The Bottleneck Trade: 5 Small- and Mid-Cap Stocks With the Best Setup (CW 13, March 2026)

Most stock ideas start with a story. I think this crazy market is better understood as a game.

There are five important players right now: central banks, hyperscalers and chip customers, utilities and grid operators, governments, and the narrow suppliers that sit at the system bottlenecks. The key point is that several of these players are no longer acting in a normal discretionary way. Fed Governor Michael Barr said on March 24 that rates may need to stay steady “for some time” because inflation remains above target. At the same time, the U.S. Energy Information Administration expects electricity load to rise 1.9% in 2026 and 2.5% in 2027, with the highest pressure in regions like ERCOT and PJM, while the IEA says global data-center electricity use is on track to roughly double to around 945 TWh by 2030. Europe is running a similar strategic game in defense: the European Defence Agency estimates 2025 defense spending at €381 billion, and the European Commission’s Readiness 2030 plan aims to unlock more than €800 billion of defense spending, with SAFE as a €150 billion loan pillar.

That changes the payoff matrix. Hyperscalers cannot rationally underinvest in AI infrastructure if rivals keep spending. Governments in Europe cannot rationally underprepare in defense if peers rearm faster. Utilities cannot ignore load growth and reliability stress. In repeated-game terms, the dominant strategy is continued capex in a few specific lanes, even if the macro backdrop stays awkward. The bargaining power therefore shifts away from the glamorous end-product and toward the chokepoints: electrical gear, grid contractors, semiconductor inspection and metrology, defense electronics, and qualified missile/space components. That interpretation is reinforced by current supply-chain data: Broadcom said this week that TSMC capacity remains a bottleneck through 2026, while SIA and WSTS still point to a 2026 semiconductor market approaching $1 trillion.

That is the framework I used here. I also applied two hard screens. The stock had to still look like a genuine small- or mid-cap today, not six months ago, and it had to show clearly strong three-year performance. After re-checking the numbers, the five names I would keep are Powell Industries, MYR Group, Camtek, HENSOLDT, and Ducommun.

1) Powell Industries (NASDAQ: POWL)

Powell is still the cleanest power-infrastructure bottleneck name on the list.

It remains a true mid-cap at roughly $3.9 billion. Its three-year return is about +1,357%. Operationally, the latest quarter was strong: revenue rose 4% year over year to $251.2 million, gross profit rose 20%, net income rose 19%, new orders jumped 63% to $439 million, and backlog reached $1.6 billion. Cash and short-term investments were just over $500 million. Most importantly for the thesis, Powell said data-center orders in the quarter were well above $100 million and included its first data-center megaproject booking.

Game-theoretically, Powell benefits because customers do not want to be the player that loses time on energized capacity. When power delivery becomes the gating factor for AI, utility, LNG, and industrial projects, the supplier of critical switchgear and electrical systems gains leverage. That is exactly the sort of non-optional spend I want in this environment.

2) MYR Group (NASDAQ: MYRG)

MYR is less flashy than Powell, but it fits the same broad game: grid and electrical infrastructure spending that customers cannot postpone indefinitely.

MYR’s market cap is about $3.2 billion, and its three-year return is roughly +137%. Its 2025 numbers were strong and current: fourth-quarter revenue rose 17.3% to $973.5 million, full-year revenue reached a record $3.66 billion, full-year net income rose to $118.4 million from $30.3 million, and backlog hit a record $2.82 billion. Management explicitly tied its confidence to increased electrification demand and continued investment in electrical infrastructure.

The reason MYR belongs here is that the grid side of the bottleneck is still underappreciated. Everyone wants to own AI demand. Fewer people want to own the contractors that actually expand and connect the physical network that makes rising power demand investable. In a market where reliability and load growth matter more than perfect macro timing, that is a good place to be.

3) Camtek (NASDAQ: CAMT)

Camtek is my preferred small-/mid-cap semiconductor pick because it sits in inspection and metrology rather than in the most crowded compute narrative.

