Series — Tenbagger Analysis 2026 (Part 2 of 2) ① Korea’s 27 Tenbaggers of 2023-2026 — the KOSPI + KOSDAQ side ② This post — the US side: 40 names, the AI-infra spine, and the pair-trade with Korea

Data Pipeline — Say It Up Front

• Universe: 5,843 US tickers. Backfilled 2023-01-01 → 2026-04-23 via yfinance with auto_adjust=True (split/dividend-adjusted → matches realized investor returns).
• Coverage after resume: 1,690 tickers with pre-2023-06 history. Post-2024 IPOs and delisted names drop out naturally.
• Filter: first close ≤ 30 days after 2023-04-24, held through 2026-04-23, ≥10× total return.
• Output: 40 tenbaggers confirmed.

The Distribution — 40 Names, 19.3× Mean

• Mean multiple: 19.3× (Korea: 17.4×)
• Median multiple: 15.7× (Korea: 13.8×)
• Max: AAOI at 69.5×
• Median 3-year return: ~1,480% (Korea median ~900%, so US runs ~1.6×)

US doesn’t just have more tenbaggers than Korea — the winners run further. The right tail is fatter.

The Structural Map — 9 Buckets

Three stories jump out.

One. AI infrastructure is a third of the list. Not “AI” in the loose sense — AI infrastructure: optical transceivers, nearline HDDs for training data, high-multilayer PCBs, data-center switchgear and transformers, hyperscale electrical construction. This is picks-and-shovels that paid.

Two. Biotech single-asset theses delivered 8 names — every one starting from $1-$16 with “all-or-nothing” trial binaries. That’s a structural profile the Korean market basically doesn’t produce at this size.

Three. Quantum, insurance turnarounds, and specialty minerals each printed standalone winners that don’t fit the consensus narrative. The 2023 list was not “AI only.”

AI Infrastructure — The Full Stack

Break the 13 AI-infra names down by sub-layer:

This is the exact Korean pair-trade map:

The pattern: US primary leads, Korea follows with a 3-6 month lag. When the US primary overshoots, the Korean derivative peaks. When the US primary corrects, the Korean derivative over-reacts on the way down.

This is not a narrative claim — it’s visible in the multiple dispersion. Isu Petasys peaked after AAOI. HD Hyundai Electric peaked after POWL. The lag is real.

Quality Check — Who Already Overshot

The sell-side target data matters here. Of the 40 names:

Key read: the AI-infrastructure primaries are where the sell-side has already caught up. If you missed it, chasing here is chasing a stock trading above its analyst consensus target. That’s rarely the asymmetric trade.

Where the consensus is still behind: crypto→AI compute pivots, biotech single-assets, quantum beta. These are the 2026-2027 continuation candidates, not the infrastructure primaries.

Factor Decomposition — Entry-Point Profiles

At the April 2023 start point:

Two practical reads:

1. The AI-infra names were already earning money in 2023. They weren’t lottery tickets. The “quality + earnings visibility” screen would have caught them. The biotech and space tenbaggers wouldn’t have passed — those were pure narrative/catalyst trades.
2. Entry-point price matters. Every biotech tenbagger started between $1 and $16. The deep value recoveries started sub-$5. Low nominal price was a real factor in the 10× realization — not because low price is magic, but because it indicates pre-catalyst pricing.

What Korea’s Tenbagger List Does Not Have

Cross-reference Part 1. Two entire archetypes are structurally missing from Korea:

1. Biotech single-asset 10× at scale. US produced 8 names from $1-16 starting points on clinical binaries. Korea has Peptron and HLB at 7× — but nothing that ran $1 → $337 like PRAX. Why? Korean biotech gets re-rated during phase 2 (pre-data) and sells off before phase 3 readout. The US market allows the full ramp through approval. Different market microstructure, different distribution of outcomes.

2. Survivor recovery 10×. CVNA went from $3.55 and bankruptcy rumors to $180+. Korea doesn’t produce this because names under administrative-issue designation typically go straight to delisting rather than restructure. The US Chapter 11 + tradeable credit spread market creates a set of “near-death → survive” names that simply don’t exist on KOSDAQ.

