I. Today's Market: The Nasdaq Fell 5.52%, Its Worst Day in About 18 Months
Unit: %, single-day change on November 5; all three readings are negative, so the zero line sits at the right edge and every bar runs left on one shared scale. The Nasdaq reads −5.52%, the S&P 500 −4.33% and the Dow −3.44%.
All three indexes fell: the Nasdaq closed at 24,314.99, below its 200-day moving average, for its largest one-day drop in about 18 months, while the S&P 500 closed at 7,355.84 and the Dow at 51,523.06. The Philadelphia Semiconductor Index fell about 12% in a day with all 30 members down, and the fear index closed near 28, up about 70% in a day. Only the day before, all three indexes had rallied, the Dow by 3.03%.
This time the market's worry was not AI demand but whether data centers get built. After the midterm results, the market repriced the AI data center trade on fears that construction could be slowed or halted at the local and policy level, and one investment bank estimated that about 30% of new U.S. data center capacity planned for 2028 is at risk. Nvidia closed at $181.60, down 9.26%, and three AI power stocks fell together: Vistra 6.15%, Oklo 13.20% and TeraWulf 13.85%. When a theme is repriced, first separate a worry about demand from a worry about delivery, because the two hit different links in the supply chain.
II. Stock in Focus: Its Peer Set Split Three Up and Three Down, and AppLovin Rose Most
Unit: %, single-day change on November 5. These six are grouped by hand on the nature of the business, not an industry field; the set spans positive and negative, so a zero line is drawn and all six share one scale. AppLovin sits at one end on +31.04% and The Trade Desk at the other on −3.17%, 34.21 percentage points apart.
AppLovin moved sharply against its industry that day, but it was not the only riser. Two averages have to be kept apart: these six averaged +10.31%, while the application software industry averaged −2.12%, so the denominators differ; Magnite in the same set also rose 20.75% and Unity 14.93% on its own results, while the tech sector as a whole averaged −5.49%. AppLovin closed at $356.10, adding about $25.8B to a $108.9B market value in a day on volume 4.0 times its average, with the price still at 25% of its 52-week range. Earnings drove it: third-quarter revenue was $2.18B, up 55.5%, and fourth-quarter revenue guidance of $2.38B came in above the $2.25B estimate; net income was $1.44B, up 72%, and operating margin rose from 76.8% to 78.2%. ⚠ But the new generation of predictive models is only partly rolled out, and its full effect shows up in later quarters. Alongside it, Apple fell 1.53% while consumer electronics averaged −2.75%, the smallest decline among the core technology stocks.
Five Dimensions: Relative Strength and Peer Rank Full, and Volatility Control Weakest at 19.6
Unit: 0 to 100, 0 at the centre and 100 at the rim, all five axes on one scale. Relative strength and peer rank each read 100, valuation heat 91.9 and trend position 25.5, and the weakest is volatility control at 19.6.
Read the full strength score against the low trend position. Relative strength is 50 plus five times the excess over the industry, and AppLovin beat the application software industry by 33.16 percentage points, which works out to 215.8, capped at 100; just 10 percentage points of excess fills the axis, so a score of 100 cannot tell +12 from +33. Trend position 25.5 shows that even after the jump, the price is still in the lower part of its 52-week range. A full score only says the stock cleared the bar; how far it cleared it lives in the raw move, and where it started lives in trend position.
III. What to Watch Today: Payrolls and Vistra, and Whether the Guidance Scope Changes
At 13:30|October nonfarm payrolls The market expects about 65,000 jobs after 29,000 in September; about 70% of bets point to a December hike, so a strong reading means higher rates
At 15:00|Vistra (VST) third-quarter earnings call No release time was announced, and it usually comes before the call; full-year adjusted EBITDA guidance is $6.8B–$7.6B
Same day|Oklo (OKLO) and TeraWulf (WULF) earnings The two fell 13.20% and 13.85% on November 5
Next Tuesday, November 10, at 13:30|October CPI CoreWeave (CRWV) earnings follow after the close that day
Next Friday, November 13, at 13:30|October PPI Jobs and inflation data land back to back within two weeks
Unit: billions of dollars, quarterly adjusted EBITDA from ongoing operations, on the same basis as the company's full-year guidance, with bars from zero on one shared scale. 2025Q3 reads $1.581B, 2025Q4 $1.742B, 2026Q1 $1.494B and 2026Q2 $1.767B.
$1.581B is today's year-on-year base Last quarter's $1.767B was up about 31%; the four quarters form a sawtooth, not a line, driven mainly by seasonal power prices in Texas and the East, so one softer quarter is not a weaker business
Why not revenue Last quarter's revenue actually fell year on year ($4.017B against $4.250B) because wholesale trading and hedge accounting distort it; by segment, Retail was the largest at $773M but grew only 2.2%, while Texas ($311M) grew 119% and the East ($642M) 53.6%, and the East holds the PJM nuclear fleet behind the 2,609MW, 20-year Meta contract
⚠ Whether the guidance scope changes The current $6.8B–$7.6B excludes the Cogentrix acquisition closed on October 30 and the Meta contract, so once they are folded in, the base changes and year-on-year growth no longer compares directly
IV. Frequently Asked Questions
Q: AppLovin already beat this quarter, so why does fourth-quarter guidance matter so much?
A: The quarter is what already happened, while guidance is the company's commitment for the next one. AppLovin's third-quarter revenue of $2.18B grew 55.5%, and its fourth-quarter revenue guide of $2.38B also topped the $2.25B estimate; the two layers together produced a 31.04% gain on a day the tech sector averaged −5.49%. In an earnings rally, check first how far guidance beats expectations, because a quarterly beat paired with guidance below expectations makes the two layers cancel out.
Q: AppLovin added $3.2B to its buyback authorization. Does that mean it will buy back $3.2B of stock?
A: No. A buyback authorization is a ceiling, not shares already bought; actual third-quarter buybacks were $1.4B. The authorization sets an allowance, and how much is bought and when shows up in the buyback amounts disclosed in later quarterly reports. When reading a buyback announcement, separate the authorized amount from what is actually executed, because the two are often far apart.
Q: Vistra's Texas segment grew 119%, so why is the East the one to watch?
A: Texas's 119% came largely from power prices and weather, which swing every year; the East holds the PJM nuclear fleet where the 2,609MW, 20-year contract signed with Meta earlier this year sits, and a further 1,200MW from the Comanche Peak nuclear plant is under contract to supply power from 2027Q4. Contracts mean that growth depends on agreements rather than weather. When reading segment growth, separate what prices delivered from what contracts lock in: the first swings every year, the second is held in place by the contract term.
The data in this article is as of the November 5 US close and all times are UTC; guidance and expectations are company or market statements, not published figures.
Disclaimer: This article is compiled and written by the MEXC RealStocks team. The data in this article is based on the closing of the US stock market on November 5, 2026. The content is a compilation of public market information, and individual stocks are publicly discussed targets, which do not represent the recommendation or opinion of MEXC and do not constitute any investment advice. More US stock content: @MEXC | @Alpha_MEXC | @MEXC_Research