Altair
Skip to content

Trading Analysis Report: UBER

Analysis date: 2026-10-21 · Model: claude-opus-5-5 · AI-generated, not financial advice.

  • Analysis date: 2026-10-21
  • Rating: Hold
  • Generated: 2026-10-21 12:19:27
  • 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: Hold

Executive Summary: Hold UBER at or below standard weight and do not short. If you are overweight, trim to standard today. Stay at standard only if a drop to $62 is comfortable and a drop to $60 is survivable across your whole book, otherwise trim to that size. If you are flat or underweight, open nothing before the Q3 print. After the print, build in quarter-size tranches only on a clean checklist and a tape that holds the reaction-day low. A daily close below about $65.4–65.5 cuts exposure to roughly half, and a close below the monthly SuperTrend at $60.38 cuts to minimal exposure. Horizon is about 3 months through the Q3 print and its follow-through, with a target of about $70.2.

Current Price: 68.21

Price Target: 70.2

Confidence: Medium

Time Horizon: 3 months

IV. Risk Management Team Decision

Aggressive Analyst

Aggressive Analyst: Neither of you has spoken yet, so I'll answer the arguments I expect, and then say why I back the Hold even though I'm the one paid to be aggressive.

Start with what I expect from the conservative side: price below the 10 EMA, 50 SMA and 200 SMA, weekly and daily SuperTrend both down, MACD about to cross bearish, z-scores only around minus 1.1 to 1.2 so "not even oversold yet," and a Fed market pricing a December hike. That's a fair list, but nearly all of it is lagging evidence. The weekly SuperTrend stop is at 83.86, almost 19% above the price. It tells you where the stock has been, not what it's worth. The stock is down about 22% from its January closing high and about 32% from its 52-week high, while operating income grew 30% year over year last quarter and TTM free cash flow sits somewhere between $7.2B and $10.1B against a $139B market cap. That's a 5% to 7% FCF yield. The multiple has already absorbed a lot of the rate fear and the robotaxi fear. Selling after a 30% de-rating, into a support cluster that has held four times, is selling at the point where the reward for being wrong is largest.

I also won't accept "the $5.4B investing outflow is scary." Unexplained isn't the same as bad. Operating cash flow was $2.86B in Q2 and up about 12% year over year, and assets and liabilities grew together. That looks at least as much like an investment or securities purchase as a problem, and the Q2 10-Q is already public. The trader's plan makes the starter conditional on exactly that check. It costs nothing to read it today.

The conservative case is weakest on shorting. The plan says don't short, and I agree. A short from 68 targets maybe 65.5, so about 2.7 points of reward. Against that, you face a weekly TD count of 7, a daily count of 6, a stock sitting on its lower Bollinger band, a CEO who put $10M of his own money in at $70.96, and frustrated retail holders who could be squeezed. Bears get the least reward right at support, and the risk is a gap up on a good Tesla or Uber print.

Now the neutral view, which I expect to be that it's fairly priced, so wait. The probability-weighted value of $70.2 against 68.21 is only about +3%, and I won't dress that up. But this Hold isn't passive. The stop is 65.41, about 4.1% down. The 50 SMA is about 6.5% up and the 200 SMA about 8.5% up, so roughly 1.6 to 2 times reward to risk on a technical basis. The first resistance near 71.5 is only about 1.2 to 1. The weekly TD 9 is about two weeks out, and the monthly SuperTrend at 60.38 is still intact. The plan puts capital to work when evidence turns, which means the 10-Q check, a hold above 65.41 after the Tesla reaction, and sizing that survives a gap to $62-63. That keeps the upside available without paying for hope.

I'd push the neutral side on one thing. Waiting for every input, including the earnings date, the 10-year yield and the Q3 guidance, means buying after the repricing. The 10-Q detail and the post-Tesla reaction are knowable within a day or two. Don't hide behind the ones that aren't.

I'll concede a few weak spots. The 9-11% event volatility is bigger than the 4% stop distance, so a stop-out could be noise, which is why the starter is only 25% of standard. The options-alert post with its $72.50 and $76.50 targets is promotional, not independent sentiment. The sentiment read is low confidence, one day of StockTwits with eight tagged posts, though "dead money" and "I'm about to quit" are what washed-out sentiment sounds like. The December-hike pricing is unverified against FRED. And the President sold $2M the day after the CEO bought.

