I’ve been watching Meta Platforms closely since late September, and the price action over the past few weeks has been nothing short of brutal for anyone who bought the post-Muse strength. The stock looked clean when it was trading near $752. Then earnings hit, capital spending guidance jumped again, free cash flow turned negative for the first time in over a decade, and the chart basically got steamrolled. What was once a clean uptrend now sits at a real crossroads. Cloudflare, on the other hand, took a similar software-sector punch but came out looking very different. Buyers showed up exactly where they needed to. That contrast is what makes this week interesting.
Why Meta’s Setup Has Changed Completely
Let’s start with the facts on Meta. When the Muse AI product launch was still fresh, the stock was sitting at $752 and up roughly 38 percent from its August low. Momentum was strong. Relative strength looked solid. The technical picture suggested higher prices were likely if the company could keep converting AI spending into real user products. Then the third-quarter report landed on October 21 and everything shifted.
Revenue came in at $64.6 billion, up 26 percent and ahead of the $63.3 billion consensus. Fourth-quarter guidance of $73 billion to $76 billion was roughly in line with expectations. On the surface those numbers look fine. The problem sat deeper in the income statement and cash flow statement. Operating income dropped 37 percent to $13 billion because of a nearly $10 billion legal settlement charge. Strip that out and operating income still grew 11 percent with adjusted EPS at $7.40. But the real shock for many investors was the capital expenditure number and the free cash flow result.
Capex hit $34.9 billion in the quarter. Full-year guidance was raised to a range of $135 billion to $145 billion. Free cash flow turned negative by $2.1 billion compared with a positive $10.6 billion in the same period last year. That is the first negative free cash flow quarter Meta has reported since 2012. For a company that has spent more than a decade generating enormous cash, that number landed like a punch. The market did not like it.
After an initial after-hours bounce, the stock gapped lower and kept selling. From the October 13 high of $762.50 the shares have dropped about 14.5 percent while the broader market has barely moved. That kind of relative weakness is hard to ignore if you trade with a technical framework.
The Levels That Actually Matter Right Now
On September 28 the trader stop level was clearly defined at $685. When the stock gapped down through that price on October 22 and opened near $676, any disciplined trader following the plan was out on the open. That is exactly how risk management is supposed to work. You set the level before the news hits so you do not have to make emotional decisions in real time.
Today the stock sits near $643. The 50-day moving average has rolled over and now acts as overhead resistance around $664. Until price can reclaim that level with conviction, the short-term technical structure remains broken. On the downside the 200-day moving average around $633 and the bottom of the old Muse gap near $625 form the next important zone. The stock already tagged $633 on Friday and bounced a little. That bounce is still tentative.
Momentum has deteriorated fast. RSI has fallen from the 70s in late September into the mid-30s. That is a full flush in roughly three weeks. A low RSI on a broken chart simply confirms that selling has been heavy. It does not automatically mean the selling is finished. In my experience, charts that look this washed out can either reverse hard or keep grinding lower until a real catalyst appears. Betting on the V-shaped recovery here feels like the lower-probability move.
The set-up is broken, and $664 is the level the stock has to take back before that changes.
For longer-term investors the $626 area remains the more important weekly close level. A decisive weekly close under that price would fill the Muse gap completely, take out the 200-day, and reopen the old trading range that dominated earlier in the year. Until that happens, the longer-term structure is damaged but not fully destroyed.
What Investors Are Really Debating
Valuation-focused investors will point out that Meta still generates enormous operating leverage once the legal charge is stripped out. They will also note that more than 10 million people in the United States and Canada set up Muse agents in the first six weeks. That adoption number is real. The company also created a new non-voting Class C share class, which gives it more flexibility to raise capital if needed for the AI build-out.
Those points matter. They just do not matter much to a technical trader who watches price, volume, and moving averages for a living. When the trend changes and key levels break, the job is to step aside, not to argue about the long-term story. I have found that mixing the two timeframes usually leads to larger losses than either pure approach would produce on its own.
Right now there is simply no clean long setup for traders. Anyone who respected the $685 stop is already on the sidelines. The only remaining question is whether the stock can stabilize above the 200-day and eventually reclaim the 50-day. Until that happens, the path of least resistance still looks lower or sideways at best.
Cloudflare Tells a Completely Different Story
While Meta was getting cut in half from its recent high, Cloudflare was busy defending its trend. The stock was written up in late September near $324 after it pushed to new highs. The thesis was simple: Cloudflare runs a global network that sits between the internet and its customers. It sells security tools and a developer platform that lets companies build and run applications on that same network. The AI angle is traffic. Non-human traffic already passed 50 percent of the network in the second quarter, and daily AI agent requests were up more than 1,700 percent year over year.
Valuation is the obvious bear case. At roughly $326 the stock trades around 260 times the midpoint of this year’s non-GAAP EPS guidance. That multiple leaves zero room for error. Cloudflare reports third-quarter results on November 5. The market will demand proof that the growth justifies the price.
Price action since late September has been messy but constructive. The stock ran to about $361 by early October, then sold off hard with the rest of the richly valued software group. It dropped 19 percent and touched $285 on October 20. That low was higher than the early September trough near $270. The sequence of higher lows remained intact.
