The bar — composite fair-value estimate

The headline number on the Sick of Gurus card blends three independent valuation methods into one estimated intrinsic value: Graham's bond-yield-corrected margin-of-safety formula, a 2-stage discounted cash flow (DCF), and a Lynch PEG=1 growth cross-check. A confidence badge shows how closely the three agree — wide disagreement (e.g. one distorted-EPS ticker) is flagged rather than silently averaged away. Price-gate and expensive-zone lines are simply that blended value adjusted by your margin-of-safety reading (see "Price-gate and expensive-zone lines" below). This card also shows the Sick of Gurus Score and Margin of Safety for the same ticker — see those two entries below for what each includes. The product is the bar: both gates, every time. Not a second product name.

Price-gate and expensive-zone lines

Price-gate and expensive-zone lines are set automatically , not user-adjustable — a default +30% vs estimated value (price / discount gate) and a −15% line (expensive zone), applied consistently across every ticker so the bar means the same thing every time. The colored bar shows where the ticker sits right now (the pointer) against those two lines. This is an educational illustration of the desk's estimate, not a recommendation to buy or sell.

Sick of Gurus Score (0–100)

One number, deliberately — not three numbers you have to reconcile yourself. It blends quality signals (moat, management, growth, profitability, balance-sheet strength) with value signals (margin of safety, valuation multiples, cash-flow-based intrinsic value) into a single 0–100 read: roughly the combination of philosophies behind Buffett, Munger, Graham, Fisher, and a dozen others who've actually put up real, verifiable track records. Extra signals folded in when data is available: DCF margin of safety, Magic Formula cheapness (Earnings Yield + Acquirer's Multiple), Accruals Ratio, Shareholder Yield, and the Accounting Manipulation Filter (see below — up to −20 points).

Graduated, not pass/fail (17 Aug 2026): earlier versions gave each quant threshold — P/E ≤20, P/B ≤1.5, EV/EBITDA ≤12, and so on — a hard 100-or-0 score, which meant a genuinely wonderful business trading at a premium (a Google, Microsoft, Visa-type quality compounder) could get dragged to a mediocre score by 2-3 classic-value misses even with excellent fundamentals everywhere else. Every quant check now gives full credit at the threshold and tapers smoothly rather than falling off a cliff — modestly rich still costs some points, but only wildly overpriced costs all of them. Bands and weighting (Quality 80% / Value 20%) are unchanged; only the cliff was smoothed out.

Momentum is shown elsewhere on the site for context only — it's a price signal (has the stock been going up or down), not a value signal (is the business good or cheap), and never feeds the score.

Bands: 90–100 Exceptional · 75–89 Quality core · 60–74 Contender · 40–59 Speculative · 0–39 Avoid. The screen-result banner is separate: score at least 70 and price at least 30% cheaper than estimated value — both, or it does not pass. Educational — not a recommendation to buy or sell.

Business style: a second read, next to the band — what kind of company this is, not whether it Passes. A name can be Exceptional and a quality cyclical, or Quality core and a core quality compounder. The nine styles are Core quality compounder, Quality compounder, Quality growth, Quality cyclical, Recovery, Turnaround value, Turnaround, Special situation, and Speculative. Filter Watchlist, Top Picks, and Alerts by style. Educational — not a recommendation to buy or sell.

Coverage screen (3 Sep 2026): three factors — Forward P/E ≤ 20, net margin ≥ 15% (banks / insurers / REIT omit), ROIC ≥ 15% (banks use ROE). PEG left the gates that day and remains Lynch display only (the PEG=1 IV leg above). A name with Fwd P/E 22 does not sit on the bar, even with a wide margin of safety.

The two gates — intrinsic value & Margin of Safety

Intrinsic value blends three independent methods, averaged: Graham's bond-yield-corrected formula (growth = min of 5-year analyst consensus, 5-year EPS CAGR, and 12% — cyclicals 4%, two guidance cuts 3% — never a single-quarter YoY print; the classic, most conservative read), a 2-stage DCF (explicit FCF years 1-5, fading to a GDP-proxy terminal rate), and a Lynch PEG=1 cross-check (fair P/E ≈ growth rate, the least conservative of the three). Published IV is capped at 1.25× the median Street target; if the desk number sits under 60% of that target it is flagged, not lifted. Names with Forward P/E above 40 or net margin under 5% show “Cannot be valued reliably” instead of a precise figure.

Filings in CNY, DKK, KRW and other non-USD books are converted at a live FX rate into the quote currency before the DCF and cash-based ratios (Cash per Share, P/C, P/FCF, Earnings Yield, Acquirer's Multiple, Shareholder Yield) run. Graham and PEG still use OVERVIEW EPS, which Alpha Vantage already reports per share in the quote currency. If a rate cannot be fetched, those cash-flow fields stay blank rather than mixing currencies — they are not skipped just because the books are in yuan or kroner.

Margin of Safety is how far the current price sits below (or above) that blended intrinsic value. The price-gate and expensive-zone lines translate a Margin of Safety threshold into a dollar illustration of the desk's estimate — not a trade instruction.