Put together, the high-reward move is to be ready, not to be early. Hold what you have at or below standard weight, don't add into Tesla, Meta and the FOMC, and keep a pre-committed trigger to start buying if the 10-Q is clean and 65.41 holds. If it breaks, cut to half. That keeps exposure to a 15x forward-earnings, cash-generating company at the support, with the downside defined. I back the Hold. Aggressive Analyst: I'll give back what you've earned first, since it makes the rest cleaner. The squeeze story is gone. We have no short interest data, and frustrated holders are longs, so that's supply risk, not fuel. I also overreached on "washed out" and "sitting on the lower band." RSI is 41, the z-scores are around minus 1.1, and the band is about a point (1.6%) below. You're also right on the stop. It's close-based, a print can gap through it, and cutting at 62 is cutting after the damage. So I'm dropping "defined downside at 4.1%" and taking the neutral framing: 65.41 is a review level, and size is the real risk control. My 1.6 to 2x also used the 50 and 200 SMA as targets in a bearish stack. Probability-weighted, it's more like 1.2 to 1.5. That's a thin edge, which is why the starter is 25% of standard and not more.

Now where I think you're wrong. Conservative, you said the TD exhaustion only arrives if the stock falls further first. Check the thresholds. The daily 9 needs closes below 70.35, 69.96 and 68.94 on the 22nd, 23rd and 26th, and today closed at 68.21, so a flat stock completes it. Weekly 8 needs the week ending the 30th below 68.11, a tenth of a point under today. Weekly 9 needs the week ending Nov 6 below 71.51. The count completes if the stock stays weak, not if it flushes to 62. It's a reversal watch, not a buy signal, but it will finish around the FOMC at roughly this price. That undercuts the idea that you have to wait for a much lower, cleaner entry.

Neutral, your trigger is better than "holds above 65.41," which I'll concede is nearly vacuous. But look at the math. Buy a reclaim at 69.6 with a 65.41 stop and you risk about 4.2 points, or 6%, to reach 71.5-71.9, which is only 1.9 to 2.3 points away. That's roughly 0.5 to 1 against you until you get to the 200 SMA, where it's about 1.05. Confirmation isn't free, and you pay for it in reward-to-risk. So I'd define the trigger with two routes. One is a retest of 66-67.4 that holds with RSI staying above its early-October lows of 34-36, the bullish divergence the technical report described. Buying around 66.5 against 65.41 risks about 1.1 points to the middle band's 3.1, close to 3x. The other is the reclaim, but with the stop re-based under the 67.15 band, not 65.41. Both give a stop a couple of points away instead of four.

I'll give you the FOMC. Nothing fires before the statement on the 29th. I won't give you the print, though. Conservative, your "no starter until after guidance" means being flat or underweight through a 9-11% event. Neutral showed that variance is symmetric. A 25% starter that gaps to 61-62 costs about 2.3 to 2.7% of a standard position, and a gap up of 9% makes about 2.3%. Your dilemma, that a harmless starter is too small to matter, misses what the starter is for. It's the first tranche of a staged entry, and it keeps you from being entirely wrong-footed by a gap up. Sitting modestly under standard is also a view, and the plan's own numbers don't support a bearish view. The four prior tests of the 65.6-67.4 zone each bounced roughly 6% to 17% off the low within a week. That's four observations, and a fifth test may fail, but that's an argument for the stop, not for stepping aside.

On fundamentals, your EV point is fair but barely moves the yield. FCF over a $148.5B EV is about 4.9% on the vendor's $7.24B and 6.8% on my derived $10.1B, versus 5.2% to 7.3% on market cap. I'll add a caveat nobody raised: we can't see stock comp, so FCF is probably flattered. Revenue deceleration from 20% to 12% is real, and so is the sequential margin dip. But Q2's 13.3% operating margin is still above every quarter of 2025, which ran 8.3% to 12.3%. A post-selloff valuation isn't a floor, I agree. The question is whether Q3 shows growth stabilizing, and that's what the print tests. One retail post claims gross bookings rose 24% against 12% revenue growth. It's unverified and I won't rest anything on it, but it belongs on the 10-Q checklist.

