Where price and value disagree.

3413.TW · Taiwan · NT$35.14B mcap

Terminal perpetuity is 77% of value with no balance-sheet floor; the +178% headline rests on a pinned β and intact 16% margins — contingent, not owned. Fair value.

3217.TWO · Taiwan · NT$17.17B mcap

Looks cheap, but the upside hinges on a 25% terminal margin already rolling over; at a realistic 22% the gap nearly closes. Fair value, not a margin-of-safety bet.

011760.KS · South Korea · ₩303.3B mcap

General trading company (종합상사) — not the carmaker. The FCFF screams cheap (₩52.9k vs a stock at 0.43× book), but it's a trapped-NAV ceiling: the family pays out ~10% and the cash a minority actually receives is worth only ~₩9–10k, below market. It earns its cost of capital with a live Value-Up catalyst, so it's a watchlist, not a value trap — needs a filed payout target ≥30% or a buyback-and-cancel to unlock.

6781.TW · Taiwan · NT$102.5B mcap

AI-data-center BBU engine (~70% of revenue, AWS-dominant) stapled to a cyclical e-bike packer. Cheap at the BASE 2.31% TWD risk-free (+47%) but only fair at the normalized 4.30% bookend and −43% in the compounded bear — the cushion is mostly a low-rate + sustained-execution bet. ~40% ROIC and net cash, but Simplo-controlled with a fresh non-compete waiver, and gross margin already fell during the +59% boom. Fair value, sized for risk.

6757.TW · Taiwan · NT$24.91B mcap

Taiwan's only LCC, a well-run net-cash monopoly — but the intrinsic is a peak-earnings, suppressed-rate ceiling. 65.5%-controlled by China Airlines: the ~50% dividend a minority actually receives is worth ≈ market gross, ~NT$42 net of Taiwan's 21% withholding. Not a value trap — fairly valued. Watchlist; alert ~NT$42–43.

META · USA · $1.54T mcap

Reels works; the AI capex bill is real. Fair value, not a bargain.

NKE · USA · $66.15B mcap

Brand is real; turnaround is slower than the bears need. Mostly priced — not a screaming bargain.

AKZA.AS · Netherlands · €8.39B mcap

World's #3 paints maker mid-merger of equals with #2 Axalta — $25B combined EV, $600M run-rate synergies. Bear-case standalone (raw-mat shock, pricing-lag) still clears at +18%. Deal-close case +32%. Activist Cevian's 10% stake endorses the strategy.

TYRES.HE · Finland · €1.42B mcap

Post-Russia rebuild story. Romania greenfield ramping; market has sat at €10 for 3 years while the inflection started Q2 2025. Modest upside after correcting an understated beta — too thin a margin to justify a starter position.

053800.KQ · South Korea · ₩549B mcap

Korea's #1 cybersecurity name, but ~45% of equity value is net cash a politician-founder hoards. The market already pays near the full NAV ceiling while only the ~2.2% dividend is a hard minority return — the gap is trapped cash, not safety. Watchlist until a treasury cancellation or Value-Up plan routes the cash out.

5283.TW · Taiwan · NT$4.68B mcap

Cheap at book (P/B 0.95, P/E ~9) but a Tsai-family-controlled (~58.5%) cash trap: the ~90% cash-dividend cut routes retained earnings into insurer subordinated bonds, so the FCFF intrinsic is a ceiling a minority can't reach — only a ~0.8% dividend (~0.6% net of 21% Taiwan withholding) actually arrives. Net-cash, below-book, and the YAMADA value brand cap the downside, so watchlist, not trap. Needs a restored dividend (≥NT$2/sh) or a buyback-and-cancel.

VIRP.PA · France · €3.01B mcap

French animal-health specialist with structural growth. DCF lands roughly at the tape — fair value, not a margin-of-safety bet.

5511.TWO · Taiwan · NT$8.12B mcap

Hard ~NT$50 backlog-plus-cash floor, but future public-tender work is ~zero-NPV (competitive bidding drives ROIC→WACC), so fair value is backlog-at-par ~NT$70–90. Roughly fair; dividend cut 24% and order intake decelerating. Watch.

069620.KS · Korea · ₩1.50T mcap

The clean DCF (~₩192k, ≈ sell-side targets) is real, but a 25% governance haircut for founding-family tunneling and ~30% holdco treasury entrenchment leaves only single-digit upside. Fair value.

6691.TW · Taiwan · NT$78.13B mcap

Taiwan's dominant (>80% share) advanced-packaging cleanroom + MEP EPC contractor, at a guided 2026 revenue peak. The +9% base cushion is thin and rides the TWD rate regime — cheap at the local-bond rate, −20% if rate suppression unwinds. A rate-sensitive hold, not a robust buy.

DEC.PA · France · €4.10B mcap

JCDecaux now sits near fair value after β triangulation flipped the SELL. Holding signal: BUY case requires either ad-cycle reacceleration or further DCF tightening.

2536.TW · Taiwan · NT$6.69B mcap

The "PE 6" cheapness is a peak-EPS mirage; normalized, a sub-WACC levered developer worth ~its risk-adjusted NAV. Roughly fair.