For a Korean investor who wants exposure to these profiles: XBI for biotech breadth, or direct ADR positions in SMMT / DAVE / CVNA for the concentrated bets. Not replicable via domestic names.

The Insurance Turnaround Cluster (Added After Resume Backfill)

When the data pipeline’s backfill completed (1,479 → 1,690 tickers), one new tenbagger showed up: KINS (Kingstone Companies) at 13.4×, a NY-state regional P&C insurer that went from $1.3 → $17.

Combined with ROOT (auto AI insurance, 14.8×), this is a quiet Insurance Turnaround mini-theme. The 2022-23 combined-ratio spike above 110% forced P&C discipline — reinsurance rates hardened, underwriting tightened, and 2024-25 ROE recovered to 20%+.

• ROOT → telematics maturity
• KINS → regional consolidation beneficiary

Korean pair: Meritz Fire, Hanwha General Insurance — similar ROE normalization, but multiples only ran 3-4× (no domestic tenbagger equivalent).

Actionable — Three Framings for a KR-Based PM

A. Pair-trade discipline on the AI-infra stack. When US primaries trade below analyst consensus targets, Korean derivatives are on borrowed time. Current read: AAOI, CRDO, LITE, POWL, IESC all below target → Isu Petasys, HD Hyundai Electric, Korean power-grid complex in take-profit zone. This is the single most actionable insight in the dataset.

B. Next-cycle candidates. Where is sell-side still behind?

• Edge-AI semis — smaller InP/GaN names (Navitas, Power Integrations) below AXTI’s scale
• Quantum infrastructure — RGTI alone printed 38.5×; IONQ / QUBT / QMCO still have multiple headroom
• AI robotics — Symbotic and others not yet tenbaggers, but candidate narratives for 2026-2028
• Crypto→AI pivots with remaining upside — IREN, APLD, WULF

C. Structural gaps. US biotech single-asset and survivor-recovery exposure cannot be replicated through KR-listed names. Either ADR direct (SMMT, PRAX, CVNA, DAVE) or XBI for breadth.

Risk Flags

1. Forward P/E above 100× — PLTR 107×, AAOI 100×, LITE 102×, WULF 685×, ASTS 99×. Consensus downgrade = -30 to -50% risk.
2. Already below analyst target — WDC, STX, LITE, IESC, POWL, AAOI, AXTI, TTMI, BWAY, RKLB. Market ran ahead; trailing stops matter.
3. Sub-$1 starting price — check short interest and SEC filings before extrapolating (RGC has TCM-pump flags).
4. Crypto→AI pivots — dual exposure to Bitcoin and hyperscaler capex. Either breaking down = -50%.

Bottom Line

40 US tenbaggers over 3 years. 33% are AI-infrastructure full stack — optical, storage, PCB, power, electrical construction. Korea’s 27 tenbaggers are the 2nd-order derivative layer of this same trade (HBM, high-layer PCB, transformers). The US infrastructure primaries are now trading through analyst targets — take-profit zone. The relocation is into crypto→AI pivots, biotech single-assets, quantum, and insurance turnarounds, where consensus is still catching up.

The pair-trade map is the single most transferable output: Korean AI-derivative names follow US primaries with a 3-6 month lag, and over-react on both ends.

What’s Next

1. KR-US pair cointegration backtest. AAOI weekly vs Isu Petasys weekly at lag=0/30/60 days — is the relationship tradeable?
2. Reverse-screen Russell 2000 using the April-2023 tenbagger profile (ROE / Fwd P/E / Rev YoY / 2yr return) — find the 2026 candidates before they break out.
3. Biotech ADR event calendar — SMMT, PRAX, SLNO, SYRE upcoming trial readouts.
4. AI capex cycle top check — MSFT / META / AMZN / GOOGL 2026 capex guidance vs 2025 actuals. The primary signal for when the US infra names will finally correct.

This closes the Tenbagger Analysis 2026 series. Part 1: Korea’s 27 tenbaggers and why power grid quietly beat AI.

All names listed are idea-generation candidates, not recommendations. Nothing here is investment advice.