It still fits the size rule, with a market cap around $8.2 billion, and its three-year return is about +549%. The company’s 2025 report was excellent on the operating side: fourth-quarter revenue rose 9% to a record $128.1 million, full-year revenue rose 16% to a record $496.1 million, non-GAAP net income rose 15% to $159.0 million, and cash, deposits, securities, and equivalents reached $851.1 million. Management said it expects another double-digit growth year in 2026 and pointed directly to aggressive capacity expansion by customers. Camtek’s end markets include advanced interconnect packaging, heterogeneous integration, memory and HBM.

This is classic bottleneck logic. When chipmakers and advanced-packaging players are in an arms race, they can delay some things, but they cannot afford lower inspection quality or weaker yield control if competitors keep scaling HBM, chiplets, and advanced packaging. That makes process-control spending structurally stickier than many investors assume.

4) HENSOLDT (Xetra: HAG)

This is the best non-U.S. addition after re-checking both the accessibility and the size filter.

On Xetra, HENSOLDT’s market cap is roughly €8.3 billion, which keeps it in mid-cap territory by a practical screen, and its three-year return on the German listing is about +124%. Its 2025 results were strong: order intake rose 62% to €4.71 billion, order backlog rose to €8.833 billion, revenue increased to €2.455 billion, adjusted EBITDA rose to €452 million, and adjusted free cash flow reached €347 million. For a Europe-based retail investor, it is also much easier to buy than a small Nordic subcontractor or obscure OTC line because it trades on Xetra.

HENSOLDT fits the game because Europe is no longer playing a one-shot defense spending game; it is playing a repeated one. Once governments start moving toward joint procurement, air defense, sensors, and readiness goals, they need qualified electronics and sensor suppliers with real installed positions. In that setup, defense electronics can be a better strategic niche than chasing the largest prime contractors after they have already rerated.

5) Ducommun (NYSE: DCO)

Ducommun is the smallest company on the list, and that is part of the attraction.

Its market cap is about $1.86 billion, and its three-year return is roughly +135%. The latest results support the case: fourth-quarter revenue rose 9.4% to a record $215.8 million, remaining performance obligations reached a record $1.106 billion, and the quarter’s book-to-bill ratio was 1.3x. Management said defense growth was driven by missile platforms, fixed-wing aircraft, and rotorcraft, while military and space represented $706.5 million of year-end backlog.

Why does DCO fit the framework? Because defense supply chains are not commodity markets. Qualification matters, switching costs matter, and missile-related bookings matter. In a sector where governments want faster delivery and redundancy, second-tier component and subsystem suppliers can end up in a stronger bargaining position than the market expects. Ducommun is not the prettiest story stock here, but it is one of the cleaner bottleneck businesses.

Final thoughts: own the constraint, not the story

The common thread across all five names is simple: they operate where spending is becoming less discretionary and more strategic. That is the heart of the bottleneck trade. In a normal cycle, investors can get away with chasing whichever narrative sounds biggest. In this cycle, I think the better approach is to own the companies that sit where the system cannot move forward without them.

That is why this list is not filled with the most obvious mega-cap winners. The real opportunity, in my view, sits one layer below the headline names: in the electrical equipment that powers new capacity, the contractors that connect and expand the grid, the inspection tools that protect semiconductor yield, and the defense suppliers that sit inside Europe’s and America’s rearmament push. These are not “story stocks.” They are leverage points inside markets where customers are under pressure to keep spending.

If the game-theoretical framework is right, these businesses do not need a perfect economy to work. They only need the other players in the system to keep following their dominant strategy: hyperscalers keep building, utilities keep upgrading, governments keep rearming, and manufacturers keep protecting yield and delivery. As long as that remains true, the bottlenecks should continue to capture a disproportionate share of value.

That is why Powell, MYR, Camtek, HENSOLDT, and Ducommun stand out to me now. They are not identical bets, but they all sit on the right side of the same market structure.

And in this market, structure matters more than stories.

In a market full of noise, I’d rather own the chokepoints.

Disclaimer: This is not financial advice. This article is for informational purposes only and reflects personal opinion, not a recommendation to buy, sell, or hold any security. Investing involves risk, past performance is not a guarantee of future results, and you should do your own research and consider consulting a qualified financial advisor before making any investment decisions.