Buy the Dip After a Market Crash? What 81 Big Nasdaq Drops Show

Why test it

"Buy the dip" is one of the most repeated ideas in trading, and on an index that went up about tenfold over our sample, it sounds obviously right. But "dip" almost never gets defined, and the claim almost never gets tested against a time horizon and a benchmark. We locked in all three before looking at any results:

• A big down week is a week where QQQ's total return was −5% or worse.
• The horizons are the next week and the next four weeks.
• The benchmark is every week in the sample, plus random sets of weeks of the same size.

What we found

Of the 1,400 weeks between November 1999 and September 2026, 81 were down 5% or more, or 5.8% of all weeks.

• A week down 5% or more (81 weeks)+0.72%
• Any week (1,400 weeks)+0.23%

Exact values from table 2 in this article. Bar lengths share a zero baseline; no new calculation or simulated price path. Read the table and its stated sample and limits in this article.

On average, the next week did bounce harder: +0.72% against +0.23%. When we drew 81 weeks at random 10,000 times, about 9% of the draws did as well, so the bounce is suggestive, not conclusive.

But it wasn't more likely to be an up week. 55.6% of those next weeks went up, compared with 56.0% normally. The higher average came from a handful of big rebounds, not from more of them.

It was also a much bumpier week. The standard deviation of next-week returns after a big down week was 5.58%, against 3.36% normally. The best next week gained 17.9%. The worst lost 13.4%.

Four weeks later, the edge was gone. The average four-week return was +0.87%, basically the same as the +0.86% of any four weeks, and the share of positive four-week stretches was actually lower: 58.0% against 62.6%.

Why the average went up but the odds didn't

A higher average with the same share of up weeks means the shape of the distribution changed, not the direction. After a big drop, both tails get fatter: some weeks rebound hard, others keep the decline going. The rebounds were big enough to lift the average, but they didn't happen any more often than usual.

As a trader, that matters more than the average itself. A strategy that lives off a few big rebounds has to survive the big extensions in between, so its results depend a lot on position size and on whether you can sit through a −13% week, the worst case in our sample.

Dip or crash? The trend made the difference

Next, we split the big down weeks by where the index stood compared with its 40-week moving average, a common stand-in for the long-term trend. Five of the 81 weeks came too early in the data to have a 40-week average, so the table below covers 76.

The few sharp drops inside an uptrend recovered well. The many inside a downtrend didn't. Most big down weeks came in clusters during bear markets: 41 of the 81 fell in 2000, 2001 and 2002 alone, and 8 more in 2008.

Fourteen cases is a small sample, but the direction of the result makes sense. Buying a sharp drop is a bet that the trend is still intact, and that bet fails once the trend has turned.

That's the practical line between a dip and a crash. A dip is a sharp fall in a market that's still above its long-term average. A crash is a run of falls that's already pushed it below. During the 2000–2002 and 2008 bear markets, every big down week looked like a buying opportunity at the time, but buying them earned nothing on average over the next four weeks. A 40-week average won't catch the exact turn, but it would have kept a dip buyer out of most of those weeks.

What this means if you trade NQ

• A big down week isn't a buy signal on its own. It lifted the next week's average but not the odds, and it nearly doubled the volatility.
• Context decides. Inside an uptrend, sharp drops tended to recover. Inside a downtrend, they were usually followed by more of the same.
• Size for the volatility. If you do trade a bounce, remember the week after a big drop has historically been way more volatile than normal, so the same position carries a lot more risk.
• Be clear about the horizon. A one-week bounce and a four-week recovery are different claims, and most "buy the dip" advice never says which one it means.

How this connects to the price-structure test

In another test, we found that Nasdaq-100 weeks closing below the prior week's low were followed by above-average weeks. That result and this one point the same way: on this index, over this period, short-term weakness was more often followed by a bounce than by more downside, but only in the short run, and with more risk.

If you want to test a dip rule yourself

Write the rule down before you look at a single chart. A complete rule has five parts:

1. The trigger, for example a week that closes down 5% or more.
2. The filter, for example only when the index is above its 40-week average.
3. The entry, for example the next week's open.
4. The exit: a fixed horizon, a price level, or both.
5. The benchmark: what the same money would have made in random weeks or by just holding.

Then count every case, including the ones that hurt. A rule that only works once you take out the bad years doesn't work.

Limits

• Weekly data. A "dip" on a daily or intraday chart is a different event, and this test doesn't cover it.
• Overlapping returns. Four-week returns overlap from one week to the next, which overstates how independent the data points are.
• One index, one period. The Nasdaq-100 eventually recovered from every decline in the sample. That's not guaranteed.
• Price-only four-week returns. The four-week numbers leave out dividends, which barely matters at QQQ's yield.

How we did it

• QQQ weekly prices from Alpha Vantage (TIME_SERIES_WEEKLY), adjusted by us for the two-for-one split of March 20, 2000; weekly returns include dividends on their ex-dates (Alpha Vantage DIVIDENDS).
• Big down week: a total return of −5% or worse. Trend: close above or below the 40-week simple moving average.
• Random comparison: 10,000 draws of 81 weeks, with a fixed seed.

Sources

• Alpha Vantage. TIME_SERIES_WEEKLY and DIVIDENDS (QQQ). Data retrieved October 2026.

FAQ

Does buying the dip work on the Nasdaq? After weeks down 5% or more since 1999, the next week averaged +0.72%, but the four-week result was no better than average.

What counts as a dip? There's no standard definition. We used a weekly drop of 5% or more. Whatever definition you use, lock it in before you test it.

Should you buy the dip during a stock market crash? Not based on this. Big down weeks with the Nasdaq-100 below its 40-week average, like in 2000–2002 and 2008, were followed by −0.11% on average over four weeks, and only half were higher.

Is buying the dip better in an uptrend? In our data, yes: drops above the 40-week average were followed by +2.93% on average over four weeks, against −0.11% below it, on small samples.

How often does the Nasdaq-100 fall 5% in a week? 81 times in 1,400 weeks, or about 6% of weeks, mostly bunched up in 2000–2002 and 2008.

Is the one-week bounce worth trading after costs? Maybe, but the average rests on a few big rebounds and comes with almost double the usual volatility. Costs are small for one round trip in index futures. The real risk is a −13% week like the worst one in our sample.

---

Research and education only. Not financial advice. Past results do not predict future returns. A backtest is not real performance.

How this article was made: drafted with AI assistance from public data and the sources listed above; facts and figures checked by Gabriel Décarie and Karel Panneton.