AutoHedging

Research · Fund study · NASDAQ: QPUX

Why QPUX lost 94% of its value, and what it would take to come back

Written 17 November 2026. Prices are closing prices to 16 November 2026, adjusted for the fund's 1-for-3 reverse split in March 2026.

Two things happened, and they multiplied. The quantum computing stocks the fund follows fell by about half. The fund is built to deliver twice their move each day. Over fifteen months that turned a fall of about half into a loss of 94%, and the same daily reset is what makes the way back so long. Company results explain little of it: the two largest holdings raised their revenue outlook this month.

$3.59QPUX close, 16 November 2026
-94%since its first trading day, 7 Aug 2025
-98.5%from its highest close, $242.92 on 14 Oct 2025
$13.9Mnet assets on 16 November

What the fund is

QPUX is the Defiance 2X Daily Long Pure Quantum ETF. Its stated aim is a return of two times the performance of a group of “pure quantum” companies for a single day. It holds no shares. It keeps its money in Treasury bills and gets its exposure through swap contracts with banks, reset every day.

The ten largest lines the fund listed on 17 November 2026, as a share of net assets. There were 29 lines in all.
HoldingHow it is heldShare of net assets
U.S. Treasury bill, due 19 November 2026Owned outright79.4%
QuantinuumTwo swaps24.2%
IonQTwo swaps22.0%
Rigetti ComputingSwap12.6%
Horizon QuantumSwap12.2%
D-Wave QuantumSwap11.4%
IQM Quantum ComputersSwap10.3%
Quantum Computing Inc.Swap8.6%

The fund names eleven companies in its target group and says it weights them roughly equally: IonQ, D-Wave, Quantinuum, Infleqtion, Arqit, Rigetti, Quantum Computing Inc., IQM, Xanadu, BTQ and Horizon Quantum. Its own page warns that “for periods longer than a single day, the Fund will lose money if the Target Portfolio's performance is flat.” (Defiance, QPUX page)

Why it fell: the stocks, then the daily double

To separate the two causes we rebuilt the fund's arithmetic from its first day. We took the four holdings that traded for the fund's whole life (IonQ, D-Wave, Rigetti and Quantum Computing Inc.), weighted them equally, and worked out two lines: what the four did, and what twice their move each day adds up to.

5 10 25 50 100 200 400 Sep Oct Nov Dec Jan '26 Feb Mar Apr May Jun Jul Aug Sep Oct Nov Four stocks, equal weightThe same at twice the daily moveQPUX 7 August 2025 = 100. The scale is a ratio scale: each step up is a doubling or more.
Source: Yahoo Finance daily closes; the basket and the daily double are our own arithmetic. The basket is rebalanced to equal weight every day. It is a stand-in for the fund's group, which changed over the period.
The same three lines as numbers
PeriodFour stocks, equal weightTwice their move each dayQPUX itself
Since the first day, 7 Aug 2025-46%-90%-94%
Since QPUX's highest close, 14 Oct 2025-76%-97.7%-98.5%

What the table says, in three steps.

  1. The stocks fell 46%. That is the starting point, and it would have hurt any holder.
  2. Twice the daily move turned that into 90%. The fund compounds its doubled move day after day. It rose more than four times between August and October 2025, then lost twice the stocks' fall each day on the way down. Resetting daily also costs something every time prices swing. When a holding drops 10% and then rises 10%, it ends 1% down. At twice the size it drops 20% and rises 20%, and ends 4% down. These stocks swung like that for fifteen months, at about 93% a year, several times an ordinary stock index.
  3. The last few points are costs and differences. QPUX fell 94%, a little more than the daily double of our four. Fund expenses of 1.29% a year, the cost of the swaps, and the fact that the fund's real group included other, newer names account for the gap. We cannot split those three apart.

None of this is a fault in the fund. It did what its prospectus says it does. The result over fifteen months is what a daily double produces when the thing it follows falls by half and swings violently on the way.

The stocks it follows

Each company in the fund's target group: its highest close in the last two years, its close on 16 November 2026, and the rise needed to get back to that high.
CompanyTickerHighest close16 Nov 2026From the highRise needed to regain it
IonQIONQ$82.09 on 13 Oct 2025$26.45-68%+210%
QuantinuumQNT$83.17 on 6 Jul 2026$31.75-62%+162%
Rigetti ComputingRGTI$56.34 on 15 Oct 2025$9.06-84%+522%
Horizon QuantumHQ$38.39 on 22 Jun 2026$11.00-71%+249%
D-Wave QuantumQBTS$44.78 on 15 Oct 2025$8.67-81%+416%
IQM Quantum ComputersIQMX$13.91 on 6 Jul 2026$6.42-54%+117%
Quantum Computing Inc.QUBT$25.68 on 18 Dec 2024$5.25-80%+389%
InfleqtionINFQ$19.87 on 2 Jun 2026$7.90-60%+152%
Arqit QuantumARQQ$58.27 on 13 Oct 2025$13.45-77%+333%
XanaduXNDU$36.12 on 1 May 2026$2.03-94%+1,679%
BTQ TechnologiesBTQ$14.14 on 8 Oct 2025$1.31-91%+979%

Most of the older names peaked in the same week of October 2025. The companies that listed in 2026 (Quantinuum, Horizon Quantum, Infleqtion, Xanadu) peaked in May to July 2026 and have fallen by 60% or more since. Every company in the group is down by more than half from its high.

