Ask a room of swing traders whether QQQ is riskier than SPY overnight and almost every hand goes up. Ask them by how much, and the room goes quiet. That gap — between a confident intuition and an actual number — is where a lot of position sizing goes wrong.
The SPY vs QQQ overnight question deserves a real answer, so this post puts the two side by side on the terms that matter to someone holding from the close to the open. Not ten-year returns. Not expense ratios. Overnight behavior: what can move the price while you’re asleep, how often, and how concentrated that risk is.
The Volatility Gap Is Smaller Than You Think
Start with the number everyone cites. Measured from April 1999 through early August 2026, QQQ’s annualized standard deviation of returns runs roughly 27.0% against SPY’s 19.2%. On more recent windows the spread narrows: one comparison puts trailing annualized volatility at about 20.3% for QQQ and 15.2% for SPY. Either way, QQQ moves more. That part of the folklore holds.
But Nasdaq’s own index research complicates the story in a useful way. Comparing one-year rolling volatility of the Nasdaq-100 against the S&P 500 from December 31, 2007 through December 31, 2025, the Nasdaq-100 averaged only about 2.9 percentage points higher. And the correlation of daily returns between the two indexes over that period sits near 93%.
Ninety-three percent correlation is the number worth sitting with. On the overwhelming majority of sessions, these two instruments are telling you the same story with slightly different emphasis. If your overnight thesis is “the market gaps up tomorrow,” you are not meaningfully diversifying by expressing it in QQQ instead of SPY. You are mostly just adding leverage to the same idea.
SPY vs QQQ Overnight, Side by Side
| Dimension | SPY (S&P 500) | QQQ (Nasdaq-100) |
|---|---|---|
| Holdings | ~500 names, all sectors | ~100 non-financial names |
| Top 10 weight | Roughly one-third of the fund | About 47.3% of the fund |
| Largest single position | Well under 8% | NVIDIA near 8.9% |
| Long-run std. deviation | ~19.2% (since Apr 1999) | ~27.0% (since Apr 1999) |
| Max historical drawdown | About -55.2% | About -83.0% |
| Implied vol gauge | VIX | VXN |
| Dominant overnight catalyst | Macro prints, Fed, geopolitics | Single-company earnings and guidance |
| Risk shape | Diffuse — many small contributors | Lumpy — few large contributors |
The last two rows are the ones that should change how you trade, and they’re the ones nobody puts in a comparison table.
Concentration Is the Real Variable
QQQ’s top ten positions represent roughly 47.3% of the fund, with the top five alone near 30%. NVIDIA sits around 8.9%, Apple near 7.2%, Microsoft near 5.0%. In practical terms: on any given evening, a single earnings release can move close to a tenth of your position’s underlying value before the market reopens.
SPY has concentration too — that’s the honest caveat, since the same megacaps sit at the top of the S&P 500. But SPY dilutes each one across roughly 500 names and eleven sectors. A disappointing quarter from one semiconductor company is partially offset by 490-odd other holdings that didn’t report that night.
This produces two genuinely different risk calendars:
- SPY’s overnight risk is spread thin and macro-shaped. CPI, payrolls, FOMC, an overnight geopolitical headline. Roughly a dozen high-impact dates per quarter, all of them known in advance and all of them on a public calendar.
- QQQ’s overnight risk is lumpy and idiosyncratic. Add eight or ten megacap earnings dates on top of the macro calendar — and those releases land after the 4:00 p.m. close, precisely inside the window you’re holding through.
If you don’t already know what actually sets the opening price while you’re asleep, that distinction won’t feel real. It becomes very real the first time you hold QQQ through a guidance cut you didn’t have on your calendar.
The VXN/VIX Ratio: A Regime Gauge, Not a Signal
There’s a cheap way to check whether the market agrees with you about which is riskier right now: compare VXN (30-day implied volatility on the Nasdaq-100) to VIX (the same for the S&P 500).
A ratio near 1.1 or 1.2 is ordinary — tech is expected to be a bit choppier. On June 16, 2026, that ratio reached 1.64, the widest reading since July 3, 2017. By July 9, VXN closed at 26.91 while VIX traded near 15 — an 11.8-point spread. Early August 2026 has VIX back in the mid-teens.
A stretched VXN/VIX ratio doesn’t tell you tech will fall. It tells you the options market is pricing a much wider distribution of tech outcomes than broad-market outcomes — which is exactly the input an overnight position sizer needs.
Used as a regime gauge, it’s straightforward: when the ratio is elevated, the same dollar position in QQQ carries materially more implied overnight range than the same dollar position in SPY. That should shrink the QQQ size, not the conviction. For context on how badly this can go when sizing is treated as an afterthought, see our breakdown of why stop-losses fail overnight.
Which One Fits Your Overnight Process?
SPY tends to suit you if…
- Your edge is reading macro conditions and broad positioning rather than individual company fundamentals.
- You want an overnight risk calendar you can fully enumerate a month ahead.
- You size positions by fixed dollar risk and want fewer surprises inside that risk budget.
- You’re still building the discipline to sit through a red pre-market print.
QQQ tends to suit you if…
- You actively track megacap earnings dates and are willing to flatten or halve size ahead of them.
- You want more overnight range per unit of capital and have deliberately reduced share count to compensate.
- You have a view on the tech complex specifically, not just on equities broadly.
Notice that both lists are about process, not preference. The instrument you choose should follow from the risk you can actually monitor, not from which ticker has the better ten-year chart. QQQ has compounded faster over the last decade — that’s a holding-period argument, and it says almost nothing about whether you should carry it through tonight’s close.
SpyTomorrow’s nightly pressure readings are built on SPY for exactly this reason: a diffuse, macro-driven instrument is the one where a structural read on overnight conditions is most likely to mean something. They’re decision support for sizing and pre-commitment, not a signal to follow blindly — and we publish the misses alongside the hits in our audit of the last 24 readings so you can judge for yourself.
Three Questions Before You Switch
- Can you name every megacap earnings date in the next three weeks? If not, you’re not equipped to carry QQQ overnight — you’re carrying unmeasured event risk.
- Did you adjust share count, or just swap the ticker? Moving a SPY-sized position into QQQ at the same dollar amount silently increases your overnight risk. Scale by expected range, not by capital.
- Is your thesis about tech, or about the market? If it’s about the market, the 93% correlation means QQQ mostly adds noise and drawdown depth without adding information.
The honest conclusion is less dramatic than the framing usually suggests. These are not two different strategies. They’re the same strategy at two different intensities, with one of them carrying a lumpier calendar of things that can happen after the bell. Choose the intensity you can size correctly and the calendar you can actually track — and then stop relitigating the choice every week.
This article is for educational and informational purposes only. It does not constitute financial advice, investment recommendations, or a solicitation to buy or sell any securities. Past performance and historical data do not guarantee future results. All trading involves risk, including the potential loss of principal. Always do your own research and consult with a qualified financial advisor before making trading decisions.
