Invesaro publishes investment theses and settles them against the market in public: with a date, a basket of companies and a verdict.

Investment thesis

The Humanoid Race: Regardless of Who Wins, Precision Gear and Actuator Suppliers Win

LivePublished July 14, 2026150-day horizontechnology

Watch this thesis

In short

What is happening
Mass humanoid robot production starts (Tesla, Figure, Chinese OEMs)
Why it matters
Mass humanoid production plans for 2026 collide with tight precision reducer supply, driving orders to a handful of Japanese and Chinese suppliers.
Who benefits
Harmonic Drive Systems Inc., Nabtesco Corporation, THK Co., Ltd., Leader Harmonious Drive Systems Co., Ltd.
What must hold
Assumptions are statements that must stay true for the thesis to make sense.
Assumptions: 5. See the assumptions
When is the verdict
The horizon is the time we give a thesis. That many days after publication we settle it: we compare the basket of companies with the market benchmark and give a verdict.
December 11, 2026, 150 days after publication
Our score at publication
A score from 0 to 100 given on the day of publication, before the market has verified anything. On the results page we check in public whether higher-scored theses do better.
50 out of 100

Basket return

-18.4%

equal weight, since publication

The market over the same window

+0.5%

benchmark for this basket

Edge over the market

-18.9%

in percentage points

Companies in the basket

6324Harmonic Drive Systems Inc.6268Nabtesco Corporation 6481THK Co., Ltd.688017Leader Harmonious Drive Systems Co., Ltd.

The argument

The year 2026 marks the inflection point where mass humanoid robot manufacturing transitions from demo reels to physical assembly lines. Tesla is commissioning Optimus production lines, Figure is scaling industrial deployments following a $40B valuation round, Agility Robotics operates its dedicated RoboFab, and Chinese OEMs (Unitree, UBTech, AgiBot) leverage state subsidies and mandated output targets. While market attention remains fixated on AI compute and foundation models, the most severe physical scaling bottleneck resides in precision mechanics.

A humanoid robot incorporates 30 to 40 actuators: dozens of rotary joints utilizing strain wave (harmonic) or cycloidal (RV) reducers, alongside linear actuators powered by planetary roller screws. A standard industrial robotic arm uses 4 to 6 precision reducers; a single humanoid requires several times that amount. Manufacturing these motion components demands micron-level tolerances, decades of metallurgical know-how, and specialized high-precision machine tools. Consequently, the supplier base is intensely concentrated: Harmonic Drive controls an estimated two-thirds of the strain wave market outside China, Nabtesco holds ~60% of the heavy cycloidal RV market, and planetary roller screws are produced by only a handful of global specialists. Adding greenfield capacity requires 18 to 24 months, meaning supply cannot swiftly adjust to demand spikes.

What this thesis rests on

Each one is a statement that has to be true. We go back to every live thesis on a schedule and check these against public sources, so each leg carries either what we found or the date we look next.

Archive: reviews of this thesis are paused. Last review September 17, 2026.

Harmonic Drive Systems raises or reaffirms its FY2026 full-year order intake and operating profit guidance in the results released between July 2026 and November 2026, rather than cutting it.
financial · Open
Checked September 17, 2026, nothing published either way
Harmonic Drive Systems and Nabtesco each report order backlog at the end of the September 2026 quarter above the level reported at the end of the March 2026 quarter.
operational · Open
Checked September 17, 2026, nothing published either way
Leader Harmonious Drive Systems reports year-on-year revenue growth of at least 10% in its 2026 half-year and third-quarter reports.
financial · Supported

A source confirms this · August 31, 2026

2026 half-year report filed 27 Aug 2026: operating revenue RMB 348.55m, +38.64% YoY; net profit attributable to shareholders RMB 70.11m, +31.25% YoY. Above the 10% threshold for the half-year leg; the third-quarter report is not yet due.
See the source
Neither Harmonic Drive nor Nabtesco reports a decline in precision reduction gear average selling prices or segment gross margin attributed to Chinese competition in the results published through December 2026.
competitive · Supported

A source confirms this · August 31, 2026

Nabtesco H1 to 30 Jun 2026 (tanshin, 31 Jul 2026): Component Solutions revenue ¥45,521m, +23.6% YoY, segment operating profit ¥4,841m, +160.8%, on good precision reduction gear demand from Chinese/Korean robot makers. No selling-price or gross-margin decline cited; full-year OP guidance raised to ¥32.6bn.
See the source
Tesla, Figure or Agility Robotics does not announce before mid-December 2026 a delay of its 2026 humanoid production ramp to 2027 or later, nor in-house production of rotary joint actuators replacing external gear suppliers.
operational · Open
Checked September 17, 2026, nothing published either way

Show the causal chain step by step

Mass humanoid robot production starts (Tesla, Figure, Chinese OEMs)
Single humanoid requires 30-40 actuators and dozens of precision gears
Surging demand for harmonic drives, RV reducers, and roller screws
Concentrated supply: micron tolerances, deep know-how, 18-24mo capex lead times
Expanding backlogs, extending lead times, and pricing power for motion suppliers
Harmonic Drive, Nabtesco, THK, Leader Drive: order acceleration and margin expansion

This thesis, over time

What has happened since publication. Every entry is computed as of its own date, from price history.

September 27, 2026

Day 75 of 150: basket -18.4%

The thesis is halfway to its verdict. The equal-weight basket stands at -18.4% since publication day. The benchmark (S&P 500) returned +0.5% over the same window, so the basket trails its market by 18.9 percentage points. The strongest contributor is 6268 (-12.1%), the weakest is 688017 (-25.6%). 75 days remain until the verdict.

September 17, 2026

Review on day 65: nothing moved

We went back to this thesis on day 65 of 150 and checked each of its 5 assumptions against public sources. No news either way: 5. The equal-weight basket stands at -18.4% since publication day. The benchmark (.INX) returned +0.5% over the same window, so the basket trails its market by 18.9 percentage points. 85 days remain until the verdict.

August 31, 2026

Review: 2 of 5 assumptions confirmed

We went back to this thesis on day 48 of 150 and checked each of its 5 assumptions against public sources. Confirmed by the sources: 2. No news either way: 3. The equal-weight basket stands at -16.4% since publication day. The benchmark (.INX) returned +2.2% over the same window, so the basket trails its market by 18.6 percentage points. 102 days remain until the verdict.

August 20, 2026

Day 38 of 150: basket -14.1%

The first quarter of the horizon is behind this thesis. The equal-weight basket stands at -14.1% since publication day. The benchmark (S&P 500) returned +1.4% over the same window, so the basket trails its market by 15.5 percentage points. The strongest contributor is 6481 (-3.0%), the weakest is 6324 (-22.7%). 112 days remain until the verdict.

July 29, 2026

Basket fell below -20% since publication

This basket is at -21.9% since publication, 15 days after the thesis was posted. The benchmark (S&P 500) returned -3.0% over the same window, so the basket trails its market by 18.9 percentage points. The strongest contributor is 6481 (-15.1%), the weakest is 6324 (-27.9%). This is not a change of status: the horizon keeps running and the verdict only lands once it expires.

July 17, 2026

Basket fell below -10% since publication

This basket is at -11.2% since publication, 3 days after the thesis was posted. The benchmark (S&P 500) returned -1.1% over the same window, so the basket trails its market by 10.0 percentage points. The strongest contributor is 6481 (-7.0%), the weakest is 688017 (-15.7%). This is not a change of status: the horizon keeps running and the verdict only lands once it expires.

This is analysis, not investment advice and not a recommendation to buy or sell anything. We publish it and track it in public, mistakes included. Any decision is yours and yours alone.