Altair
Skip to content

Trading Analysis Report: QCOM

Analysis date: 2026-10-14 · Model: claude-opus-5-5 · AI-generated, not financial advice.

  • Analysis date: 2026-10-14
  • Rating: Underweight
  • Generated: 2026-10-14 15:03:06
  • TradingAgents 0.6.0: anthropic, deep claude-opus-5-5, quick claude-sonnet-5-5
  • Analysts: market, sentiment, news, fundamentals; research debate rounds 5, risk debate rounds 5
  • Data vendors: core_stock_apis yfinance, technical_indicators yfinance, fundamental_data sec_edgar,yfinance, news_data yfinance, macro_data fred, prediction_markets polymarket

V. Portfolio Manager Decision

Rating: Underweight

Executive Summary: Underweight QCOM. If you hold it, trim to about 50% of a standard allocation. Sell the first piece near $172-173 on Thursday and Friday without waiting for a bounce, place limit sells for the rest up to about $178, and sell anything unfilled at market by the close on Monday Oct 26 (or two sessions before a confirmed earnings date, if earlier); if you are flat, stay flat and place no bids before the print. Control risk through position size, not stops: a daily close below $166.9 only finishes the trim, a close above $194.48 means reassess rather than stop out, and the reduced position rides through the Q4 FY26 print, with post-print rules written in advance for the next 1-3 months.

Current Price: 172.62

Price Target: 162.5

Confidence: Medium

Time Horizon: 1-3 months

IV. Risk Management Team Decision

Aggressive Analyst

Aggressive Analyst: Neither of you has spoken yet, so I'll make the case and then take on the objections I expect.

The decision is a trim to about half a standard allocation, not an exit. I'd call that disciplined aggression: act on the evidence now, keep half the upside, and hold dry powder for a better price.

The evidence points one way. Revenue growth went +5.0%, then -3.5%, then -4.0%, while implied opex growth went +12%, +16%, +22%. That is the opposite of operating leverage. Operating margin fell from 30.5% to 16.3% in six quarters, and Q3 operating income was down 41%. Q3 free cash flow was $495M, down 81%, with capex doubling. Over nine months, financing outflows of about $10B ran roughly $3.2B ahead of FCF, so cash is falling and the current ratio has gone from 2.82 to 2.02. At $172.62, the stock is paying for a recovery in a cost line that is still moving the wrong way.

The tape agrees. Weekly and daily SuperTrend are both down, MACD has made 13 straight lower readings, and OBV fell with price. Today closed below the 50-SMA and the 10-EMA. RSI is 42, so there's been no washout. The 10/13 bounce topped out at 180.78 and closed at 176.86, then gave back 2.4% today. A stock that can't hold a bounce after a headline legal win is telling you something about who is selling.

Here is how I'd answer the likely pushback.

The conservative analyst will say a 16.9x forward P/E, a 2% dividend that's well covered, and a sound balance sheet make this a hold. I'd agree the dividend is safe at roughly 37 to 42% of FCF, and that's exactly why we trim and don't exit. But the 16.9x rests on $10.20 of forward EPS, and nobody can say whether that's GAAP or non-GAAP. The latest quarter's $1.87 annualizes to about $7.50, or roughly 23x. A 2% yield doesn't compensate you for a 14.5% drawdown in 14 sessions. "Sound" describes the level of the balance sheet, and the direction of cash, liquidity and returns versus FCF is negative.

The conservative will also say selling now means missing a rebound. We keep half the position, so a rip back toward 200 still pays us. If the stock bounces toward 176 to 181, we sell the second half of the trim into that strength at better prices. The plan benefits from a bounce and doesn't need a collapse.