Confidence badge: High when the methods land within 15% of each other · Medium when they differ but stay inside 60% · Low — methods disagree (more than 60% apart); Margin of Safety and the Estimate marker use the midpoint; the gradient paints the interval as a shaded band, not a second headline number · Limited data when only one method resolved · Cannot be valued reliably when Forward P/E is above 40 or net margin is under 5%.

52-week range position (shown when available): where the current price sits within its own ~12-month trading range — a lightweight, honestly-labeled proxy for "valuation vs. own history," not a substitute for a true historical P/E band (which needs point-in-time EPS data this site's free data tier doesn't provide).

When the Margin of Safety reading is unreliable: a Leverage-Distortion Gate (added 22 Aug 2026) checks D/E >1.5×, Net Debt/EBITDA >4×, or a trailing P/E under 8× paired with a P/B over 8× — the signature of a P/E propped up by debt-funded buybacks rather than genuine value — and, separately, a detected cyclical earnings peak. When any of these trip, the Margin of Safety reading is treated as neutral (50/100) instead of counted as a discount or premium, and a flag explains why — the Sick of Gurus Score itself isn't penalized, only the price read is held back from being trusted at face value.

Watchlist, bulk rows, portfolio analysis, and daily alerts score every ticker without a human setting the six qualitative sliders (Moat, Management, Circle of competence, Predictability, Revenue quality & growth durability, AI-proof moat) — those default to neutral, excluded from the Quality Composite. The single-ticker card works the same way now (fixed 25 Aug 2026): an untouched slider (still at 5/10) is excluded too, not counted as a real reading — a note above the sliders shows which state you're in. The moment you actually move any of the six, it's included as your real assessment for that ticker.

Guru Take & Guru Watch

Pick a guru from the avatars up top and their own published-checklist logic runs against the same ticker data — different philosophy, different checklist, same numbers. The Guru Watch table runs all 34 gurus' logic at once and ranks their labels BUY → WATCH → SELL. Those labels are an educational illustration of that manager’s published checklist. They are not the desk's recommendation to buy or sell, and not personalized advice.

Guru Score is a directional 0–100 read, not the same rigor as the Sick of Gurus Score: a base of 80/55/25 by checklist tier, nudged ±15 by how far the shared Margin of Safety reading sits above or below that guru's own typical bar (used internally to shape the checklist label only — no price shown).

Guru Take deliberately shows no dollar price target — only that manager’s BUY/SELL/WATCH checklist label and at least three specific comments on the ticker you searched, in that investor's own documented style. The Sick of Gurus card above is the single source for current price, intrinsic value, and the price-gate / expensive-zone bar.

Behavioral & risk flags

Burry (inventory/receivables outpacing sales), Klarman (high leverage), Dalio (portfolio concentration), Spier (debt-funded operations — CFO not covering opex/capex), short interest (≥5% of float or ≥5 days to cover), and net insider selling all apply a real penalty to the Sick of Gurus Score when checked. Camilo's event-driven catalyst flag and Spier's separate checklist-complete flag are shown for context only and never scored.

Accounting Manipulation Filter

A ten-point, pre-DCF sanity checklist, added 17 Aug 2026: Cash Conversion (FCF/Net Income persistently <0.8×), CFO vs Net Income (net income rising while operating cash flow falls), Accruals ((Net Income−CFO)/Assets persistently high — the same Sloan-style signal already feeding the score via the Accruals Ratio field, shown here as its own explicit check too), Receivables vs Sales (A/R growing faster than revenue), Inventory vs Sales (inventory piling up without matching sales), SBC (stock-based comp/revenue rising alongside a growing diluted share count), GAAP vs Non-GAAP (a large, persistent gap between reported and adjusted earnings), real CapEx (maintenance CapEx running below D&A — FCF may be inflated by underinvestment), Acquisitions/Goodwill (growth that's consistently bought via M&A rather than organic), and Share count (diluted shares rising over 5-10 years — EPS growth may be a buyback/dilution artifact, not real earnings power).

The idea, in Buffett's own words: reported accounting figures are the starting point for economic thinking, not a substitute for it. A DCF built on manipulated or low-quality FCF is precise but wrong — this filter exists to catch that before the valuation math even starts.

Now scored: each box checked costs 2 points off the Sick of Gurus Score, capped at −20 total for all ten. It's applied separately from the Burry (inventory/receivables) and Klarman (leverage) penalties below — those are philosophy-level red flags on the business itself, this filter is the underlying data-quality check on whether the numbers those philosophies rely on can be trusted in the first place, so nothing here double-counts the other flags.

Auto-detection (17 Aug 2026): 8 of the 10 boxes now auto-check themselves from the same income statement/balance sheet/cash flow data Sick of Gurus already fetches for every ticker — Cash Conversion, CFO vs Net Income, Accruals, Receivables vs Sales, Inventory vs Sales (when the business reports inventory at all), real CapEx, Acquisitions/Goodwill, and Share Count. Each auto-checked box is tagged "(auto-detected — verify)" — treat it as a fast first read, not a substitute for actually looking at the filing if something looks off. SBC and GAAP-vs-non-GAAP can't be honestly auto-detected from free data (the provider only returns GAAP figures, and doesn't break out stock-based comp cleanly) — those two stay manual and are tagged accordingly.