PetrusOne

How the screen works

PetrusOne pulls an issuer's most recent annual figures straight from the SEC's XBRL company-facts API — the same structured data behind every 10-K — then scores three pillars a lender cares about and blends them into a single letter grade.

The three pillars

• Leverage (45%) — Debt / EBITDA. How many years of cash earnings the debt represents.
• Coverage (45%) — EBITDA / Interest. Whether earnings comfortably clear the interest bill.
• Liquidity (10%) — Current ratio. A modifier, not a driver: current assets include inventory, which is not liquidity, and healthy investment-grade issuers routinely run current ratios below 1.0.

Scoring

Each pillar is placed into one of seven bands scored 0 (strongest) through 6 (weakest). The pillars are weight-blended, and the result maps back to a band from AAA to CCC. A blended score below 3.5 falls on the investment-grade side of the scale. When an issuer doesn't report a usable tag for a pillar, that pillar is dropped and the remaining weights are renormalised — the sheet says so when this happens.

One override applies: an issuer covering interest less than three times is capped at BB regardless of how strong the rest of the profile looks. Coverage is the binding constraint on whether debt actually gets serviced.

Period integrity

Every figure is anchored to a single fiscal year, fixed by the most recent annual balance sheet date. Issuers retire and replace XBRL tags over time, so taking the latest available value for each metric independently can silently pair one year's earnings with another year's interest expense. Figures outside the anchor period are discarded rather than substituted — a missing number is safer than a number from the wrong year. EBITDA is built from operating income where reported, and rebuilt from pretax or net income where it isn't, since some issuers present no operating income subtotal.

Known limits

• No business risk. This is the largest gap. Rating agencies weight industry cyclicality, competitive position, and management record heavily alongside the financials. A screen built purely on ratios will read an airline or a cruise operator as stronger than the market does, because the numbers look fine right up until a demand shock arrives.
• EBITDA carries no adjustments or add-backs, so it won't tie to a company's own guidance.
• Operating leases, pensions, and off-balance-sheet obligations are not capitalised into debt.
• Issuers with captive finance arms show consolidated debt against industrial earnings, which overstates leverage.
• Banks, insurers, and REITs don't fit a corporate leverage framework and will screen oddly.
• The screen reads the last annual period only — it's a snapshot, not a trend.
• Thresholds are calibrated against observed issuers and are a point of view, not a standard.