The neutral analyst will say to wait for earnings and decide with real information. But the print is a roughly 10% two-sided gap in a name with a 1.65 beta, and the date isn't verified. We also can't verify the earnings base, because there's a $7.4B Q2 net income distortion and an implied Q4 FY25 loss. Holding full size through that is a bet, not patience. Opex has to turn first, and the burden of proof is on the print. If it's good, we hold 50% and can add back at 100% if opex is at or below about $3.4B and revenue is -2% or better. If it's bad, we've already dodged half the damage. Because the date is uncertain, I'd start executing now and not wait for a bounce that may not come.

The neutral may also say the Arm verdict de-risks the story. It does, strategically, but retail summaries say damages were $2 nominal, Arm won some claims, and the royalty question is open. I can't verify those details from the tools. What I can see is the price reaction, and the market isn't rewarding the win. Good news that fails to lift a stock is a classic distribution tell.

I'll concede some things. The macro read is headline-based because FRED failed, and the 86% "another hike" market is thin at under $1M of volume, though the 97% no-cuts market is deeper. Today's bar may still be incomplete. The 171.02 low has held twice, and a bounce from the 168 to 171 zone is plausible. The reward-to-risk on the reassessment level isn't spectacular, with 194.48 about 12.7% above price and the rebuild zone at 152 to 158 about 8 to 12% below. That is why this is a trim and not a short. I'm also not leaning on insider selling, since it looks plan-based and is a weak signal.

For anyone applying this: if you're flat, stay flat and rest bids in the 152 to 158 area. If you hold a full position, sell down toward half, with the first half of the trim near 172 to 173 and the rest on limits into 176 to 181, finished a session before the print. Treat a daily close above 194.48 as the signal to reassess. The 167 to 168 area is within one ATR of price, so it's noise and not a stop.

The cautious-sounding choice of holding everything and hoping the print fixes it is actually the more exposed one. I'm for selling.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: We've converged on most of the plan. All three of us trim to about half, nobody holds full size into the print, and nobody is shorting. Neutral is right that "hold everything" was a straw man, so I'll drop it. What's left is execution and how much of the bear case is proven, so I'll concede where you've earned it and hold where you haven't.

Conservative, you're right that "start now because the date is uncertain" and "sell the second half into a rally" pull against each other. The fix is a first tranche now, limits for the rest, and a hard deadline of October 28 or two sessions before a confirmed date, then market. Neutral's 175 to 178 range is also better than my 176 to 181, since the last bounce died at 180.78 and the 10-EMA is 177.99. I'll take it. 194.48 is a reassessment level, not a stop, and nobody should put a buy-stop there against a half position.

I'll also retire two of my own points. The 23x annualizes the weakest quarter in the dataset, so it's a stress case, not a base case. And "distribution" overreached, because 11 to 12 million shares isn't heavy supply. The defensible claim is that nobody is paying up for the win. The one bullish tape item in the sentiment report, the $182.50 calls, expires Friday and is now about 5.7% out of the money.

Now where I hold. You both say the FCF collapse could be working capital, so look at which way the evidence points. Current liabilities rose $1.6B in Q3. A rise in payables, accruals or deferred revenue is a source of operating cash, so it would have flattered the $991M, not depressed it. The one kind of jump that wouldn't touch operating cash flow is debt moving into the current bucket, and that's a refinancing question, not a timing excuse. Non-cash current assets rose only about $0.8B, so I don't see a working-capital drain big enough to explain this. I'll grant the statements don't reconcile, so I can't prove it. Even if you write off Q3 cash flow as timing, operating income is down 23% over nine months, capex is up 101%, and Q2 FCF was already down 18% before Q3 hit.

Neutral, on the TTM point, nine months of financing outflow is $10.05B on its own. Setting that against a full year of FCF at $10.24B mixes periods. Add a fourth quarter of outflow and you're above FCF unless Q4 was unusually light. Cash is also down from $7.2B in December to $4.5B. I agree we can't say whether buybacks, debt or cash is absorbing the gap. But "we don't know what's giving" argues for less size, not for full size.

