Anatomy of a Squeeze: NVIDIA, Autumn 2022

By October 2022, NVIDIA had lost roughly two-thirds of its value from its late-2021 high. Demand from crypto miners had collapsed. Gaming chips were piling up unsold. Growth stocks as a group were deep in a bear market. The chart looked like a former leader that investors had stopped wanting to defend.

And then the chart went quiet. This case study walks through what you could actually see that autumn — before anyone knew what 2023 would bring. It separates what the setup genuinely told you from what only hindsight can claim. Everything here is hindsight-labeled history, not a method for predicting the next one.

The Setup (What You Could See at the Time)

• The decline stopped making progress. After one last flush in October 2022, price stopped making meaningful new lows. Selling that had been relentless for nine months stalled into a sideways range.
• The swings shrank. Weeks of smaller and smaller moves pulled the Bollinger Bands (guide) to their tightest levels in many months. The bands are an envelope drawn around price — wide when swings are big, narrow when they're small. Bands this narrow are called a squeeze. The chart that had been front-page dramatic became boring.
• A base with a ceiling formed. A base is a sideways resting range that forms after a decline. Rallies kept stalling at the same broad ceiling, while dips found buyers at higher levels than the October low — the higher-low structure covered in Support & Resistance.
• Volume dried up on the dips. Down days inside the range came with far less trading than the panic days had — a volume dry-up (guide).

The honest read at the time: a deeply beaten-down leader had stopped going down and was coiling up. That's all. History says a squeeze often comes before a larger move. It says nothing about which direction.

The Concept

Squeezes matter because volatility moves in cycles: quiet stretches tend to come before loud ones. Add a base with higher lows, and chart readers see a market that may be running out of motivated sellers. But the direction question is only answered by the break itself — which side of the range gives way, and how much volume shows up when it does.

NVIDIA's autumn wasn't unique, either. In hindsight, Tesla's mid-2019 base traced a similar arc: months of compression, a breakout that fall on rising volume, then a long ride higher along the upper band (shown briefly in the video above; the fuller story is in the Tesla trend study). Two different years, one measurable condition — and only hindsight tells us which one became famous.

What Happened (In Hindsight)

In January 2023, price broke the top of the range on rising volume, then kept building higher bases. The business reason — an AI demand boom, confirmed by that May's earnings guidance — arrived after the chart had already turned. Over 2023, the stock roughly tripled.

Two cautions before that story sounds too clean. First, nobody watching the squeeze in November 2022 knew the AI boom was coming. The setup and the reason it worked were only connected afterward. Second, this example is remembered because it ended spectacularly. Studying only the winners has a name — survivorship bias — and it is exactly how bad instincts get built.

When This Setup Fails

The identical picture — beaten-down former leader, quiet base, tight bands — goes wrong all the time:

• The break goes down. The same squeeze in a company whose business keeps getting worse breaks the floor instead. 2022 alone produced many such bases in former growth leaders that broke down and kept falling.
• The fake-out. Price pokes above the ceiling on weak volume, slips back inside the range, and traps the people who jumped in early — the failed-breakout pattern from the breakout case study.
• Endless chop. Some squeezes just... continue. A chart can stay quiet far longer than anyone's patience.

Careful students of this setup handled the uncertainty with structure. Before the break, they picked the price that would prove the idea wrong — the floor of the base. And they demanded strong volume at the break before trusting it.

The Lesson

• Knowable in advance: the squeeze, the higher lows, and the price that would prove the idea wrong.
• Knowable only in hindsight: the direction, the size of the move, and the reason for it.
• The skill: reading conditions and defining risk — not calling bottoms.

Study It Yourself

In Ticker Coach, pull up 2022's other beaten-down growth names and find their autumn bases. Some resolved like NVIDIA; many broke down. Compare the two groups honestly — and notice how similar they looked before the break. That discomfort is the education.

Hindsight, educational walkthrough — not a prediction, recommendation, or trade record. Squeezes resolve in both directions; past patterns do not guarantee future results.