Here is the key question that Friday’s big stock-market selloff raises.

Was Friday’s selling that chopped the Dow Jones Industrial Average and the Nasdaq Composite Index by more than 800 points each a market top or just another one-off?

The short answer is: It’s not clear yet.

But we will know soon enough. Futures trading late Sunday will offer the first signal of what’s to come. The early indicators so far are bullish, with the Dow, Standard & Poor’s 500 Index and Nasdaq-100 showing decent opens.

The reason we don’t know yet if we’re dealing with a market top is that a top takes time to reveal itself.

Often a big runup is followed by a pullback and then resumes. A top slowly gives up its gains, unlike market bottoms which hit a low, bounce up some distance, fall a second time but, critically, don’t drop below that first low.

You could see how stocks bottomed after the April tariff tantrum, when the Trump Administration hurriedly told investors the first proposed tariffs were just that: proposals.

And that realization prompted investors to jump right back into stocks. The market jumped sharply and continued to rise, basically until early in October.

The Standard & Poor’s 500 Index, as an example, bottomed on April 7 at 4,835 and is up 35.5% from that point. It’s up 11.4% for the year, which reminds us that the April slump was just plain nasty.

The catalyst for Friday’s drubbing was President Trump’s criticism of China’s moves to make rare earth metals more difficult to obtain and threatening U.S. companies.

Later, the president imposed 100% tariffs on imports from China, and stocks took an abrupt dive that led to their biggest one-day losses since this past spring.

Technology shares overall fell nearly 4%. Consumer Discretionary stocks were off 3.3%, and only one S&P 500 sector was ahead on the day: Consumer staples. Think Walmart , one of just three Dow stocks that closed with gains on Friday.

The others were McDonalds and Coca-Cola. At the same time, high-flying stocks slumped. Coinbase, the giant crypto marketplace, dropped 7.8% on the day. Tesla was off 5.1% on the day and is off 7.1% for October. (It is up 2.4% on the year.)

The start of a top?

Which leads to us trying to understand tops.

Market tops have some common elements:

• A stock or an index goes on a pretty good run.
• But the size of the daily gains start to shrink.
• Trading volumes start to decline.

In fact, tops kind of sneak up on you. The 2007-2008 crash was derived from over-heated speculation in real estate, especially residential real estate, with a financial system that seemed clueless about the risks that came with making loans with no particular care.

The Oct. 19, 1987, market selloff that saw the Dow fall 22% IN A DAY was the product of something else: poorly understood structural problems in the financial sector complicated by a depression that hit the energy industry. But the seriousness of the issues didn’t receive their due attention until that summer.

The Dow was up nearly 44% on the year at its 1987 peak on Aug. 25, 1987. On Oct. 18, the gain had been cut down to 18.5%. The Oct. 19 selloff wiped out the gains for the year. The S&P 500’s decline was 20.5%; the Nasdaq was off just 11.4%.

Amazingly, the Dow and S&P 500 ended 1987 with small gains. The Nasdaq’s loss was 5.4%.

The market top to beat all market tops

You can see a market top pretty clearly with the Nasdaq Composite Index in March 2000. In 1998, the index jumped 39.6% as money poured into Techland as the Internet emerged as a transformational event. In 1999, however, a huge rally turned into the dot-com bubble, and the index jumped . . . 85.6% to 4,069.

So, of course, 2000 was going to be great. And it seemed, at first, to be great. The Nasdaq was up 24% by March 9. But the gains were starting to diminish. The volume was waning. And on March 10, the index closed up just 2 points to 5,049.

On Monday, March 13, the Nasdaq fell 141 points. The next day saw the index fall 200 points. On the third day, the loss was 124 points. From the March 10 top until March 2009, the Nasdaq fell 74.9%.

The index would not close above 5,000 again until 2015.

The worries about bubbles now

There are a lot of people on Wall Street who are wildly optimistic about stocks. Big tech stocks. Multi-trillion-dollar stocks. They see billions of dollars flowing into artificial intelligence investments for years and years to come, product demand and profits practically guaranteed to boot.

The market value of the 68 stocks in the S&P 500 technology sector, plus Tesla, Google-parent Alphabet and Facebook-parent Meta Platforms represent about 45% of the market value of the entire S&P 500 Index, according to Standard & Poor’s.

Related: One move just sent AMD into a whole new league

That suggests a floor under the S&P 500 Index (and, of course, the Nasdaq and Nasdaq-100 indexes).

But the bubble worriers are worth listening to. So, to give them their due, look to what happened Wednesday and Thursday ahead of Friday’s selloff.

• The S&P 500 index was pushed to a record close of 6,753 on Wednesday, its eighth gain in nine sessions.
• It hit a 52-week high of 6,754.58 on Thursday but closed at 6,735.11, an 18-point drop.
• Gains have shrinking since the S&P 500 rose 1.5% on Aug. 22. No daily gains in September and so far in October have been larger than 55 points, or 0.9%, reached on Sept. 11.

Most analysts have been waiting for third-quarter earnings reports to come out in volume. That will start this week when JPMorgan Chase reports on Tuesday.

But worries are growing. “Investors have a hard time in the middle of this excitement distinguishing between the good ideas and the bad ideas,” Jeff Bezos, Amazon’s founder, said last week. The Bank of England warned of the possibility of a sharp correction.

At the same time, there have been large, surprise bankruptcies:

• Tricolor Holdings, a nationwide operator of 65 used-car dealerships, collapsed in September after its lender reported alleged fraudulent activity.
• First Brands Group, a Cleveland-based auto-parts manufacturer, sought bankruptcy on Sept. 29 because its $10 billion in financing was falling apart. Worse, there was $2 billion that was missing.

But the explosive growth in artificial intelligence  — huge investments in data centers, semiconductor manufacturing and growing debt loads  — are causing deep concerns.

If there’s nothing to stop the AI boom, the market top will be deferred until next year or maybe later. It depends on the earnings, when the government shutdown ends, and, probably, whether China and the United States can negotiate a workable trade deal. The key issue is the terms under which U.S. can buy rare earths ores. Talks have been going on for months, but there still is no deal. Just promises of a framework.

If cooler heads prevail, maybe the optimism that has prevailed on Wall Street since April and up until this past week will reignite. Many firms have been boosting their targets for the end of 2025 and beyond.

So, was Friday’s drubbing a market top? Maybe.

Related: How often does the S&P 500 finish October higher?