Solar just passed coal for the first time in American history. I screened 27 stocks down to three toll booths — the wafers, the switchgear, and the batteries.
In May, solar made more of America's electricity than coal for the first time ever — 12.8% vs 12.2%. And when America builds new power now, 91% of it is solar and batteries.
Panels are the crowded, cutthroat part of the boom. I ran 27 supply-chain stocks through two filters — sales growing every single quarter, and a product so scarce the whole industry lines up for it.
Three passed: GLW (the only US solar wafer source), FPS (grid switchgear, up every quarter), EOSE (the only zinc long-duration battery). Full 27-name scorecard at the bottom.
The gold rush rule: don't dig for gold. Sell the shovels.
In every boom, the famous names fight each other to the death while their suppliers quietly get rich. Solar is no different: dozens of companies make panels, and they compete each other's prices into the ground. But a panel cannot exist without a wafer (the silicon slice inside it), cannot touch the grid without switchgear (the industrial breakers and transformers that connect it), and cannot power your house at night without a battery. Those three parts are in shortage. The companies that sell them are toll booths — they get paid no matter which panel brand wins.
The screen: 27 stocks, two questions. Is revenue growing every quarter, back to back — not just "up from last year," which seasons and hype can fake? And does the company sell something scarce? Two yes answers or it's out. The full scorecard is at the bottom.
In a gold rush, most miners go broke — the people selling picks and shovels get rich either way. These three sell the "shovels" of the solar boom: the silicon slices panels are made from, the electrical gear that plugs everything in, and the batteries that store sunshine for night.
The wafer monopoly. 80% of its capacity is already reserved — five years out.
Corning is the only company in America making the silicon slices every US-built panel starts from — and buyers already reserved most of the next five years. It is the only flour mill in a country that just decided to bake all its own bread. The ovens are running a little late.
The switchgear compounder. $2B of orders waiting — and they come in 2x faster than they ship.
Every data center and solar farm needs giant industrial switches and transformers to plug into the grid — and America cannot make them fast enough. Forgent grows every quarter because customers cannot afford to wait, and it is sitting on two billion dollars of orders. The flag: the owners who built it are cashing out while it is hot.
The zinc battery bet. $24B of customer interest — priced like it's going to zero.
Eos makes big batteries from zinc instead of lithium — cheaper, fireproof, all-American, and they store power longer. Customers want $24 billion worth, and new rules punish its Chinese-supplied rivals. But it still spends far more than it earns and is selling new shares to survive the ramp. It works big or it doesn't work at all.
One boom, three roles. The anchor, the engine, and the flyer.
The build-out pays all three no matter which panel brand wins: the wafer before a panel exists (GLW), the switchgear to plug anything in (FPS), the battery to keep lights on after sunset (EOSE). But they are not the same kind of bet — they ladder by risk, and they do different jobs side by side.
Think of it like a road trip crew: GLW drives the car (steady, gets you there), FPS reads the map and sets the pace (fast but needs attention), EOSE bought the lottery ticket at the gas station. Same destination, three different jobs — and you would not want three of any one of them.
All 27 stocks, two filters. Only three check both boxes.
Growing every quarter? — sales up sequentially, quarter after quarter, not just "up from last year." Sells something scarce? — a product the build-out cannot get anywhere else. Two checks = pick. Scarce but not growing = watch list. Neither = cut. Size is market cap — smaller companies move more, both directions.
