SOXL Leveraged ETF Trap: Chip Index Flat, 3x ETF Down 29%?

A giant 3x balloon swelling above a semiconductor chip with a steeply falling stock chart below

The Index Is Flat, but the 3x ETF Fell 29%?

✦ Same Index, Different Results

vs. July 29 close

SOXX (1x)-0.01%
Nov 12
SOXL (3x)-28.76%
Nov 12
SOXX (1x)-3.68%
Nov 13
SOXL (3x)-36.60%
Nov 13

On July 29, 2026, the iShares Semiconductor ETF (SOXX), a 1x chip ETF in the US, closed its regular session at $155.00 (retroactively adjusted for the 3-for-1 split on November 5). About four months later, on November 12, it closed at $154.99. That's a change of -0.01%.

The underlying index was right back where it started.

Meanwhile, the Direxion Daily Semiconductor Bull 3X ETF (SOXL), which tracks three times the same index's daily return, fell from $91.99 on July 29 to $65.53 on November 12. That's a drop of 28.76%, or about 29%. What on earth...??

The NYSE Semiconductor Index that both products follow is made up of 30 large US-listed chip stocks. Same underlying index, right? But the index hasn't moved while the 3x product's principal alone has shrunk by nearly 30%...

That's a hard one for investors to wrap their heads around.

SOXX climbed from $155.00 on July 29 to $196.50 on October 5 (+26.77%), then slid to $131.90, including a one-day plunge of 12.15% on November 5. On November 12 it got back to $154.99. It just followed the underlying index's path.

SOXL, on the other hand, ran from $91.99 to $164.27 on October 5 (+78.57%), then crashed to $40.87 on November 5 and stopped at $65.53 on November 12. SOXX made it back to where it started, but SOXL got crushed.

And a day later, on the November 13 close, SOXX was at $149.30 (-3.68% from late July) and SOXL at $58.32 (-36.60%), so the gap widened even more.

This has happened before. From May 29 to October 1 this year, SOXX gained 1.27%, from $189.69 to $192.11, while SOXL lost 31.49%, from $224.34 to $153.69. For all of 2024, SOXX rose 12.92% while SOXL actually lost 12.31%.

Look at the past five years through the November 13 close, with dividends reinvested. 1x SOXX returned +166.73% in total, and SPY, which tracks the S&P 500, gained +72.55%. Meanwhile SOXL couldn't even tread water at -4.83%.

✦ SOXL, SOXX and QQQ Over the Last 22 Trading Days

Regular-session closes, October 15, 2026 = 100

SOXLSOXXQQQ
3870101Oct 15Oct 29Nov 13
Nov 13 closes: QQQ 93.39, SOXX 87.02, SOXL 54.33. SOXL hit its low of 38.07 on Nov 5.

Put the last 22 trading days (October 15 to November 13) next to the broad market and the gap is easy to feel. On a closing basis, the Nasdaq 100 ETF (QQQ) fell 6.61% and 1x SOXX slid 12.98% over that stretch, while SOXL plunged 45.67%.

Trap #1: Why Does a Daily 3x Target Backfire?

The core reason a 3x product shrinks even when the index comes back to the start, or even rises, is how it's run. Here's the thing: this ETF isn't built to match 3x over the long run, but to match 3x of each day's return.

✦ Index Back to Square One, 3x Product Down

1x ($100)
-10% → +11.11% → $100
3x ($100)
-30% → +33.33% → $93.33

Say a $100 1x product drops 10% on day one to $90. If it rises 11.11% the next day, it's back to $100. A 3x product, though, falls 30% on day one to $70, and even a 33.33% bounce the next day only gets it to $93.33.

The index is back at the start, but $6.67 has vanished from the account. The bigger the price swings up and down, the faster the losses snowball... People usually call this 'negative compounding.'

✦ Gain Needed to Get Back to Even After a Drop

PeriodDropGain needed
SOXX Oct 5 → Oct 16-16.31%+19.48%
SOXL Oct 5 → Oct 16-42.98%+75.37%
SOXL 52-week high → Nov 13-80.69%+417.83%

If you think about it, the gain a 3x product needs to get back to even after a drop has to be far higher. From October 5 to October 16, SOXX fell 16.31%, from $196.50 to $164.46, and needed +19.48% to get back to even.

Over the same stretch, SOXL got cut 42.98%, from $164.27 to $93.67, so getting back to even takes a whopping +75.37%.

And there's more. From its 52-week high of $302.00, SOXL is down 80.69% at the November 13 close of $58.32. To get back to that high, it would have to climb +417.83%. That's some serious negative compounding.

Let's think about the worst case. Say the underlying index closes down more than 33% in a single day. Then the entire fund can lose 100% of its assets in a single day. A 3x leveraged ETF is a bet where you take on that kind of risk too.

But most investors don't know any of this and just assume it's a product that 'moves three times as much as the underlying asset.'

✦ 3x ETF One-Year Returns by Index Volatility

Assumes the index returns 0% or +10% over one year. Volatility is the index's annual volatility.

Index volatility 10%
Index at 0%
-3.0%
Index at +10%
+29.2%
Index volatility 25%
Index at 0%
-17.1%
Index at +10%
+10.3%
Index volatility 50%
Index at 0%
-52.8%
Index at +10%
-37.1%
Index volatility 75%
Index at 0%
-81.5%
Index at +10%
-75.4%
Index volatility 100%
Index at 0%
-95.0%
Index at +10%
-93.4%

Convert it to annual volatility and the size of the trap becomes clear. Excluding costs and dividends, if the index is flat after one year (0%), a 3x product loses 17.1% at 25% annual volatility, 52.8% at 50% volatility, and 81.5% at 75% volatility.

Even if the index gains 10% over the year, a 3x product still loses 37.1% when volatility hits 50%.

