Quantum BiasNvidia Stock and the Magnificent 7 vs the S&P 500: Worth the Risk?
What the market's biggest winners, Nvidia first, actually put their holders through.
In this article
All seven beat the index over their full history, and Nvidia did the best of all since 2012, but none of them made it easy. Since 1999 (or their listing), Apple, Microsoft, Nvidia, Amazon, Alphabet, Meta and Tesla each fell between 63% and 94% from a peak at some point, against a worst drop of 55% for the S&P 500. Each one beat the index in only 50% to 54% of weeks. And they're the seven survivors everyone knows about today, picked with hindsight.
| Question | Data | Answer |
|---|---|---|
| How did the seven largest U.S. stocks compare with the S&P 500 in return and risk? | Weekly total returns, Nov 1999 – Oct 2026 (later starts for Alphabet, Tesla, Meta) | Higher returns, 1.6 to 3.4 times the volatility, drawdowns of 63% to 94% |
Why look at the giants
If you trade NQ or ES, these seven companies aren't just stocks. Together they carry a big share of both indexes: on September 30, 2026, the seven together, counting Alphabet's two share classes, weighed 37.6% of the Nasdaq-100, as shown in our concentration article. Their moves are the index's moves.
They're also the stocks most traders are tempted to own directly. So it's worth asking: what did holding them actually look like, compared with holding the index?
Full history
| Since | Annual return | Volatility | Worst decline | Date of the low | Worst week | Weeks beating SPY | |
|---|---|---|---|---|---|---|---|
| S&P 500 (SPY) | Nov 1999 | 8.5% | 17.6% | −54.6% | Mar 2009 | −19.8% | — |
| Apple | Nov 1999 | 25.9% | 36.3% | −81.1% | Apr 2003 | −50.7% | 53.6% |
| Microsoft | Nov 1999 | 11.7% | 28.0% | −68.7% | Mar 2009 | −19.0% | 50.1% |
| Nvidia | Nov 1999 | 35.6% | 59.8% | −88.3% | Oct 2002 | −38.9% | 53.3% |
| Amazon | Nov 1999 | 17.0% | 47.5% | −94.4% | Sep 2001 | −32.0% | 52.0% |
| Alphabet | Sep 2004 | 24.7% | 30.4% | −63.3% | Nov 2008 | −15.4% | 52.9% |
| Meta | Jun 2012 | 24.2% | 38.2% | −76.0% | Nov 2022 | −23.7% | 51.1% |
| Tesla | Jul 2010 | 42.7% | 55.4% | −72.2% | Jan 2023 | −25.9% | 50.8% |
Exact values from table 2 in this article. Bar lengths share a zero baseline; no new calculation or simulated price path. Read the table and its stated sample and limits in this article.
Three things jump out.
The returns were insane. Nvidia compounded at 35.6% a year for 27 years. Tesla at 42.7% since 2010. That's why they're the giants.
The drops were brutal. Amazon lost 94.4% between 1999 and 2001. Nvidia lost 88.3% by 2002. Apple lost 81.1% by 2003, including a single week of −50.7% in September 2000. Even in the last few years, Meta fell 76% into November 2022 and Tesla 72% into January 2023.
Week to week, it was close to a coin flip. None of the seven beat the S&P 500 in more than 54% of weeks. Their outperformance came from a small number of huge weeks, not from steady leadership.
Nvidia stock: the best return, and one of the deepest drops
Nvidia is the clearest example of the trade-off. Its 35.6% a year since 1999 is the highest of the four stocks that traded through the whole period, and its 59.1% a year since 2012 tops all seven. But its shares lost 88.3% from the 2000 peak to October 2002, and they fell 65.9% again between November 2021 and October 2022. Its weekly volatility, 59.8% a year, was more than three times the S&P 500's, and its beta of 1.73 made it the most sensitive of the seven to the market. Holding Nvidia meant sitting through two collapses of more than 60%.
Exact values from table 2 in this article. Bar lengths share a zero baseline; no new calculation or simulated price path. Read the table and its stated sample and limits in this article.
The same window for everyone
To compare fairly, here's the period since June 2012, when all seven were trading:
| Annual return, Jun 2012 – Oct 2026 | Worst decline in that period | |
|---|---|---|
| S&P 500 (SPY) | 15.1% | −31.8% |
| Apple | 23.1% | −43.6% |
| Microsoft | 24.4% | −35.0% |
| Nvidia | 59.1% | −65.9% |
| Amazon | 24.9% | −54.8% |
| Alphabet | 24.6% | −41.9% |
| Meta | 24.2% | −76.0% |
| Tesla | 44.0% | −72.2% |
Even in one of the best stretches ever for big tech stocks, every one of the seven had a worse drawdown than the index. Microsoft came closest.
