The Magnificent Seven Stocks, Measured 2015 to 2026
The Magnificent Seven are seven giant US technology-led stocks: Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta Platforms and Tesla. From the end of 2014 to August 2026 every one of them rose more than the S&P 500, and every one also suffered a deeper fall from a high than the index did.
The Magnificent Seven is a market nickname for seven very large US companies whose shares led the stock market through the 2010s and early 2020s. The name describes what happened, and on the numbers below it happened with a great deal of pain along the way.
Who is in it
Seven companies, eight share lines. Apple (AAPL), Microsoft (MSFT), Alphabet (GOOGL and GOOG, two classes of the same company), Amazon (AMZN), Nvidia (NVDA), Meta Platforms (META, formerly Facebook) and Tesla (TSLA). This page uses GOOGL for Alphabet throughout.
The label is usually credited to Bank of America strategist Michael Hartnett in 2023, borrowing the title of the 1960 western. It is a nickname, not an index: no committee maintains it, and nothing decides when a company joins or leaves. That makes it different from the S&P 500 or the Nasdaq 100, which have written rules.
Index weights are not quoted here. The seven carry a large share of the S&P 500’s value, but the exact share changes daily with their prices, and this page quotes only figures it can recompute from its own files. S&P Dow Jones Indices publishes the current weights.
How it works
The group matters because of how the big indexes are built. The S&P 500 weights each company by its market value, so the largest companies move the index most. When seven of the largest rise or fall together, they can decide a day, a month or a year for the index on their own.
That link runs both ways for anyone holding an index fund. Owning the S&P 500 or the Nasdaq 100 already means owning a large slice of the seven. Buying them separately on top of a fund doubles down on the same bet, which is where concentration risk comes from.
They are not one business. A phone maker, a software and cloud company, a search and advertising company, an online retailer, a chip designer, a social media company and a carmaker share a nickname, not a product. What they shared from 2015 to 2026 was a result.
A worked example
Take a hypothetical $1,000 at the close on 31 December 2014, split equally, about $142.86 in each of the seven, and hold it to 10 August 2026. Prices are the daily closes from the Yahoo Finance feed, adjusted for splits, with dividends left out.
The seven slices end up worth about $73,910.53 together. Put the same $1,000 into the S&P 500 at its closing level of 2,058.90 and it ends at 7,753.11, which on price alone turns $1,000 into about $3,765.66.
Now look inside the $73,910.53. Nvidia’s slice alone is about $61,995.95, because its split-adjusted price went from $0.50 to $217.55. That one slice is 83.9% of the whole basket. The other six, from Meta’s $1,089.32 to Tesla’s $3,187.94, share the rest.
So “the Magnificent Seven did well” mostly means one stock did extraordinarily well. Anyone studying the group should check whether a finding survives with Nvidia removed, because a result that disappears without it is a finding about one company.
Why the seven dominate a backtest
They are the stocks people test first. They are liquid, famous and on every chart platform’s front page, so a swing rule is often tried on them before anything else. That is a list chosen by its outcome, and it flatters any rule that buys dips.
The swing backtests on this site started there. In August 2026 ten rules were run on the seven from January 2015 to August 2026 in TradingView’s Strategy Tester. A run passed only if at least 60% of trades closed higher, the profit factor was 1.3 or more and there were at least 30 trades.
The original data
Every one of the seven beat the index on price over the whole window. From the close on 31 December 2014 to 10 August 2026, the S&P 500 rose +276.57%. Meta, the weakest of the seven, rose +662.52%. Microsoft rose +989.47%, Apple +1,017.09%, Alphabet +1,247.45%, Amazon +1,692.11%, Tesla +2,131.56% and Nvidia +43,297.17%.
Every one of the seven also fell harder than the index at its worst. The S&P 500’s deepest fall from a prior high in the window was −33.9%, from February to March 2020. Each of the seven did worse, and for six of them the worst fall ran through 2022.
Meta fell −76.7% from September 2021 to November 2022. Tesla fell −73.6%, Nvidia −66.4% and Amazon −56.1%, all from 2021 highs. Even the gentlest, Microsoft at −37.6% and Apple at −38.7%, were worse than the index’s worst. Apple’s came earlier, from October 2018 to January 2019. These falls are measured from close to close, and intraday lows, which the measure ignores, went at least as deep. A reader holding any one of the seven through that stretch would have watched well over a third of the position’s value disappear at some point in the window.
The backtests tell a narrower story than the prices do. Across 70 runs, ten rules on seven stocks, 30 cleared all three bars. Alphabet and Meta passed on 6 rules of 10, Microsoft, Amazon and Nvidia on 4, and Apple and Tesla on 3. Only the RSI(2) rule passed on all seven, and three of the ten rules passed on none.
So even on the decade’s winners, most rules did not clear the bar on most stocks. The per-stock figures are in the published price summary, with the daily closes behind it and the backtest results beside them. All of it is past data, price only, and none of it is a forecast.
When it fails
It fails as a forecast. The name was coined after the gains, from a list of stocks that had already won. The same selection made in an earlier decade would have picked a different group, and nothing in one decade’s result obliges the next decade to repeat it.
It fails as diversification. Six of the seven had their worst fall in overlapping stretches between July 2021 and January 2023, so they were not six separate bets. In a broad decline tied to technology or interest rates they have tended to fall together.
It fails as a test universe. A rule that works on these seven has been tested on the ten-year winners. When the same swing code ran on five large caps that had lagged, the pass rate for its three best rules collapsed, which the backtesting page explains in general.
And it fails as a unit. The equal-split example above is 83.9% one stock. Averages across the seven say little about any one of them, and the gap between Meta’s +662.52% and Nvidia’s +43,297.17% is wider than the gap between Meta and the index.
Related
The S&P 500 page explains the index the seven are measured against, including its own worst years. Concentration risk covers what happens when a few holdings decide a portfolio’s result, which is the practical risk of owning the seven twice.
The Nasdaq 100 is the index where the seven weigh most heavily. And drawdown explains how to read the falls measured above, and why a −76.7% fall needs a gain of more than 300% to recover. Nothing here suggests buying or selling any of the seven.
Treat the name as a description of the past decade, not a category with a future attached. Before buying any of the seven, or a fund heavy in them, look at the size of each one’s worst fall on this page and decide whether that fall is survivable in your account.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.