The NYSE Semiconductor Index's annualized volatility was 34.76% from its 2021 launch through 2024 (its highest over 2021 to 2024 was 44.38%), and SOXL has recently swung wildly, with a -38.71% plunge and a +22.25% surge in single days.

The more volatile the semiconductor index gets, the more SOXL investors' principal slowly melts away.

〈Why SOXL is a trap for retail investors〉
* From the Options Trading IQ YouTube channel.

Trap #2: The Daily Bill From Total Return Swaps and High Rates

On top of the 'negative compounding' we just covered, there's one more structural cost leaking out of the fund every day. SOXL holds about half of its $17.7 billion in fund assets directly in chip stocks.

The rest sits in short-term cash-like assets such as money market funds (MMFs), which are pledged as collateral to nine big investment banks, including Barclays, Goldman Sachs and JPMorgan.

In return, it signs total return swap contracts that reset its exposure to 300% at the market close every day.

✦ Good to Know

What is a total return swap?

A total return swap is a contract where the bank passes the index's gains and losses to the fund, and the fund pays the bank an interest-like cost in return. Using this contract, the fund can create price moves far bigger than its actual stock holdings would produce.

On the surface, SOXL's official net expense ratio is 0.75% a year (gross expense ratio 0.91%, or 0.71% excluding the fees of underlying funds such as MMFs), more than double the 0.33% of 1x SOXX.

On top of that, the swap financing cost, the price of borrowing 2x or more of extra exposure from the banks, comes out of the fund's net asset value (NAV) separately every day.

SOFR (the Secured Overnight Financing Rate, the US short-term funding benchmark) is sitting around 3.90% a year (3.90% on November 12, 30-day average 3.92%). In a high-rate environment like that, the financing cost paid to the banks is bound to go up too.

So it piles up as a fixed expense that eats into the fund's returns every day.

While SOXL Plunged 59.8%, South Korean Investors Got Trapped

✦ Money Poured In, Yet the Price Still Fell

SOXL

Global net inflows$7.3 billion
October
South Korean retail net buying$4.88 billion
Oct 10 - Nov 9
SOXL price-59.79%
Same period

Lots of investors do what's called 'averaging down': adding money during a decline to lower their average cost. Recently, SOXL took in a whopping $7.3 billion in net inflows across the global ETF market in October alone.

Yet over that exact same October (from $147.86 on September 30 to $77.05 on October 30), SOXL's price plunged 47.89%.

South Korean retail investors were especially aggressive buyers. Over the month from October 10 to November 9, they net bought $4.88 billion (about 6.6 trillion won) of SOXL, the most of any overseas stock and 7.7 times the $633.5 million for second-place Alphabet.

In the single week of November 2 to 6, 74.0% ($1.63 billion) of South Korean retail investors' total net buying of US stocks ($2.20 billion) was concentrated in SOXL alone.

South Korean investors' share of global SOXL holdings had already been climbing steadily, from 22.2% in March last year to 30.6% at the end of August this year.

Yet in the very month they poured in about $4.9 billion, October 10 to November 9 (October 9 close $139.76 to November 9 close $56.20), SOXL's price actually plunged 59.79%.

At the period's low on November 5 ($40.87), the drop reached 70.76%, and on November 13 it was still sitting at $58.32 (-58.27% from October 9).

The money they put in to lower their average cost took the full 3x decline, so they got stuck in a trap where adding more money only makes the losses bigger.

〈Ben Felix on the truth about leveraged ETFs〉
* From the Ben Felix YouTube channel.

When Leveraged ETFs Work, and the Turning Point Ahead

So are leveraged ETFs trash you should never buy? Not at all. In a market where the index keeps running in one direction, a leveraged ETF can shine.

When an uptrend (or a downtrend, for an inverse ETF) rolls on without a big pullback, 'positive compounding' kicks in, where today's 3x gain stacks on top of yesterday's bigger principal.

In fact, in 2023, when SOXX rose 67.13%, SOXL returned +227.02% for the year, well above a simple 3x (+201.39%). That's the kind of market where leveraged ETFs are strong.

On the flip side, when the trend turns and a deep bear market hits, recovery takes a long time too. In the 2022 chip bear market, SOXX fell 35.09% while SOXL collapsed 85.67%, and it took almost four years to fully regain its December 2021 high.

✦ Factors That Could Be SOXL's Turning Point

October inflation data
Producer prices +6.5% from a year earlier, 10-year yield at 5.15%
Nov 17
NVIDIA Q3 earnings
Dec 8-9
Fed FOMC meeting

What will decide how SOXL does from here is whether the chip sector can break out of the wild swings since October and turn into a one-way uptrend. NVIDIA's Q3 earnings on November 17 look like the first gate.

It'll be the first report card since CEO Jensen Huang resigned, with Colette Kress in as interim CEO.

On top of that, the October inflation data came in hot. Consumer prices rose 0.4% from the previous month, and producer prices were up 6.5% from a year earlier, the highest since November 2022.

The 10-year Treasury yield has climbed to 5.15%, and the Fed's FOMC meeting on December 8 to 9 is waiting too. 3x compounding only swings back in your favor once rates and volatility calm down and a one-way trend opens up.

Bottom line: leveraged ETFs are definitely risky products. Unlike unleveraged stocks or 1x ETFs, losses pile up in choppy, sideways markets, and in a downturn the negative compounding builds so fast that recovering your principal gets very hard.

If you're interested in leveraged ETFs, maybe it makes sense to trade them only when you expect a strong uptrend (or a strong downtrend, if you trade inverse ETFs).

This isn't investment advice. You know that, right? Let's each protect our own precious money.

That's all for this post. See you next time.

Related posts

No comments:

No comments yet. Want to leave the first one?