| Stock | Size | Growing every quarter? | Sells something scarce? | Verdict |
|---|---|---|---|---|
| ✓✓ — The picks: growing AND scarce | ||||
| GLW solar wafers | $160B | ✓ | ✓ | PICK |
| FPS switchgear | $13.4B | ✓ | ✓ | PICK |
| EOSE zinc batteries | $1.7B | ✓ | ✓ | PICK |
| ✓ on scarcity only — watch for growth to flip | ||||
| FSLR solar panels | $24.6B | ✗ | ✓ | WATCH |
| GEV turbines & grid gear | $260B | ✗ | ✓ | WATCH |
| ETN electrical equipment | $140B | ✗ | ✓ | WATCH |
| PWR grid construction | $98.6B | ✗ | ✓ | WATCH |
| POWL switchgear | $8.5B | ✗ | ✓ | WATCH |
| VRT data center power | $117B | ✗ | ✓ | WATCH |
| ✗✗ — failed both filters | ||||
| NXT panel trackers | $16.4B | ✗ | ✗ | CUT |
| SHLS site wiring | $1.7B | ✗ | ✗ | CUT |
| TE panel factories | $1.9B | ✗ | ✗ | CUT |
| ENPH home inverters | $5.7B | ✗ | ✗ | CUT |
| RUN home solar | $2.9B | ✗ | ✗ | CUT |
| FLNC battery projects | $3.0B | ✗ | ✗ | CUT |
| TSLA EVs + storage | $1.5T | ✗ | ✗ | CUT |
| HUBB grid parts | $25.3B | ✗ | ✗ | CUT |
| CEG nuclear power | $104B | ✗ | ✗ | CUT |
| VST power plants | $55.2B | ✗ | ✗ | CUT |
| TLN nuclear power | $16.6B | ✗ | ✗ | CUT |
| NRG power plants | $29.1B | ✗ | ✗ | CUT |
| NEE utility + renewables | $185B | ✗ | ✗ | CUT |
| FCX copper | $85.3B | ✗ | ✗ | CUT |
| HL silver | $10.4B | ✗ | ✗ | CUT |
| PAAS silver | $18.6B | ✗ | ✗ | CUT |
| AG silver | $8.5B | ✗ | ✗ | CUT |
| WPM metals royalties | $50.8B | ✗ | ✗ | CUT |
The watch-list six own something scarce but their sales breathe with seasons or big project timing — one strong quarter flips any of them to a pick. The bottom eighteen either ride the boom without owning anything scarce, or their sales follow metal prices and weather instead of demand.
The next six weeks decide who was right.
| Date | What happens |
|---|---|
| July 21 | EOSE — the new-share sale closes. How much gets bought at $5.48 tells you what big money thinks of the ramp. |
| Late July | GLW — Q2 earnings (expected). The number to watch: is the wafer factory ramp back on schedule after the delay? |
| July 30 | FSLR — earnings, plus the pending tariff ruling around it. A clean print flips the biggest watch-list name toward a pick. |
| ~Late August | FPS — fiscal Q4 report (expected). The streak is the story: quarter six of growth in a row, or the first crack. |
Each pick has a near-term test: does Eos raise its money, does Corning fix its factory delay, does Forgent keep the streak. Dates marked "expected" are estimates, not confirmed.
The three picks, scored. Quality is a fundamental read, not a price call.
| Ticker | Price | ~Mkt cap | Quality | Risk / reward |
|---|---|---|---|---|
| GLW | $186.44 | ~$160B | 8.0 | Lower risk — real monopoly inside a giant; solar upside diluted by the conglomerate |
| FPS | $41.53 | ~$13.3B | 7.0 | High — fastest grower in the screen, priced for perfection, PE overhang above |
| EOSE | $4.35 | ~$1.7B | 5.0 | Speculative — real tech monopoly, real burn; dilution in progress, lottery-ticket sizing |
Own the tolls, not the traffic.
Solar passing coal isn't a projection — it happened in May. And 91% of new US power capacity is already solar + storage. The build-out is funded and under way.
GLW = the anchor (wafer monopoly, sleeps well) · FPS = the engine (the compounding, watched quarterly) · EOSE = the flyer (multiple-or-nothing). Different bets, same tailwind.
Each pick proves itself inside six weeks: EOSE's share sale closes July 21, GLW's wafer-ramp answer lands with Q2 earnings in late July, FPS faces streak-quarter six in late August.
The screen picked them. The conviction call is mine to add.
Want to see what price I bought in at — or the other stocks I'm in? Click the button below to join the Discord.
Want to see what price I actually bought in at?
Join the Discord to find out! →Ember analysis of EIA data (solar 12.8% vs coal 12.2%, May 2026) · Corning Q1 2026 results & Springboard updates · Forgent Power FQ3 2026 results (May 14) & June follow-on filing · Eos Energy Q1 2026 results, Line 2 launch (June 17), Frontier Power PO (June 18) & July rights offering · quarterly financials via stockanalysis.com (S&P Global Market Intelligence) · live prices via brokerage feed, July 14, 2026 intraday.