The Magnificent 7
WeightingFrom January 2015, hold Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta and Tesla in equal weight (one-seventh each), rebalanced back to equal weight on the first trading day of each January. Benchmark is QQQ (the Nasdaq-100 ETF).
See this in the terminal — live earnings bands, fair value and rates.
Start free trialRead the story
Seven stocks. That's the whole strategy. And it buried the index everyone owns.
The rule. From January 2015, hold Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta and Tesla in equal weight (one-seventh each), rebalanced back to equal weight on the first trading day of each January. Benchmark is QQQ (the Nasdaq-100 ETF).
What actually happened.
- $10,000 became $435,804. The same $10,000 in Nasdaq-100 (QQQ) became $78,732. This strategy beat the benchmark by $357,072.
- Compounded at 37.9% a year vs 19.2% for the benchmark.
- Best year: 2020 (+161%). Worst year: 2022 (-46%). This is not a smooth ride.
The catch. At its worst, this strategy fell -50% from a prior peak — deeper than the benchmark's -35%. Would you have held through that? Most people don't. That's the whole game.
The takeaway. It worked here — but "it worked in the past" is the most expensive sentence in investing. The edge is real only if it survives costs, taxes, and the drawdown above. Show your work, then decide.
Holding right now (2026-01-02): AAPL, AMZN, GOOGL, META, MSFT, NVDA, TSLA.
Drawdown from prior peak
Calendar-year returns
| Line | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26* |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Magnificent 7 (equal wt) | +42% | +37% | +50% | -1% | +51% | +161% | +53% | -46% | +113% | +61% | +23% | +10% |
| Nasdaq-100 (QQQ) | +10% | +7% | +33% | -0% | +39% | +48% | +27% | -33% | +55% | +26% | +21% | +21% |
* Partial year — the backtest does not cover this whole calendar year, so it is excluded from the best and worst year figures.
Holding now · 2026-01-02
This is the definition of hindsight: the “Magnificent 7” label was applied AFTER these names had already won. Picking them in 2015 would have required foresight nobody had. How we test
Rebalance: annual (first trading day of January), equal weight · Window: 2015-01-01 → 2026-09-28 · Last run Sep 28, 2026 · JSON