SOXL turns $10,000 into $131,000—almost nobody held

SOXL turned – Direxion’s 3x leveraged semiconductor ETF SOXL surged from a $17.24 starting point on May 27, 2025 to a $225.79 close on May 26, 2026—about a 1,209.99% gain over the exact one-year window. But the run’s violence, volatility swings, and day-to-day mechanics mea

A $10,000 bet on May 27, 2025 on Direxion Daily Semiconductor Bull 3X Shares—SOXL—turned into about $131,000 thirteen months later. The chart has the kind of punchline that travels fast: the fund closed at $225.79 on May 26. 2026. after starting from a $17.24 reference price on May 27. 2025. delivering a 1. 209.99% gain on the exact one-year window repeated in the headlines.

But the second half of the story is the part people don’t screenshot as easily. Almost nobody you know actually held it the whole way.

SOXL opened 2026 at $47.24 and is up 377.96% through May 26. Most of those gains arrived in violent stretches—one trading day on May 26 worth 18.49%, and a week of 48.65% that came after months of pain.

If you owned it through the run, you felt like a hero in your group chat. If you sold in March, you watched the same momentum play out from the outside—close enough to understand it, far enough to miss it.

The reason matters, because it changes what the headline “1,210%” is actually telling you.

SOXL isn’t a normal stock bet. It’s a 3x daily leveraged ETF tracking the ICE Semiconductor Index—the same basket that the unleveraged iShares Semiconductor ETF (NASDAQ:SOXX) follows.

So the cleanest check is simple: compare the two over the same window. SOXX returned 174.11% over May 27, 2025 to May 26, 2026. Three times 174.11% is 522—not 1,209.99%. That gap—about 688 percentage points—is where the mechanics live.

Both funds were exceptional. But only one of them is the kind of number that makes someone quit their job, and only one of them is capable of being something else entirely if the path changes.

For the same $10,000 invested in SOXX, the value became about $27,400. Both moves reflect strength. The difference is how leverage compounds when the market cooperates—and how quickly it can punish when it doesn’t.

Three things had to happen at once for the SOXL run to land the way it did.

First, the underlying story was real. Semiconductor demand kept ratcheting up on AI capex. NVIDIA (NASDAQ:NVDA) told investors that AI capex will grow 3x to 4x by the end of the decade. and the hyperscaler order books continued to validate that through 2025 and into 2026. Earnings from the largest weights inside SOXL’s basket kept beating and kept guiding higher—the SOXX story. in other words.

Second, volatility behaved. That’s not a side note for leveraged ETFs—it’s the engine. The VIX averaged 18.18 over the last twelve months and sat at 16.59 at the time of this comparison, at the 34.6th percentile of recent readings.

Leveraged funds hate chop and love sustained directional moves through low volatility. Over most of the twelve-month window, that’s exactly what happened. There was one violent disruption: a VIX peak of 31.05 on March 27, 2026. Readings were elevated from early March through April 7, where the bleed happened and where many holders capitulated.

After that disruption, the seven weeks since became the cleanest possible setup for 3x daily exposure—steady decline in the VIX from 25.78 on April 7 down to 16.59 as semis ripped on the other side.

Third, the bounce was enormous because the base was so low.

SOXL was at $128.32 on April 24 and at $225.79 on May 26, a 75.96% one-month move. The Moomoo community pointed out that SOXL’s April monthly gain of about 165% was the best single month since inception. That month was helped along by a 38% April for the PHLX Semiconductor Index, at a level not seen since February 2000.

A 3x fund recovering from a deep drawdown can generate outsized percentage numbers on the way back without ever reclaiming its old high in dollar terms. The path is what creates the math. The story can look simple in hindsight, but the numbers inside the run were anything but.

A longer view makes the “screenshot economy” feel even sharper. $100 invested in SOXL 15 years ago is now worth about $22,838, an average annual return of 44.22%, and you can fit several round-trips of total devastation inside that path.

Which is why the question becomes less “did SOXL go up?” and more “how many people stayed through the whole ride?”

The moment in the market where most people admit they didn’t is often the same moment where optimism spikes hardest.

Almost everyone was briefly the “right” kind of trader for a while.

Reddit sentiment flipped from a score of 18 (very bearish) on the morning of May 8 to 88 (very bullish) by that evening, the same week as the biggest activity spike in the dataset. On May 8 at 6pm ET, a wallstreetbets window showed 588 upvotes and 90 comments.

A widely circulated bearish post titled “Jr. Burry SOXL short” appeared in the same week the fund proceeded to do what it did.

One Moomoo user, quoted in news aggregations, wrote: “I keep selling… but it keep going up! 😢 am I too early?” It’s a single sentence that captures the retail experience of this kind of rally—selling for fear, rebuying for hope, and watching the price stay one step ahead.

Options markets showed the same tug-of-war. Open interest sat in the 1.0M to 1.2M contract range through April and May. Put-heavy days arrived at 65.77% puts on May 15 and 59.75% puts on May 12.

People hedged. People shorted. And the move kept going.

So what would have to stay true for the pattern behind the 1,210% number to keep working?

Three conditions sit at the center of the checklist—and each one is tied to what makes leveraged math either reward or punish.

First, the AI capex cycle has to keep absorbing the supply. The simplest tell is hyperscaler guided capex on quarterly calls and the data-center revenue trajectory across the basket’s top holdings. Jim Cramer’s May 26. 2026 show flagged Micron as the day’s notable spike and asked whether NVIDIA has become a compounder—framing that is the right question. but also the wrong moment to assume the answer is yes forever.

Second, volatility has to stay tame. With the VIX at 16.59 and trending lower, the math is working for leverage today. A sustained move back above the low 20s—like the regime from early March through early April—would change the outcome. Under that kind of environment, the same daily compounding that produced 1,210% can start subtracting from an account on flat weeks.

“Watch the VIX, not the S&P” is the practical instruction embedded in the numbers.

Third, the price has to keep behaving as it is behaving now. SOXL closed at $225.79 on May 26 and started the year at $47.24. The same 75% monthly move that lifted the fund from $128 to $225 would take a holder buying today to something near $400. But doing that would require the same combination of low volatility and one-directional semiconductor strength.

That’s also why earlier warnings stayed relevant. 24/7 Wall St.’s May 7 piece described SOXL as “a tactical instrument for active traders due to volatility decay and symmetric downside. making it unsuitable for buy-and-hold investors. ” and that framing “has not changed” because the chart looks different than it does for long-term investors.

The honest read is that the 1,210% number is real. The mechanism that produced it is regime-dependent. And the regime is still cooperating—at least for the moment.

If the VIX climbs back into the 20s and stays there, or if a single bad guide from the basket’s largest weight cracks the AI capex story, the same 3x daily structure that minted the screenshot will run the projector in reverse.

That is the indicator. That is the contract.

Anyone telling you otherwise is selling you the chart.

SOXL Direxion Daily Semiconductor Bull 3X Shares semiconductors leveraged ETF VIX AI capex NVIDIA volatility decay SOXX ICE Semiconductor Index