How much of their movement was the market?
Each stock's weekly correlation with the S&P 500 was between 0.45 and 0.61. Roughly speaking, the market explained only about a fifth to just over a third of their week-to-week moves. The rest came from company news, earnings and sentiment.
Their sensitivity to the market, or beta, ranged from about 1.0 for Microsoft and Alphabet to 1.58 for Tesla and 1.73 for Nvidia. In a week when the S&P 500 fell 5%, Nvidia fell about 8.7% on average. That's just the market part, before any Nvidia-specific news.
What a 90% drop means in real life
A 94% drop, like Amazon's, needs a gain of more than 1,600% to get back to even. An 88% drop, like Nvidia's, needs more than 700%. Amazon's weekly close only got back to its December 1999 peak in October 2009. Investors who bought the top waited almost ten years, through the 2001 low and the 2008 crisis.
The index's worst drop, 54.6%, needed a gain of about 120%. That's a very different experience, and it's why holding the index through crashes is way more common than holding a single stock through one.
The survivorship problem
This is the most important caveat, and it applies to every "Magnificent Seven" chart you'll ever see. These are the seven biggest companies today. In 1999, nobody knew which companies would become giants. Several of the biggest companies of 2000 didn't stay on top. Cisco, briefly the most valuable company in the world in March 2000, spent more than two decades below its 2000 peak.
A fair test of "buying the giants" would buy the biggest companies at each point in time, including the ones that later shrank. That's very different from buying, with hindsight, the ones that won.
What this means for traders
- Size single stocks for their drawdowns, not their returns. A position that can drop 70% to 90% needs to be way smaller than an index position for the same risk. Our weekly range article gives range-based sizing ratios.
- If you trade NQ or ES, watch them as drivers. With weights of 7% to 8% each, a big move in Nvidia or Apple moves the index directly, no matter what the other 98 members do. Our semiconductors article measures the link for chip stocks.
- Don't confuse outcome with process. Holding these seven through their worst periods took conviction that, in real time, looked exactly like stubbornness.
How to check it yourself
- Download weekly adjusted prices for a stock and for SPY.
- Compute each one's running peak and how far below it the price is each week. The minimum is the worst drawdown.
- Count the weeks the stock beat SPY.
- Do the same for a stock that was big in 2000 and isn't a giant today, to see the other side of survivorship.
Limits
- Survivorship. The sample is picked with hindsight, so results overstate what a real-time investor could expect.
- Weekly closes. Intraday and daily drops were deeper in some cases.
- Different start dates. Alphabet, Tesla and Meta listed after 1999, so their full-history numbers don't include the 2000–2002 bear market.
How we did it
- Weekly adjusted prices (dividends and splits included) from Alpha Vantage (
TIME_SERIES_WEEKLY_ADJUSTED), through October 9, 2026. - Annual return: compounded weekly total return. Volatility: weekly standard deviation × √52. Beta: slope of weekly stock returns on SPY returns.
- Worst decline: largest fall from a running peak, on weekly closes.
Sources
- Nasdaq. (2026, September 30). Nasdaq-100 Index (NDX) fact sheet. indexes.nasdaqomx.com. Accessed October 12, 2026.
- Alpha Vantage. TIME_SERIES_WEEKLY_ADJUSTED (SPY, AAPL, MSFT, NVDA, AMZN, GOOGL, META, TSLA). Data retrieved October 2026.
FAQ
Did the Magnificent Seven beat the S&P 500? Over their full histories through October 2026, all seven had higher annual returns than SPY.
What's Nvidia stock's biggest drawdown? On weekly closes, Nvidia fell 88.3% from its 2000 peak to October 2002, and 65.9% from November 2021 to October 2022.
How much did Tesla fall from its peak? Tesla fell 72.2% on weekly closes into January 2023.
Is it riskier to own one of these stocks than the S&P 500? Yes, on this data: 1.6 to 3.4 times the volatility, and drawdowns between 63% and 94% against 55% for the index.
Do the Magnificent Seven beat the market every week? No. Each one beat SPY in only 50% to 54% of weeks.
Research and education only. Not financial advice. Past results do not predict future returns.
How this article was made: drafted with AI assistance from public data and the sources listed above; facts and figures checked by Gabriel Décarie and Karel Panneton.
The week just closed, summarized, and the economic calendar for the week ahead.