On the $5.4B, here's what you both missed. The Q2 10-Q was public well before September. The CEO bought $10M at 70.96 and the President bought $5.3M after that disclosure, so they knew what the outflow was. If it were a capital-allocation problem they were about to defend, buying into it with their own money would be odd. The CEO also sold $43.7M at 94-101 and bought at 71, about 25 to 30% lower. That's price-sensitive behavior. I agree it's not a floor, and both are underwater. But I read it as the best evidence in the file that the outflow isn't a red flag.

So the three of us agree on more than it sounds: no short, no adds into Tesla, Meta and the FOMC, trim any overweight to standard, and read the 10-Q today as a two-way input. Where I stand is this. Hold at standard if the book can absorb a gap to 61-62, and don't go underweight on a view the expected value doesn't support. Check the caller's other holdings for stacked rate and AV exposure. Confirm the Tesla and earnings dates. After the FOMC, start 25% on a clean 10-Q plus one of the two triggers above. Cut to half on a close below 65.41, and exit to minimal exposure below 60.38. Being ready beats being early, but being flat when the stock gaps up is just a different way of being wrong. Aggressive Analyst: Let me start with what I'm giving up, because you've both earned some of it. Neutral, you're right that my retest route doesn't work as I laid it out. A 1.1-point stop is about 0.6 of an ATR, and a pullback to 66.5 would almost certainly drag RSI back to the mid-30s, so "RSI holds above the October lows" is a coin flip, not a filter. I'm dropping it. You're also probably right that the 17% bounce came off the Q2 print, which makes it a catalyst, not support. Conservative, you're right that the insiders who bought also approved whatever the $5.4B was. Their buying says they like the decision, not that it's a benign securities purchase. And the 10-Q has been public since August, so reading it is diligence, not edge.

But follow that last point through, because it matters more than either of you has let it. The market has had that 10-Q for two and a half months. The stock rallied from 68 to 80 after it, then fell 15% with nothing new in the file about the company. The headlines on the way down are Waymo, Tesla, rates and a California driver union. A clean 10-Q doesn't touch those. So the 10-Q should be a veto, where an ugly answer kills the starter. A clean answer isn't a reason to buy, and I wouldn't make it a hurdle the plan has to clear.

Conservative, on the trim of a fifth, Neutral already showed it's the starter's own math in reverse. Your bear case centers on 68.4, which is today's price, so a blanket trim is a directional view your own numbers don't support. "We can't see the book, so fail the test" is also backwards. We can't see it, but the caller can, so give them a rule they can apply. I'm with Neutral's split: overweight comes back to standard, standard holds if a gap to 61-62 is tolerable, and anything else trims to what is tolerable.

Now the starter. The FOMC statement is Thursday the 29th, so the next session is Friday the 30th. My recollection, which isn't in the file, is that Uber reports somewhere between the last days of October and the first days of November. Neutral's rule, a print more than five trading days after the statement, probably closes that window before it opens, and Conservative's gate does the same. So let's be honest that the starter is likely dead in practice. I can live with that, since the cost is bounded at about 2% of a standard position either way. Neutral also said to gate on the print, not the Fed, which tells me the FOMC wait is mostly a courtesy and the earnings date is the one fact that sets everything. Look it up first.

But if the starter is dead, the plan is incomplete, because the money is made or lost after the print. The current post-print trigger, a reclaim of 69.5-69.7 with RSI turning up, is built for a stock that drifts, not one that gaps. The file has two prints that look like reactions. In August the stock went from 68.18 to 78.03 in three sessions. In early February it broke on 63M shares to 73.92. A good print gaps over your reclaim level and the trigger fires at 73 or 74, with the 200 SMA overhead and 65.41 irrelevant. That's the cost of buying after the repricing, and the plan has it built in.

So write the post-print rule now. If revenue growth holds at or above the Q2 pace of about 12%, operating margin holds around 13% or better, guidance isn't cut, and the investing outflow has a sensible explanation, go to standard weight in tranches even into a gap, with the risk level re-based under the print-day low. If the print is muddy, hold what you have. If growth slips well under 12% or margin breaks, cut to half and watch 60.38. That's more decisive than either of your plans, and it's what being ready rather than early means: the decision is made before the information arrives.