Source: Yahoo Finance daily closes. Highs are closing highs within the price history available to us (two years, or since listing).

Revenue against outlook: what the companies report

We read the latest quarterly release of the four largest holdings. These cover the quarter to 30 September 2026 and came out between 5 and 10 November.

Third quarter of 2026. USD millions.
CompanyRevenueOn a year earlierLoss before interest, tax and non-cash itemsCash and investmentsWhat it said about the year
IonQ$173.8+336%$133.6$1,700, plus $2,000 raised in OctoberRaised its 2026 revenue outlook to $505–515. Orders not yet delivered more than doubled to $1,200.
Quantinuum$9.1+233%$80.5$1,940Raised its 2026 revenue outlook to $31–34, from $28–32.
D-Wave$4.5+21%$37.7$503No revenue outlook given. Nine-month revenue was $10.4, against $21.8 a year earlier. Orders not yet delivered were $38.3.
Rigetti$1.4-28%Operating loss $33.2$497Kept its 2026 revenue outlook of about $20. That needs about $9 in the fourth quarter and depends on one customer accepting a system.

What we take from it.

  • Outlooks went up at the two largest holdings while their shares went down. IonQ and Quantinuum both raised revenue guidance this month. Their share prices are 68% and 62% below their highs. So the fall is not explained by these companies missing what they promised.
  • IonQ's growth is partly bought. $83.5 million of its $173.8 million came from SkyWater, a chip foundry it acquired on 31 July. The business it had before grew 127%.
  • Revenue is small next to the losses. Quantinuum lost about nine dollars for each dollar of revenue in the quarter, D-Wave about eight. These are companies funded by cash raised from shareholders, and all four have more than a year of it.
  • Two results were lumpy. D-Wave and Rigetti sell a few systems a year, so a quarter with no system sale looks weak. Rigetti's full-year figure rests on one delivery.
  • The prices were set by expectations far ahead of the revenue. When share prices of companies with revenue this small fall by two thirds, it is mostly a change in what investors will pay for distant profits. News reports this month tied the latest leg down to the Rigetti and D-Wave results and to a broad fall in technology shares. We have not verified a single cause.

Sources: IonQ release · Quantinuum release (SEC) · D-Wave release (SEC). Rigetti's figures are from a published report of its release, not from the release itself. The loss column is the companies' own “adjusted EBITDA”, which leaves out share-based pay and some other costs. We did not read the results of the other seven companies in the group.

Can it come back?

There are two different questions here, and they have different answers.

Can the stocks recover?

They can. Revenue is growing at the larger companies, order books are bigger than a year ago, and all four that we read hold enough cash to keep operating. Whether investors will again pay the prices of October 2025 for them is a judgement about the future that no filing settles, and we do not make it here.

Can QPUX get back to where it was?

That is arithmetic, and the arithmetic is hard. From $3.59, the fund needs to rise 16 times to reach its first-day price of $57.29, and 68 times to reach its high. A daily double does not retrace a fall when the stocks retrace theirs, because the reset that cost so much on the way down also costs on the way up whenever the path is rough.

What a fund aiming for twice the daily move would return over one year, for a given move in its stocks and a given roughness of the path. An illustration from the standard formula, before fees. Not a forecast.
The stocks over a yearCalmer path (swings of 70% a year)As rough as the past 15 months (93%)Rougher (120%)
Flat-39%-58%-76%
Up 50%+38%-6%-47%
Doubles+145%+68%-5%
Triples5.5 times3.8 times+113%
Back to the old highs (up about 384%)14.4 times9.9 times5.6 times
  • If the stocks go nowhere, the fund still loses. At the roughness of the past fifteen months, a flat year for the stocks costs the fund about 58%.
  • If the stocks double, the fund does not quadruple. Depending on how rough the path is, it ends anywhere from slightly down to up about one and a half times.
  • Even a full recovery in the stocks leaves the fund far below its old prices. If the four older holdings all returned to their highs within a year, a rise of about 384% on average, the same formula puts the fund near 10 times today's price, about $35. Its first-day price was $57.29 and its high was $242.92.

Two more facts to weigh. The fund has $13.9 million in assets. Fund companies often close funds of that size; when a fund closes, holders are paid the net asset value on that day, and the chance of a later recovery ends with it. And the fund has already consolidated its shares once, one for three in March 2026, which raises the share price and changes nothing else.

Our reading. QPUX fell for a reason that is built into it. It can rise sharply over short stretches if these stocks rally, and it has no path back to its earlier prices that does not require the stocks to rise many times over, quickly and smoothly. The fund's own documents describe it as a tool for a single day. Nothing here says what the stocks will do.

Sources and limits

  • Fund facts, holdings and risk wording: Defiance's QPUX page, read 17 November 2026. The page shows the ten largest of 29 lines, so the weights above are incomplete.
  • Prices: Yahoo Finance daily closes to 16 November 2026, adjusted for the reverse split.
  • Our basket uses four stocks because only those four traded for the fund's whole life. The fund's actual group changed as new companies listed, so our “twice the daily move” line is an estimate of the fund's mechanics, not a copy of its portfolio.
  • The one-year table uses the usual approximation for a fund reset daily: the square of the stocks' growth, reduced by the square of their yearly swing. It ignores fees, swap costs and gaps between days, all of which lower the result.
  • Company figures are from each company's release where linked. They are as reported and unaudited.