For the better part of three years, Wall Street had a simple playbook: bet on the Magnificent Seven tech giants and watch the money roll in. That strategy is suddenly looking less magnificent.
The Roundhill Magnificent Seven ETF (MAGS), which offers concentrated exposure to the tech behemoths driving the artificial intelligence boom, has stumbled into negative territory in early 2026. Meanwhile, the broader market is quietly leaving it behind.
Here's the striking part: 493 of the 500 stocks in the S&P 500 are outperforming the Magnificent Seven so far this year. Two exchange-traded funds are capturing this divergence in real time, and the numbers tell a compelling story about what might be a genuine regime change on Wall Street.
The Numbers Tell the Story
As of Thursday, MAGS is down 5.60% year-to-date. The Defiance Large Cap ex-Mag 7 ETF (XMAG)—which tracks the S&P 500 with the Magnificent Seven deliberately excluded—is up 3.23% over the same period. That's a nearly 9-percentage-point gap, which in market terms is enormous.
MAGS holds equal-weight positions in NVIDIA Corp. (NVDA), Meta Platforms Inc. (META), Apple Inc. (AAPL), Alphabet Inc. (GOOG) (GOOGL), Tesla Inc. (TSLA), Amazon.com Inc. (AMZN) and Microsoft Corp. (MSFT). The fund carries $3.50 billion in assets under management, so plenty of investors are feeling this pain.
What Exactly Is XMAG?
XMAG launched on Oct. 21, 2024, which in retrospect was excellent timing. The fund tracks the BITA US 500 ex Magnificent 7 Index, a benchmark that includes the 500 largest U.S. publicly traded securities while explicitly excluding the seven mega-cap tech names that dominated recent years.
The index uses free-float market capitalization weighting and rebalances quarterly. As of Thursday, the fund holds 493 stocks and carries net assets of $137.79 million. It's essentially a way to bet on "everything else" in the large-cap universe.
The fund exists because some investors suspected that the Magnificent Seven's dominance couldn't last forever, and that valuation gaps would eventually close. So far in 2026, that thesis is playing out exactly as hoped.













