By Alexey Kovalyov, Founder, AstraEye Labs
Measure the move first, then set it beside the evidence around it: how it compares with its sector, the broad market or crypto as a whole, what was published about it (the news and, for a company, its SEC filings) and, for a coin, how perpetual futures traders were positioned. An AstraEye Labs report does this for U.S. stocks, ETFs and crypto in one run, with its market figures computed in code from licensed market data and public sources, and its sources named. It shows what lines up with a move; it cannot prove what caused it.
A stock gaps after earnings, or a coin falls 10% in a day, and the first question is why. This guide shows how to research a move: measure it, see whether it was the asset's own, read what was published on the same days, check positioning, and weigh the evidence, shown on a real published report.
Start with the price itself: how far the stock or coin moved and over how many sessions, how that compares with its usual swings, where it left the price against its moving averages and recent highs and lows, and whether trading volume rose with it. In AstraEye, ATLAS, the market structure and momentum desk, computes these readings in code from daily price history.
Compare the move with its sector, the broad market and, for a coin, crypto as a whole. A stock that fell on a day its whole sector fell moved with its market; one that fell while its sector rose moved apart from it. Either is a clue, not an answer: neither says what caused the move or how much of it the market accounts for. In AstraEye, relative-strength analyses compare an asset's return with a benchmark's over fixed recent windows (20 and 60 sessions for a stock against its sector, 30 days for a coin against Bitcoin and Ether), so they describe the weeks around a move rather than a single session; breadth analyses show how many of the assets in the run rose or fell, with market-wide context where it is available; and correlation analyses show how the assets you chose have moved together.
Crypto does not always move apart from stocks: IMF staff found that the correlation of Bitcoin and S&P 500 returns rose from 0.01 in 2017 to 2019 to 0.36 in 2020 and 2021 (IMF, 2022), so a coin that falls with the stock market may be moving with it.
Read the news from the days around the move and, for a company, its filings with the SEC, such as an earnings release or a report of another material event. A move that follows a filing within hours reads differently from one with no news at all. In AstraEye, NOVA, the news and sentiment desk, reads the news coverage and, for companies, their filings.
Measuring how a price reacted to news around its date is an established method: the event study, introduced by Fama, Fisher, Jensen and Roll in 1969. A company's filings are free to read yourself on the SEC's EDGAR database.
Leverage can amplify a crypto move. Perpetual futures funding shows which side was paying to hold its position. Open interest is the value of open positions in dollars, so it can rise or fall with the price alone: read beside the price, it shows whether leveraged exposure grew or shrank, not whether positions were opened, closed or liquidated. In AstraEye, DELTA, the derivatives desk, reads funding and open interest for the coins traded on the perpetual futures venues it covers.
No single reading explains a move. Put them together: the size of the move, whether the market moved too, what was published and how traders were positioned, and note what could not be measured. In an AstraEye report, SAGE, the research guide, writes the bottom line from the desks' findings, says where they agree and where they conflict, answers the questions your project set, and lists the data sources so you can check any figure against them.
These are the parts of one published report, as its public page shows them: the bottom line, the questions its project set, the analyses it used with their methods, and its data sources.
Main conclusions
Predictions
Why: RSI near 24 and nearly 30% below the 50-day average typically produce at least a mechanical bounce, but the revenue guidance cut limits the upside.
What would change it: federal drone contracts reinstated or further downgrades.
Why: A 12% single-session surge that lands 1.8% below resistance on elevated RSI tends to stall before the next leg, absent fresh catalysts.
What would change it: strong analyst upgrades or macro catalyst.
Why: Estimates have not yet moved but the company itself cut guidance, and formal analyst models typically follow within one to two weeks of a revenue guide-down.
What would change it: management clarifies contract timing recovery.
A wave of formal analyst estimate revisions for AXON, or a break above resistance for ABNB, would alter the picture most; if AXON's delayed contracts are reinstated sooner than expected, the bearish read weakens materially.
Questions this research set out to answer
Airbnb (ABNB) up 12%, Axon (AXON) down 18% after earnings: what changed?
ABNB beat expectations on both EPS and revenue, raised its outlook, and announced a $12 billion buyback, surging about 12%. AXON missed on both lines after cutting its 2026 growth forecast due to delayed federal drone contracts, falling about 18%. The divergence is fundamentally driven and reflected across every technical, sentiment, and coverage measure.
What did each company report?
Coverage reports ABNB delivered EPS of $3.16 and revenue of $4.87 billion, re-entering growth mode with its fastest revenue growth in over three years, raising its full-year revenue and margin outlook, and announcing a $12 billion share repurchase. AXON reported adjusted EPS of $1.72 and missed on revenue, disclosing a $1.2 billion investment in Skydio and cutting its 2026 revenue growth forecast to about 30% after removing roughly $130 million tied to three delayed federal Dedrone deployments.
How did revenue and earnings compare with expectations?
ABNB beat EPS expectations by about 10% ($3.16 versus $2.86 expected) and revenue by about 2.5% ($4.87 billion versus $4.75 billion). AXON missed EPS by 9% ($1.72 versus $1.90, on an adjusted basis) and also missed on revenue, though the exact revenue figures were not stated in the coverage read.
How far did each stock move on Nov 6?