You also said the stock has round-tripped to its 9/8 level, so it isn't priced for euphoria. I agree, but compare it with where the stock traded on the same fundamentals. The quarter ended June 28 was public by the late-July gap that looks like the print. The stock then closed at 146.81 and based between roughly 157 and 170 through August. At 172.62 it's about 18% above that post-print low and slightly above the top of the base. I can't say what drove September, but the filed numbers haven't improved since August. The market is already assuming opex turns, and that has to reverse a line growing +12%, +16% and +22%. Your point that a 4 to 7% edge is modest is fair. That's why this is a trim and not a conviction call.

Conservative, Neutral is right on the quarter rule. Going from half to a quarter buys about three points of standard-position protection, costs the participation we kept the half for, and triggers 3.3% below price on a stock with a 4% ATR. That's a sell-the-low rule. If 166.9 breaks, finish any unsold trim and stop there.

Neutral's catch on the post-print rule is also good. Q3's own numbers would trigger the deepest cut. I'd read it as: more of the same isn't a reason to add back, and it isn't a reason to dump into the gap either. The pre-print trim exists so we don't have to sell the gap. After the print I'd key on the FY27 opex guide and on where the stock holds, and use bounces to trim further.

The rebuild bids are where I'll fight you hardest, Conservative. We're not trimming to avoid the event. We're trimming because the price is above our estimate of value, and the event only makes the position size matter. A fill at 152 to 158 is 8 to 12% lower, below the trader's own $160 to 165 central value, near the August lows and around 2x ATR. If macro drags it there, some print risk is already priced and you're buying under value. I'll meet you on size and timing, though. Three-quarters of standard is too much to leave resting. Ladder it at roughly 10% near 157 to 158 and 10% near 153 to 155, cap the add at about 20% of standard, and cancel both on the same deadline as the sell-by date. That's a bit bigger than Neutral's starter, with the same discipline.

In practice, express everything as a share of standard allocation, since we don't know anyone's holdings. If you're full, sell about half the trim now near 172 to 173, work the rest on limits to 178, and market out whatever is left by October 28 or two sessions before a confirmed date. A close below 166.9 accelerates the unfinished trim but doesn't push below half. Reassess on a close above 194.48. If you're flat, stay flat, with small laddered bids that get canceled before the print. The cost of being wrong is that we keep half and can rebuy. The cost of waiting is taking a 10% gap at full size against fundamentals that haven't turned.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: We're close enough now that I'll skip the victory lap and spend this on the few places where I think one of you is still off, and concede where you've earned it.

Neutral, you're right that current liabilities and non-cash current assets are blunt tools, so I'll stop arguing the working-capital bridge from the balance sheet. But look at what your point implies. The June quarter filing is already public, so the biggest unknowns can be checked today: whether Q3 operating cash was timing, what sits in that opex line, and what the $7.4B in Q2 was. I'd let that drive the plan. The first tranche at 172 to 173 goes now, because the trim stands on opex, gap risk and a modest valuation edge whatever the filing says. Then the filing sizes the second tranche. If the cash weakness looks like timing and the opex step-up is largely one-time, a third is plenty, which fits your embedded-gain point. If opex is run-rate spend and cash conversion is really weakening, go to the full half. For a taxable holder with a big gain, that's a tax constraint and not a change of view, and I'm fine with it.

On cash, I used the December number too, and Neutral is right that September 2025 is the cleaner base. From there cash is down about $1.0B, while the flows imply about $3.3B, so the truth sits somewhere in the $2.3B reconciliation gap. I'll call it a monitoring item. On the FCF range, you two have both made fair points, but look at what the top end needs. Getting to roughly $10B for the year means Q4 FCF around $3.2B, within about 7% of last year's $3.4B. That comes right after a quarter down 81%, with capex running at about double last year's pace. It requires Q3 to have been entirely timing and to reverse completely in one quarter. That's possible, and I wouldn't underwrite it. So the yield is somewhere between 4% and 5.4%, and a range that wide is itself a reason to run half size.

