct balance-sheet condition, not an ordinary map coordinate.

Incomplete History

No cascade anchor has the complete 28-quarter history required for a defensible classification.

Cascade Truly-Non-Viable

At least one cascade anchor has complete history, but none has a positive cumulative value over the trailing window.

Cascade Eligible No-Disagreement

A cascade score can be calculated, but fewer than four valid anchor ranks remain, so disagreement cannot be measured responsibly.

Launch-universe exclusions

Non-U.S. domicileUnknown domicileNon-USD or unresolved currency mismatchMarket capitalization below $50 million20-day average dollar volume below $5 millionMissing market capitalization or liquidity

These are eligibility or data-quality outcomes—not OSMR economic classifications.

08 · Research and production integrity

The methodology is locked before the final evidence is regenerated.

The production system separates methodology, calibration, live scoring, historical validation and public presentation.

01

Point-in-time statements

Historical financials may enter the model only after their public filing or acceptance date—not at fiscal period end.

02

One scoring engine

Live and historical classifications must use the same pure OSMR v2 scoring implementation.

03

Frozen calibration

Empirical cascade penalties are versioned and checksummed. Weekly production runs do not recalibrate them.

04

Serialization only

Website exporters receive already-scored records. They may not recreate or alter methodology logic.

Evidence status

Historical research is published for the current v2 architecture.

The Evidence page presents payoff, drawdown, topology and robustness results generated with the current cascade × disagreement × distance system. Legacy return and Fama–French findings from earlier architectures are not used as evidence for v2.

OSMR does

  • Describe current structural conditions
  • Compare valuation with demonstrated operating output
  • Expose disagreement, exceptions and exclusions
  • Create a repeatable cross-sectional research map

OSMR does not

  • Predict the next price move
  • Produce a price target
  • Replace company-specific fundamental research
  • Provide personalized investment advice

From specification to application

Inspect the live classifications—or examine the historical evidence.

The platform shows how the methodology classifies individual companies today. The Evidence page presents the historical payoff, drawdown, topology and robustness results supporting the framework.