5519.TW · Taiwan · NT$7.08B mcap

Chen-family Kaohsiung residential developer. The screener's +48% is a peak-margin × suppressed-β × no-net-debt artifact: normalize the 24.6% FY25 operating margin to a through-cycle 15%, re-lever the −0.08 suppressed beta to 1.46, and subtract ~NT$3.75bn net debt, and the FCFF lands at NT$32.81 — fair value, +1.6%. Unlike net-cash Te Chang there is no balance-sheet floor; it is −40% in the stress regime and fails the robustness gate. The real draw is a genuine, 7-years-growing ~10% dividend — an income play, not a margin of safety. Watchlist; re-arm only if the through-cycle margin holds above 15%.

5434.TW · Taiwan · NT$87.14B mcap

High-quality AI-materials distributor — ROIC ~20%, net cash, 29 straight years of dividends. At the methodology-blended 2.65% TWD risk-free it is fair value (+1%); the eye-catching +34% only appears at Taiwan's suppressed 1.13% bond rate, and normalising toward 4.3% lands −26%. Monte Carlo is a 47% coin flip. Quality you own on a pullback, not at fair value.

3515.TW · Taiwan · NT$32.43B mcap

Sum-of-the-parts: ~36% of value is the 46%-owned, separately-listed ASRock Rack AI-server stake marked to market; the wholly-owned board/VGA stub is flat-to-declining. After the rally the market pays full parts — no free Rack. Fair value.

6414.TW · Taiwan · NT$52.5B mcap

Foxconn-controlled industrial-PC roll-up that's really a holding vehicle for two separately-listed stakes — Kontron (28%) and Marketech (38%). Consolidated FCFF flatters it; valued by its parts, the look-through NAV (~NT$347) already ≈ the price, and below it after a governance haircut. A growing, covered ~4% dividend underpins the downside, but there's no margin of safety after a +25% YTD run to the analyst target. Fair value — a buy nearer the high-NT$200s.

2546.TW · Taiwan · NT$11.0B mcap

Net-cash TSMC-fab contractor; FY2025 EPS is a non-repeatable tech-fab recognition peak (+51% revenue off a ~NT$14bn plateau). The FCFF DCF (~NT$169) is a NAV ceiling on trapped, Kindom-controlled cash a minority can't sweep; the dividend that actually reaches you discounts to ~NT$80 ≈ market. Fairly valued — a 20-year growing ~5% yielder, not a double. Watch.

BLBD · United States · $2.39B mcap

A genuinely high-quality, net-cash #1 school-bus OEM at 35–42% ROIC — but the price already embeds a year-10 operating margin near the FY25 all-time peak it has never sustained. Re-centred to a disciplined 10%, the DCF lands just below market; the whole thesis is a single bet on margin durability (bear $35 / bull $123). Fair value, no margin of safety — revisit on a pullback.

6186.TWO · Taiwan · NT$7.92B mcap

Screens at 5× PE / 12.5% yield — but that's a recognition peak (two big Taipei projects completing at once). On the through-cycle completion pipeline the DCF NAV-ceiling is ~NT$53, yet the cash that actually reaches a minority (~NT$30 net / NT$38 gross) ≈ the price. Add a circular dividend-plus-13%-raise and the worst Taiwan housing cycle since 2016 → fairly valued, no margin of safety. Watch.

010690.KS · Korea · ₩362.42B mcap

Captive 95–96% Hyundai/Kia chassis supplier, cash-flow-negative through the Georgia/Metaplant build-out, with returns on capital barely above the cost of it. At a 5.0% through-cycle margin the valuation essentially re-derives book — the ~4–5× earnings multiple is largely a correct price for trapped, leveraged equity, not a mispricing. Watchlist: the bull/bear band is very wide (₩6.7k–₩23k) on Metaplant operating leverage plus battery-case and robot-module optionality, but a growing-yet-thin dividend and no catalyst with teeth keep it at fair value.

2501.TW · Taiwan · NT$27.3B mcap

Controlled Cathay/Lin Yuan-group developer at a 4-year earnings peak. The screener's +138% is a peak-earnings × suppressed-TWD-rate × no-reinvestment-drag artifact: normalized FCFF (terminal ROC ≈ WACC, land-heavy capital intensity) values the cash a minority can capture at only ~NT$10. The asset case is real but trapped — NAV ~NT$26 book plus a disclosed NT$10.8bn investment-property surplus management won't sell — so book is a ceiling, and the ~5% dividend isn't FCFE-covered. Fairly valued, no margin of safety; watchlist. Re-arm ~NT$14 + a catalyst.

3010.TW · Taiwan · NT$36.06B mcap

A thin-margin distributor that earns roughly its cost of capital. It looks cheap only at Taiwan's suppressed 1.13% bond yield; at the honest 2.65% blended risk-free it sits below market with no balance-sheet floor — a bond-yield artifact, not a margin of safety.

3023.TW · Taiwan · NT$66.50B mcap

A genuine ~20%-ROE, net-cash compounder — but only cheap at Taiwan's suppressed 1.13% bond rate. At the honest 2.65% blended risk-free the DCF lands ~12% below the post-rally price, with ~95% of the Monte Carlo distribution under market. Fairly-to-richly valued; buy zone nearer NT$210–220.