Conservative, you've talked me down on the rebuy ladder. A single order of about 10% of standard, canceled at the deadline and on a close below 166.9, is fine. I'll point out one thing, though. The bid sits around 157 to 158 and the cancel trigger is 166.9, so to fill, the stock has to fall from above 166.9 to the bid inside one session, with no daily close in between. In practice that's a flush-day order. I'd rather own a flush than not, but I won't pretend it's central, and at 10% it's worth about a point of standard allocation either way. The principle I'll defend is that dry powder exists to be used. If you trim above your value estimate and never intend to redeploy below it, you haven't made a plan, you've made an exit.

Where I'd stay more aggressive than both of you is after the print. Neutral wants 75% after the first full session and the last 25% only on a second confirmation, like cash flow normalizing in the filings. Conservative wants 75% and then a pause. The trader wrote down thresholds for a reason: opex at or below about $3.4B and revenue at -2% or better. If those are met and the guide supports them, waiting for a filing-based second confirmation means paying up after the re-rating for information the P&L already gave. I'd go to 75% after the first settled session, and to full size once the stock holds the post-print level for a few sessions or the cash flow statement lands, whichever comes first. A pre-registered condition beats a discretionary one that arrives late.

On the other cases, I agree with Neutral. Don't sell the gap on a bad print. The whole point of trimming now is that we don't have to. If a good print sells off on heavy volume, I'd stay at half, and I wouldn't buy that dip either. That's the real version of the Arm-verdict message, where good news didn't lift the stock, and I'd trust it more than I trusted this week's 11 to 12 million share tape.

On the calendar, I'm happy with the close on Monday the 26th, or two sessions before a confirmed print. My recollection of the October FOMC is also the 27th and 28th, so a Monday finish avoids forced selling into the decision. Friday's monthly expiration, with those $182.5 calls expiring, means this week's tape is noisy either way. So don't hang the first tranche on a bounce. Work it over a session or two, because today's bar may not be final.

Here's what I'd say to anyone applying this. Think in shares of a standard allocation. If you're full, sell toward half, with a third as the floor if you have a large taxable gain or if the 10-Q shows the cash drop was mostly timing. Sell about half the trim near 172 to 173 over the next session or two. Work the rest on limits up to about 178, since the 175 to 178 area is where the 50-SMA and 10-EMA sit and the last bounce died at 180.78. Sell whatever remains at market by the close on the 26th. A close below 166.9 finishes any unfinished trim but doesn't push you below half. A close above 194.48 means reassess, not stop out. If you're flat, stay flat, with at most one small bid that comes off at the deadline.

None of this requires us to be right about the cash story, the macro read, or whether the next bounce reaches 178. It requires one thing: a 10 to 12% gap against a half position costs about 5 to 6% of standard, instead of 10 to 12% at full size, and you keep half the upside if the print is good. The cash from the trim is an option on the gap, because after the number, with the guide in hand, you can add on whichever side the market gets wrong. That is the aggressive part of this plan: we pay a small known cost to be positioned to act when the uncertainty resolves.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: Neutral has the right idea: we've been giving one-point items the same attention as five-point ones. I'll go through what's left in order of importance.

Conservative, you're right about my "option on the gap" language, and I'll retract it. Selling half doesn't cost a small known amount. On a 10% gap up it costs about five points of a standard allocation, and on a 10% gap down it saves about five. The trim is symmetric around the event. What it buys is a modest edge, two to three and a half points if the $160 to $165 central value is roughly right, plus cash that lets us size after the number instead of before it. That's a positive-expectancy variance cut. I'd still push back on "we redeploy higher after a good print and into a worse guide after a bad one." Those don't cancel. The price we redeploy at depends on how the market reacts. A bad guide with the stock down 15% is a different purchase from a bad guide with the stock flat. A full-size holder has to sell to get that choice, and that is the gap-selling we all want to avoid.