ABNB rose 11.67% on November 6 to close at 189.99. AXON fell 18.18% to close at 313.00. The prior session, November 5, had already seen AXON slip 9.81% and ABNB decline 3.54%, so over two sessions AXON dropped 26.21% from 424.18 while ABNB gained 7.72% from 176.37.
Where does each now sit against its 50-day average, nearest support and resistance?
ABNB closed about 12.4% above its 50-day average, with nearest resistance at 193.45 (1.8% above the close) and nearest support at 164.10 (13.6% below). AXON closed about 29.1% below its 50-day average, with nearest resistance at 426.91 (36.4% above) and no computed support level below.
Did analysts change estimates, ratings or targets after the reports?
Formal consensus estimates have barely moved. Both ABNB and AXON score near 49 on estimate revision momentum and are rated broadly unchanged. The analyst action scanner found two actions for ABNB and one for AXON, all within a bullish-leaning window, but that 14-day window extends well before the earnings reports and likely reflects pre-earnings positioning.
In which direction?
For both stocks the direction of estimate revisions is essentially flat. ABNB's 7-day EPS revision is negative 0.04%. AXON's 30-day EPS revision is negative 0.17% with revenue up 0.05%. Neither window isolates post-earnings revisions cleanly, so meaningful directional shifts from the reports may still be ahead.
What does the news say drove each move?
Coverage attributes ABNB's rally to its fastest revenue growth in three-plus years, the raised full-year outlook, and the $12 billion buyback announcement. Coverage attributes AXON's decline to a revenue miss, the cut to 2026 growth guidance, and the removal of roughly $130 million in revenue from three delayed federal Dedrone deployments. Coverage also notes AXON's $1.2 billion Skydio investment and AI recurring revenue crossing $100 million as partial offsets.
How widely is each explanation reported?
The ABNB growth and buyback narrative appeared across at least 25 articles from 7 sources. The AXON revenue miss and guidance cut appeared across at least 11 articles from 4 sources, with the Skydio investment covered in at least 8 articles from 5 sources. Both narratives are well-established in coverage, not single-source reports.
What would have to happen for either move to reverse?
For ABNB's rally to fade, execution on the raised outlook would need to disappoint or macro headwinds to travel demand would need to emerge, since the stock is now pressed against resistance at 193.45 with elevated momentum. For AXON's decline to reverse, reinstatement or rescheduling of the delayed federal drone contracts would address the specific revenue gap, and upward estimate revisions would confirm the street sees the miss as temporary rather than structural.
What would change this read?
This read would change if AXON's consensus estimates are revised sharply lower, confirming the guidance cut as structural rather than a timing delay, or if ABNB's resistance level breaks and holds, validating the post-earnings re-rating. The pending wave of formal analyst revisions for both names is the single most important update to watch.
The analyses behind this report, each linked to its own public page.
The newest reports in the research library, each readable without an account and each ending with the data sources it used.
How to read one, part by part, is in the guide How to research U.S. stocks and crypto with AI: one real report, step by step.
A report sets a move beside the evidence around it and says what lines up with it. Evidence that lines up is not proof: a move can have causes no data source records, and two moves on the same news can differ for reasons no report sees.
Every figure is as of the data it was computed from, and depending on the source that data can be real-time, delayed, end-of-day or periodic. A report is a record of the day it was researched and is not updated afterwards, so its figures age from that day.
A report does not say what time of day a coin's price and indicator figures were read.
Each research run is priced in Astra Credits, so it is built for reports, not for many quick lookups through the day.
It does not say what to buy or sell, and it does not place trades.
Read a published report on a move first, no account needed. Then choose the U.S. stocks, ETFs or coins you follow and run the same kind of research: every run shows its price in Astra Credits before it starts.
Measure the move first, then set it beside the evidence around it: how it compares with its sector, the broad market or crypto as a whole, what was published about it (the news and, for a company, its SEC filings) and, for a coin, how perpetual futures traders were positioned. An AstraEye Labs report does this for U.S. stocks, ETFs and crypto in one run, with its market figures computed in code from licensed market data and public sources, and its sources named. It shows what lines up with a move; it cannot prove what caused it.
It can gather and measure the evidence, if its figures come from market data rather than a language model's memory, which may not hold what was published on the day of the move. Research that computes the move, the market around it and the positioning from market data, and reads the news and filings from those days, can show what lines up with a move. No tool can prove what caused it.
Compare the move with its sector and the broad market, and for a coin with crypto as a whole. If they moved together, the asset moved with its market; if it moved apart, look for news or filings of its own. Neither says what caused the move or how much of it the market accounts for. AstraEye's relative-strength analyses make that comparison over fixed recent windows, so they describe the weeks around a move rather than a single session.
No. A report shows the move, the market around it, what was published and, for a coin, how traders were positioned, with the sources it drew on named, and its bottom line says where that evidence agrees and where it conflicts. Evidence that lines up with a move is not proof of its cause.
Yes. AstraEye Labs connects to ChatGPT, Claude, Gemini, Perplexity, Grok or Glama as an app, so the assistant can read your finished reports and their methods, ask SAGE about them, and price and start new research.
There is a free plan, and paid plans start at $19 a month (Space Analyst). New accounts get a 7-day trial with no card. Every research run shows its price in credits before it starts.