On cash flow, Conservative's 1.8x operating-cash-to-operating-income point is fair, and here's a sharper version nobody made. First-half operating cash flow grew about 7% against roughly 42% growth in operating income. That's why I'd take the low end of the FCF range, about 4.9% on EV. Q2 operating cash flow was still up about 12%, so nothing has reversed yet, but the float question belongs on the 10-Q checklist. This isn't a screaming bargain. It's a fair price for a cash-generative business with a thin edge, which is exactly why this is a Hold with a pre-made plan and not a buy.

Where I land: no short, nothing before the FOMC, overweight back to standard, hold at standard if the caller can take a gap to 61-62, and don't trim on a view the expected value doesn't support. Treat 65.41 as a review level and size for the gap. Confirm the earnings date today. Then set the post-print rule in advance so a gap doesn't leave you waiting for a reclaim that never comes. Aggressive Analyst: Neutral, I'll start with the gap path, because you're right and it changes my number. The bad outcome the plan actually fires on is cheap: the stock slides through 65.41, you cut to half, and the print hits a half position. The expensive one is the stock hanging at 65.5 to 66, never closing under the line, then gapping about 10% to roughly 59.5 with the full position on. That's close to 13% from today. I'll take your two-part test, comfortable at 62 and survivable at 60. A standard position that can't survive a 13% drawdown in a beta 1.27 stock isn't really standard, so the test also checks whether the standard weight was set right.

You've each earned other points too. Conservative, my post-print thresholds were the status quo, "about" isn't a rule, and cutting after a gap is the react-after flaw I'd already conceded on the stop. Neutral's fix is better than my draft: the checklist is a veto, the tape is the confirmation, and the first quarter-size tranche comes at the close after print day with guidance in hand.

On cash flow, conversion falling from 1.8x operating income to 1.4x is normalization, not collapse. I should still stop quoting the top of the yield range. Four to five percent on EV after stock comp is a fair price, not a bargain.

On the calendar, I called the FOMC statement Thursday the 29th. The 29th is a Thursday and the Fed announces on Wednesdays, so Neutral's 28th is very likely right, though neither of us has verified it. It only matters for counting days to the print.

On the Fed, Conservative, an 18-point rise in hold odds is a move toward the base case, not instability. I side with Neutral that October is mostly priced and December is a sizing input.

I'd still amend three things, all small.

First, the exit rules. Closing under 65.41 and falling below 60.38 are price rules that fire after the information, and we all agree a stop can't cap a gap. If a print gaps the stock through 60.38, the level tells you almost nothing and the print tells you everything. So I'd make the response depend on cause. If the checklist failed (growth below Q2's, margin breaking, guidance cut, the outflow unexplained), cut hard. If the numbers passed and the stock still gapped to about 13.5 times the vendor's forward EPS of 4.41, don't add, and don't liquidate on the line either. Hold and see whether the print-day low survives the next close.

Conservative will point to February, where the stock broke on 63 million shares and slid another 5% over seven sessions. That's fair. One print-like reaction in the file kept falling, and the other, if August 5 was the print, reversed. Two observations can't write a rule, which is why I'd anchor on the content of the print and not the price on the screen.

Second, the add rule. A 9% gap up from 68.21 is about 74.3, which already sits on the daily SuperTrend at 73.37 and the 200 SMA at 74.02. If tranche two has to clear that whole cluster, it only fires on a beat bigger than the typical event move, and you've paid 74-plus for it. I'd let tranche two fire on a second close that holds the print-day low. The 73.4 to 74 cluster is where you stop and reassess before the final tranche, not a hurdle for the second. The confirmation toll is real, and I'd rather not pay it twice.

Third, everything we've said about gap risk comes down to size, because a stop can't cap it. One instrument does cap it: a defined-risk options structure. For a flat caller who wants exposure through the print, that's the only way to own it with a known maximum loss. I can't tell you it's worth it. The file has no implied volatility, and pre-print premium usually already charges for the 9 to 11% move. If the structure costs about what the expected move is worth, you're buying the variance you were trying to avoid, and size is the better tool. But price it before calling the starter dead.