On the 10-Q, I side with Neutral on the logic and Conservative on the default. The filing can't shrink the gap, and it can't tell us whether anything reverses in Q4. But the 4 to 7% edge does rest on opex persisting, and the filing is where one-time items would show up. So if it shows both cash timing and one-time items in opex, holding two-thirds instead of half is defensible on evidence, not just tax. If it shows one or neither, we stay at half. The filing moves us within that band and doesn't take us outside it.

On the volatility argument, I'm with Neutral. A 1.65 beta is already in the standard allocation, so the chart of +100%, -41%, +37% and -14% doesn't justify going lower. I'd also point out that the same chart is why I won't go below half. A stock that can double in two months is one where sitting at zero is also a bet. The incremental reasons for half are the price above value, the opex line still accelerating, and the history of 10% gaps. The two-events point is real, but it argues against adding before the Fed, and our plan already does that. The sells and the single bid all come off by the close on the 26th, before the 27th-28th meeting, which we should still verify.

Conservative's FCF scaling to roughly $9.3B and a 5% yield is a fair middle, and I'll take it. It's real support. But a 5% yield on a number that has fallen four quarters in a row is a cushion, not a floor.

On re-entry, I've lost some of this argument. Conservative is right that "the cash flow statement lands" is a vacuous trigger, because the release normally carries it the same evening. Neutral is right that $3.4B opex is a 7% sequential drop and a full give-back of the Q3 step-up, so the trader's threshold is a "clearly good" screen, not a pass-fail test. Opex is also derived by subtraction. I'll accept Neutral's rule: 75% after the first settled session on a good print, and full size when management gives a specific opex number or direction, the stock holds the post-print level, and the 10-K doesn't contradict the release. I'd add one thing so the filing doesn't become an excuse to wait indefinitely. If it lands and is clean, we go to full size the next session. The cost of waiting is a week or two of drift, so we're arguing about a small amount either way. Check the filing date along with the earnings date.

On the bad-print rule, Neutral and Conservative have converged on cutting toward a quarter within three to five sessions if the guide shows opex still rising, and I accept the structure. I'd change one thing: don't touch the stock on gap day. That session has the widest spreads, the most algorithmic flow and the most overshoot. Start on day two, finish by day five, and cut only if the guide is bad and the stock hasn't already repriced. If it holds on a bad guide, the market is telling us it's priced, and we stay at half. This keeps the promise the pre-print trim makes, which is that we never have to sell the gap itself.

The bid is settled at about 10% of standard, off at the deadline, and not worth more argument. A 166.9 close finishes any unsold trim and cancels it, though as Conservative noted, the cancel can't protect us in the flush scenario that fills it.

The bold part of this plan isn't the selling. It's being at half going into a binary event, with no dependence on a bounce or a macro call, and then acting quickly in either direction once the guide is out. Full size into this print would be passive. It would mean taking a 10% gap against fundamentals that haven't turned, on a stock that hasn't priced much of that risk.

For anyone applying this, think in shares of a standard allocation. If you're full, sell toward half. Sell the first piece near 172 to 173 over the next session or two, since today's bar may not be final and Friday's expiration makes the tape noisy. Work the rest on limits up to about 178. Sell whatever is left at market by the close on Monday the 26th, or two sessions before a confirmed earnings date if that's earlier. A close below 166.9 finishes the unfinished trim and cancels any bid, but doesn't push you below half. A close above 194.48 means reassess, not stop out. Hold two-thirds instead of half only for a documented tax reason or if the 10-Q shows both cash timing and one-time opex. Use at most one bid of about 10%. If you're flat, stay flat.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: We're now arguing over one or two points of standard allocation, so I'll concede what I owe and spend the rest on the two places where a rule could misfire.

