Anatomy of an Earnings Gap: NVIDIA, May 2023

On May 24, 2023, NVIDIA told the market its next quarter would be far larger than Wall Street had modeled. The next morning, the stock didn't climb step by step toward a new price. It opened there, roughly a quarter above the prior close, reportedly adding hundreds of billions in market value before most regular-session traders had touched a key.

That blank space is the lesson. On the regular-session daily chart, trading never printed the prices in between — extended-hours trading did occur overnight, but the daily chart shows a tear in the tape: yesterday's value on one side, a new valuation on the other. That zone is a gap, and this case study uses it, in hindsight, to teach how students of charts think about them.

The Setup (What You Could See at the Time)

• A closed market, new information. Earnings landed after hours, when the regular market couldn't react. Overnight, buyers and sellers worked out where the stock should open the next morning.
• The open as a repricing. Through overnight trading and the opening auction, the market settled on a much higher starting price than the prior close. On the daily chart, price teleported; it didn't travel.
• Enormous volume on the gap day. The repricing day traded several times its normal volume — a genuine crowd at the new level — the confirmation signature from the Volume guide.

The Concept

A gap is the market repricing faster than continuous trading allows. The chart vocabulary:

• Gap-and-go: price never returns to the empty zone — the new information was that decisive. In hindsight, this NVIDIA gap was one: the zone never filled, and price kept climbing for months.
• Gap-fill: price later trades back through the empty zone. This happens constantly — most ordinary gaps eventually fill, which is exactly why "gaps always fill" became folk wisdom.

The honest teaching point: both outcomes are common, and which one you get depends on the quality of the news, not the shape of the gap. A guidance rewrite that resets a company's earnings power gaps differently than a routine beat.

What Happened (In Hindsight)

The gap never filled. Instead, the empty zone became the visible border between the pre-AI-repricing NVIDIA and the one the market began valuing after that guidance shock. Pullbacks in the following months stayed above the old range, and the stock kept building from the new level rather than retracing through it.

That's what makes the example useful: the chart did not predict the durability of the move. The durability came from the market deciding the company's earnings power had changed — and then continuing to defend that repricing.

When Gap-Reading Fails

• "It must fill" thinking. People who treated this as an ordinary gap were fighting a repricing backed by a real change in the business. The myth survives because ordinary gaps often do fill; applying that habit to extraordinary news is the error.
• Chasing the open. The opposite mistake: buying any big gap-up assumes the repricing is right and complete. Many earnings gaps mark exhaustion instead — the AMC study shows euphoria peaks that gapped and collapsed. Gap direction alone tells you nothing about what comes next.
• The chaotic open. The first minutes after a gap are the wildest of the day — the difference between buying and selling prices is widest, and swings are biggest, exactly when emotions are strongest.

The Lesson

• Gaps are information events, not patterns: the market repriced while you slept.
• Whether a gap fills or runs depends on the durability of the news — unknowable from the chart alone, which is why gap folklore fails in both directions.
• Watching how the market treats the gap zone afterward (does it hold as support, per role reversal?) taught more here than any rule of thumb.

Study It Yourself

In Ticker Coach, pull up a stock through several earnings dates and mark each gap. Track which filled within a month and which never did — then ask what distinguished them. It's usually the news, not the chart.

Hindsight, educational walkthrough — not a prediction, recommendation, or trade record. Gaps resolve both ways routinely; past patterns do not guarantee future results.