That puts me within rounding of both of you. Hold, no short. Overweight comes back to standard today. At standard, hold if 62 is comfortable and 60 is survivable, and otherwise trim to what is. If flat, wait, and a starter only makes sense if the print is well after the statement, which is probably not the case. After the print, the checklist is a veto, the tape is the confirmation, and the first quarter-size tranche comes at the close after print day. Exits depend on cause, not on a line.

Three facts decide this: the caller's gap tolerance, the earnings date, and what the Q2 10-Q says about the $5.4B. Get those today, and the rest is execution. Being ready beats being early, but only if the readiness is written down before the print. Aggressive Analyst: I'm taking most of what you've said.

Neutral, you're right about the 4.41. Market cap over price gives about 2.04 billion shares, so forward EPS implies roughly $9 billion of net income against $6.7 billion of TTM operating income. The 13.5x only works if the non-operating and tax items that flattered the past year keep showing up, so I'm retiring it as a cushion. The cleaner check is EV over operating income. That's about 22x now and, by my arithmetic, about 19.5x at 59.5. That's a modest discount after a 13% gap, not a bargain.

Conservative, you're right about the consequence. If the checklist passes and the stock gaps down, hold, add nothing, and cut to half on a close below the reaction-day low. When the tape and the checklist disagree, the tape wins, because the file has no consensus or guidance to say otherwise. A rule written today beats one improvised at 59.5.

On the tranches I'll take Neutral's version, with one correction. Gate full weight on conditions, not levels: a close above the 200 SMA and the daily SuperTrend flipping up, wherever they sit by then. The weekly SuperTrend is context only. But it needs about a 23% rally, not 18%. The 18.7% in the report is measured down from the stop, and 83.86 over 68.21 is about 1.23. The conclusion is the same, since a rule that waits for it never reaches full weight.

On days, I'll take Neutral's definition: the reaction day is the first full session after the numbers. My recollection, unverified, is that Uber reports before the open, which would make the print day itself the reaction day. The sequence is then a first quarter-size tranche at the next close that holds that session's low, a second on one more such close, and full weight on the gate. No guidance in hand, no tranche.

On options, you're both right. We have no implied volatility, the file's only options reference is a promotional post, and for a flat caller a call structure is the starter by another route. I withdraw it, except for the holder who can't trim because of taxes or concentration. Everyone else should trim, since that hedge is free.

I'd add one item to the veto checklist. In H1, operating cash flow grew about 7% while operating income grew about 42%. Q2 current liabilities rose $2.9B against $2.86B of operating cash flow. If operating cash flow stalls or falls year over year while operating income keeps growing, that's the payables float reversing, and it should count as a fail. The checklist is then revenue growth not below Q2's 12.2%, margin holding in the low 13s or better, operating cash flow keeping pace, guidance not cut, and the $5.4B outflow explained.

Here's what I'm still defending, and it's narrow. Don't trim on a view of the stock, because the bear case centers on 68.4 and the plan's own weighted value is about 70.2. Trim on the book, if 60 isn't survivable. We all agree on that now, so it's no longer a disagreement.

I also want the cost of this plan stated openly, since "ready beats early" only holds if the price of being ready is known. A flat caller who follows it and sees a 9% gap up to about 74 owns nothing going in. They then build through a quarter, a half and full weight at 74 or higher. They give up most of that first leg, and we accept it because the same variance cuts the other way with no edge to pay for it. That's a deliberate trade, not an oversight.

My landing point is Hold, no short, and no starter before the print. Any overweight comes back to standard today. At standard, hold only if 62 is comfortable and 60 is survivable across the whole book, including anything else that moves with rates or AV headlines, and otherwise trim to what is. A close below 65.41 is a review that cuts to about half, and it isn't gap protection, because size is. After the print, the checklist vetoes and the tape confirms. The tranches follow the sequence above, with a cut to half on a close below the reaction-day low whatever the checklist said.

The checks for today are the earnings date, the Q2 10-Q investing detail, the final 10-21 bar, and what else the caller owns. Three facts decide almost everything here: the caller's gap tolerance, when the print lands, and what the 10-Q says about the $5.4B. With those in hand the plan mostly executes itself, and the hard decisions have already been made before the information arrives.