Look at the October 20 candle itself. Price traded through the prior day’s low, hit $285, and closed back near $292. The result was a doji with long wicks on both sides. That is classic rejection of the low. The next session closed above the top of that candle. Then Friday delivered a 7 percent jump that carried the stock back above the rising 50-day moving average near $319. The 200-day continues higher around $245. That is exactly the kind of buyer tenacity you want to see in a still-intact uptrend.
How to Trade the Cloudflare Dip
The question every trader asks after a sharp sell-off is simple: when is the dip worth buying? The answer is when the larger trend remains intact and buyers show up at the expected higher low. That is what happened here.
Momentum also behaved well. RSI dropped from the 60s into the low 30s at the October low and has already recovered back into the middle of its range near 51. Price held its higher low while momentum reset. That combination usually leaves room for another push higher without starting from an overbought reading.
For active traders the logical stop sits at $299, the top of the October 20 doji. A close back under that level would mean the confirmation of buyer support has failed. You get your answer early and keep the loss small. For longer-term investors the more important weekly level is $285, the actual low that was rejected. A weekly close under $285 would break the sequence of higher lows that has defined the uptrend since spring.
- Higher lows since spring remain intact
- 50-day moving average is rising and has been reclaimed
- Momentum flushed and recovered without breaking structure
- Clear invalidation levels exist for both traders and investors
That list is why the Cloudflare setup still looks usable while the Meta setup does not. One chart is defending its trend. The other has already violated its key levels and needs to repair a lot of damage before it becomes interesting again.
Risk Management Is Not Optional
Markets change. News hits the tape. Earnings surprise. Guidance shifts. Capital spending plans expand. None of that is unusual. What separates consistent traders from the rest is the willingness to accept that a previously good idea can become a bad one in a matter of days. Pretending the original thesis is still intact after the chart has broken is how accounts get damaged.
In Meta’s case the $685 level was the line in the sand. Once price gapped through it, the trade was over for anyone following the plan. There is no need to invent new reasons to stay long. The system did its job. Cloudflare offers the opposite lesson. When a stock in an uptrend gets hit by a sector-wide sell-off and still manages to print a higher low and reclaim its moving averages, that is information. Buyers are still present. The trend has not been invalidated.
I have watched too many people refuse to honor stops because they “believe in the company.” Belief is not a risk-management tool. Price is. Levels are. If the levels break, you step aside and wait for a new setup. That discipline feels uncomfortable in the moment. Over a long enough sample it is the only approach that keeps you in the game.
Putting the Two Charts Side by Side
Meta and Cloudflare both live in the broader technology complex. Both have meaningful AI exposure. Both took hits when the market rotated away from high-multiple growth names earlier this month. The outcomes so far have been very different.
| Factor | Meta Platforms | Cloudflare |
| Recent High | $762.50 | ~$363 |
| Key Support Tested | 200-day near $633 | $285 higher low |
| Current Structure | Broken short-term uptrend | Intact higher-low sequence |
| Trader Stop Level | $685 already violated | $299 |
| Investor Anchor Level | $626 | $285 |
| Momentum Condition | Deeply oversold, still weak | Reset and recovering |
That table makes the contrast clear. One name needs to repair damage before it becomes tradeable again. The other is already showing the kind of price behavior that often precedes another leg higher, provided the broader market cooperates and the upcoming earnings do not disappoint.
What Comes Next for Both Names
For Meta the immediate task is simple. Hold the 200-day. Reclaim the 50-day. Until both of those things happen, any bounce is just a bounce inside a damaged structure. The elevated capital spending guidance means free cash flow will stay under pressure for the rest of the year. That reality is already reflected in the price. The stock can still recover, of course. Companies with Meta’s scale and user base have come back from worse. But recovery will have to show up on the chart first.
Cloudflare faces its own test on November 5. The valuation leaves little margin for error. If growth remains strong and the AI-related traffic numbers continue to impress, the stock has room to challenge the $363 high again. If the numbers disappoint, the $285 higher low will be tested quickly. The beauty of the current setup is that the invalidation level is clear and relatively close. You do not have to sit through a large drawdown to know whether the idea is working.
Perhaps the most useful takeaway is that not every sell-off is the same. Some break structure. Others simply reset momentum inside an ongoing trend. Learning to tell the difference is one of the highest-leverage skills a trader can develop. Meta currently sits in the first category. Cloudflare sits in the second. Acting accordingly is the entire job.
Markets will keep serving up these kinds of contrasts. Some names will look broken and stay broken for a while. Others will look broken for a few sessions and then reveal that the buyers never really left. Staying flexible, honoring predetermined levels, and refusing to mix long-term belief with short-term risk management remains the only sustainable approach I know. The charts do not care about the stories we tell ourselves. They only care about price and whether the levels hold.
Right now Meta needs to prove it can stop falling. Cloudflare has already shown that its buyers are still willing to step in. That difference is everything when you are deciding where to put capital and where to stay in cash. The next few sessions and the upcoming Cloudflare report will tell us more. Until then the levels are clear, the risk is defined, and the job is to wait for the market to show its hand.
