Meta (Facebook) shares (META) stock analysis, October 22, 2026: Earnings selloff lands inside an intact weekly uptrend

Meta Platforms (META) fell 5.6% to $658.61 on October 22, 2026, after a quarter that beat on revenue but arrived with a higher spending plan and a $25 billion legal verdict. The selloff has pushed the stock into oversold territory inside a weekly uptrend that is still intact. The balance of evidence for the next 2–6 weeks leans positive, with high conviction. The case rests on the trend, relative strength and history, not on the news flow.

Why it matters

The quarter itself was strong. Revenue came in at $64.6 billion, up 26% from a year earlier and above the $63.4 billion consensus, and adjusted earnings were $7.40 a share against $5.74 expected, Benzinga reports. Reported EPS was only $3.83 because of an $18.59 billion litigation settlement and a related $9.89 billion charge, The Motley Fool notes. Investors looked past the revenue beat to what the business is spending and what it may owe.

Three things weighed on the shares. Meta raised its capital spending plan, according to Barron's. A New Mexico jury ordered the company to pay $25 billion over privacy violations tied to Cambridge Analytica, and Meta plans to appeal. And the Financial Times reports a new class of non-voting shares from January to fund AI spending without diluting Mark Zuckerberg's roughly 61% voting control. None of that threatens revenue, but all of it raises the price of owning the AI build-out.

The move is the stock's own. META is down 10.6% over five days and 10.6% over a month, while SPY is up 0.45% over five days and 0.5% today. Early-session futures pointed lower and Morningstar describes a European AI-stock slide with Brent above $112. But the broad market is not what drove this decline. We have no sector ETF figure for comparison. The stock is now 15.3% below its September 24, 2026 closing peak of $777.59.

The argument is about payback. Zuckerberg told investors that the Muse AI assistant has reached product-market fit, with more than 10 million users in the U.S. and Canada in about six weeks and roughly 25% engaging daily, Yahoo Finance reports. The counterpoint, from an analyst in the same coverage, is that Muse is losing more money than any other product Meta has launched. Broader AI commentary, such as Yahoo Finance's AI stock picks, stays bullish on the physical build-out, but it does not speak to Meta's returns. On valuation, 58 analysts rate the stock a strong buy with a mean price target of about $818, which is 19.5% above the price. The targets run from $580 to $1,050, so the spread is wide and the low end sits below today's price.

What the evidence says

The long-term thesis is intact. Price is above its 40-week average and the 200-day average of $632.61, and the stock is 25.3% above its 52-week low of $525.72. The phase is a pullback inside an uptrend, not a new downtrend. Over 50 days META still leads the S&P 500 by 4.8%, and by 3.3% over 200 days.

For the next 2–6 weeks, the balance of evidence leans positive, with high conviction: three of four checks line up. A weekly-trend pullback has reached oversold, with RSI at 31.8. The stock is leading the market. History also favors the trend side: after 151 earlier one-day drops of more than 5%, the stock was higher a month later 70.9% of the time, against 61.8% for a typical month, with a median one-month move of 5.4%.

The strongest case against is that the damage is recent and sits on real fundamental questions. The stock is below its 10-week average and just under its 50-day average of $663.06. The drop followed a spending increase and a legal verdict that has not yet been appealed. Volume is currently 73% below its 20-day average, though that is an early-session figure and will change. One of the four checks did not confirm, which is why conviction is high rather than higher. It does not tip the balance because the trend, relative strength and history all still point the same way.

Levels that matter

The structure holds above $650, a round-number level just below the price. A close under it would be the first crack. The 200-day average at $632.61 sits next, and the trend ends below $609.12, a swing zone. A weekly close under that level would break the weekly uptrend. Below it, $600 and $593.70 are the next marked supports.

The breakout level is a daily close above $663.06, the 50-day average, on volume above its 20-day average. From there the first target is $688.80, a swing zone, which gives only 1.3 to 1 against a stop at $643.94. That stop is $650 minus a quarter of the 14-day ATR of $24.25. The second target is the 52-week high at $777.59, which gives 6.0 to 1. The math therefore clears the 3 to 1 bar, but only if the move runs past $688.80, with $700 and $750 as round-number resistance on the way.

What would change the picture

A weekly close below $609.12 would change the picture, because it would end the weekly uptrend that the positive reading depends on. The non-price fact that would most likely do it is evidence that AI spending is outrunning revenue, for example revenue growth slowing from this quarter's 26% while capital spending keeps rising and Muse losses widen. A court outcome that enlarges the legal bill beyond the $25 billion verdict would add to that.

What to watch

• Earnings: the quarter was reported on October 21, 2026, and the next date is not in the data. The number that matters is the capital spending guidance relative to revenue growth.
• Sector and macro: AI-related earnings elsewhere, oil above $112 a barrel and renewed tech selling in Europe moved the stock before its own news.
• Legal: Meta's appeal of the New Mexico verdict.
• Valuation: the price is 19.5% below the $818 mean analyst target, but the $580 to $1,050 spread shows how divided the forecasts are.