Conservative is right that I said FCF fell four quarters in a row, and the data doesn't show that. It's up 2.5% year over year in Q1, then down 18%, then down 81%. Two declines, accelerating, is the claim to make. I also retract "the stock hasn't priced much of this risk." I agreed the edge is only 4 to 7%, and a stock down 14.5% in 14 sessions has priced some of it. Some is priced and some isn't, which is what half a position expresses. I don't need the stronger claim, because the trim stands on three things the cash story doesn't touch. Implied opex is growing +12%, +16%, +22% against revenue of -3.5% and -4.0%. Operating income is down 41%. And the stock sits at the top of its August base with a history of 10% gaps on prints.

Neutral's repair of the pre-written test is right. The $160 to $165 central value is the persistence case, and a bad guide should be worth less. So here's my number, which is my judgment and not a tool output. A bad guide gets about 20x the $7.50 GAAP run-rate, roughly $150. The rule: if the guide shows opex still rising and the stock is still above $150 after the first full session, cut toward a quarter on days two through five, and sell whatever is unfilled at market on day five. If it has already gone below $150, hold at half. Write the number down before the print. It means a 12% gap to about $152 still triggers the cut, which is intended, because that stock hasn't repriced to the bad case.

On the re-entry gates, Neutral is right that stacking them turns the last quarter into a cap. One more point against them. As I recall, QCOM guides the next quarter and not the year, so we get one quarter of forward evidence and shouldn't ask more of it than that. I'll take Neutral's fallback. Back out the implied operating margin from the revenue and EPS guide, using the reconciliation in the release, and ask whether it clears the 16% area. My rule is 75% after the first settled session, and full size when the implied margin is rising and the stock has held its post-print level for three sessions. The 10-K can veto but can't gate, with a time box. I also suspect QCOM files the 10-K within days of the release, not weeks, though that's memory. Check last year's filing dates on EDGAR, because if I'm right the Conservative-versus-Aggressive disagreement over waiting for it disappears. Either way we're arguing about roughly a point of drift.

On half versus two-thirds, I accept Neutral's concrete test. Named one-time items must explain most of the roughly $260M step from Q2 to Q3 opex, and nameable, reversible items must explain the $991M of operating cash flow. Conservative's point also stands. Even if the filing explains all of Q3's step-up, opex is still about 16% above a year ago, which is why the ceiling is two-thirds. The 10-Q is public and takes about twenty minutes, so read it before placing the first order. Decide the target once, then don't revisit it. The wash-sale point is fair too. Pick lots deliberately and ask a tax person, but it doesn't change the target.

On the bid, Neutral's refinement, that it's live only after the trim is done, makes it nearly moot. If the trim finishes at the deadline, the bid is live for zero sessions. I'll keep it as an option and not a plan.

The other analysts have framed the aggressive position as leaning on a cash bear case we can't prove. It doesn't. I'd sell half with the cash story fully unresolved, because the case is opex, a modest valuation edge and gap size. What makes this aggressive is that we act before the catalyst instead of reacting to it. Every post-print move is written in advance, and we get back to size quickly if the guide earns it. Holding full size and hoping the print fixes a line growing 22% a year is the passive choice. Stacking so many confirmations that we're still at 75% three weeks after a good print is passive too, just slower.

The plan, in shares of a standard allocation. If you're full, read the 10-Q first. Then sell the first piece near 172 to 173 over Thursday and Friday, knowing Friday's monthly expiration will make the tape noisy. Work the rest on limits up to about 178. Sell whatever is left at market by the close Monday the 26th, or two sessions before a confirmed earnings date if that's earlier. A daily close below 166.9 finishes any unsold trim and cancels any bid, but doesn't push you below half. A close above 194.48 means reassess, not stop out. If you're flat, stay flat.

After the print, a good number with the stock holding means 75% after the first full session, then full size on the margin test and three sessions of holding. A good print that sells off, or a bad print that holds, means stay at half and don't chase. On a bad guide, don't touch gap day, and apply the $150 test on days two through five.

FINAL TRANSACTION PROPOSAL: SELL