Leveraged and inverse ETFs promise a multiple of an index's return for one day only, and the daily reset makes their longer term results depend on the path of prices, not just the destination. The ProShares UltraPro QQQ (TQQQ), which targets three times the daily return of the Nasdaq-100, returned 54.25% in the first half of 2026 and 96.70% over the year to June 30, 2026 at net asset value, while its mirror image, the ProShares UltraPro Short QQQ (SQQQ), lost 45.71% and 60.59% over the same periods [1][2]. Those figures are not three times the index and are not each other's opposite, and the reason is arithmetic: gains and losses compound on a base that changes every day. DataPorium price data for the iShares Semiconductor ETF (SOXX) shows the same effect in 2026, with a hypothetical three times daily strategy gaining 654% during a 111% rally and then losing 69% in a 29% decline [4].
How the daily reset works
A three times fund holds swaps and futures sized to deliver three times the index's move for the day. At the close it rebalances so that the next day again starts at three times exposure. ProShares states that for any holding period other than a day the return may be higher or lower than the daily target and that the differences may be significant; smaller index moves and higher volatility push results below the target, while larger trending moves and lower volatility push them above it [1][2].
A two day example
Suppose an index goes from 100 to 110 on day one and back to 100 on day two. The index is flat. A three times fund gains 30% on day one, to 130, and then loses three times the 9.09% decline on day two, or 27.3%, to end at 94.5, a loss of 5.5% with the index unchanged. A minus three times fund falls 30% to 70, then rises 27.3% to 89.1, a loss of 10.9%. FINRA's Regulatory Notice 09-31 gives the low volatility version: an index that moves from 100 to 101 and back costs an inverse fund only 0.02%, while a move to 110 and back costs 1.82% [3]. The larger the daily swings, the larger the drag, and the drag hits both the long and the short product.
FINRA's notice also documents real cases from the five months ending April 30, 2009: a two times oil and gas ETF fell 6% while its index gained 2%, and a three times financial sector ETF fell 53% while its index rose about 8% [3]. Those episodes led regulators to describe the products as unsuitable for buy and hold investors who do not monitor them daily [3].
2026 results: TQQQ and SQQQ versus the math
| Fund (as of June 30, 2026, NAV) | YTD | 1 year | 3 year (ann.) | 5 year (ann.) | 10 year (ann.) | Net expense ratio |
|---|---|---|---|---|---|---|
| ProShares UltraPro QQQ (TQQQ), 3x | 54.25% | 96.70% | 59.94% | 22.72% | 45.50% | 0.82% [1] |
| ProShares UltraPro Short QQQ (SQQQ), -3x | -45.71% | -60.59% | n/a | n/a | n/a | 0.95% [2] |
Two features stand out. First, the funds are not mirror images: TQQQ's 96.70% one year gain sits beside a 60.59% loss for SQQQ, because a short fund cannot lose more than 100% and its losses shrink as its base shrinks [1][2]. Second, the five year figure for TQQQ, 22.72% annualized, is far below the ten year figure of 45.50%, because the five year window includes the 2022 bear market, when daily rebalancing into a falling and volatile index destroyed capital that the later rally only partly rebuilt [1]. The underlying Nasdaq-100 traded at 36.90 times earnings with a 0.55% dividend yield on June 30, 2026, so a three times product on it was carrying triple exposure to a highly valued index [1]. Both funds also carry gross expense ratios near 1% (0.97% for TQQQ, 0.99% for SQQQ) before waivers, plus the financing cost embedded in the swaps [1][2].
What a three times strategy on SOXX would have done in 2026
DataPorium's daily prices for SOXX, one of the most volatile large ETFs of 2026, allow a clean illustration of path dependence. Applying a three times daily reset to SOXX's actual closes, with no fees, gives the following results [4]:
- March 30 to June 22, 2026 (SOXX up 111.4%): a 3x daily strategy would have gained about 654%, almost double the 334% that three times the period return implies, because a steady trend with low day to day reversals compounds in the holder's favor; a -3x strategy would have lost 93.4% [4].
- June 22 to July 29, 2026 (SOXX down 29.0%): the 3x strategy would have lost 69.0%, less than the 87% implied by three times the decline, while -3x would have gained 117.8% [4].
- June 30 to August 26, 2026 (SOXX down 19.6%): 3x would have lost 54.9% and -3x would have gained only 44.8%, not 58.7%, because the decline was choppy [4].
- December 31, 2025 to August 26, 2026 (SOXX up 71.1%): 3x would have gained 199.5% versus 213.4% implied, and -3x would have lost 92.9% [4].
The pattern is consistent with the FINRA examples: trending markets reward leverage beyond the stated multiple, and volatile or sideways markets punish it, in both directions [3][4]. Readers can reproduce these calculations from the daily closes on DataPorium's ETF page.
Who uses leveraged and inverse ETFs, and the cost of holding them
These products were built for traders who want a day's exposure without margin or options, and for hedgers who need a short position for a specific event. For that use the daily reset is a feature, since exposure is known at every open. For a multi month holder the same reset is a tax on volatility that compounds alongside a fee near 1% a year [1][2]. The fair case for long holding is a strong, low volatility uptrend like the one from March to June 2026, in which the products delivered more than three times the index; the problem is that no one knows in advance which stretch of the market will look like that, and the same fund gave back 69% in the following five weeks [4]. Investors may consider position sizes they could see fall by half in a month, and a plain index ETF for the part of the portfolio meant to be held for years.
Daily reset leveraged ETFs multiply a day's return, and over longer periods the path of the market, not the leverage number, decides the result.
Key takeaways
- TQQQ returned 54.25% and SQQQ lost 45.71% in the first half of 2026; over one year the figures were 96.70% and -60.59% at NAV [1][2].
- Daily rebalancing means volatility erodes both long and short products; a 10% up and down move costs a 3x fund 5.5% and a -3x fund 10.9% with the index flat.
- FINRA documented a 3x financial ETF falling 53% while its index rose about 8% in 2008 and 2009 [3].
- A hypothetical 3x on SOXX gained 654% in the March to June 2026 rally and lost 69% in the following five weeks [4].
- Net expense ratios are 0.82% for TQQQ and 0.95% for SQQQ, before swap financing costs [1][2].
Frequently asked questions
Why does TQQQ not return exactly three times the Nasdaq-100?
Because it targets three times the daily return and rebalances each day, so over longer periods the result depends on the sequence of daily moves; ProShares states that returns for holding periods longer than one day may differ significantly from the daily target [1].
Can you hold SQQQ long term?
The product can be held, but its structure works against long holders: SQQQ lost 45.71% in the first half of 2026 and 60.59% over the year to June 30, 2026, and volatility drag reduces it even in flat markets [2][3].
What is volatility decay in leveraged ETFs?
It is the loss caused by daily rebalancing when prices move up and down; FINRA's example shows an inverse fund losing 0.02% when an index moves from 100 to 101 and back, but 1.82% when it moves to 110 and back [3].
What do TQQQ and SQQQ cost?
TQQQ has a 0.97% gross and 0.82% net expense ratio; SQQQ has 0.99% gross and 0.95% net, according to ProShares [1][2].
