Altair
Skip to content

Trading Analysis Report: NOW

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

  • Analysis date: 2026-10-14
  • Rating: Underweight
  • Generated: 2026-10-15 00:24:15
  • 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

Current Price: 139.98 (close as of 2026-10-14; whether the bar was complete is unverified)

Price Target: 134

Confidence: Medium. The decision to trim is unanimous and well supported. The target and timing are less certain because several inputs are unconfirmed: the earnings date, implied move, prior guidance, the composition of forward EPS, the acquisition terms, and the caller's holdings and tax basis.

Time Horizon: 3–6 weeks (through the expected late-October Q3 release and the initial post-print read)

IV. Risk Management Team Decision

Aggressive Analyst

Aggressive Analyst: The usual instinct is that selling into a 61% rally is the timid call and that aggression means buying the breakout. I think that's backwards. The bold position is acting before confirmation, taking a clear view on a binary event, and sizing for it instead of hoping a stop will save you. That's what this trim does.

Start with the payoff. NOW closed at 139.98. The ceiling is the 147-148 zone, about 5 to 6% away, and the stock has been rejected there repeatedly. On 9/15 it printed 147.57 and closed at 141.90. On 10/12 it printed 148.14 and closed at 140.32. The 8/31 high was 149.60. The downside is two documented event gaps, with opens 8.6% and 15.3% below the prior close. A 126.81 stop sits 9.4% below, and a 15.3% gap from here lands around 118 to 119, straight through it. The stop is cosmetic against an event gap, so position size is the real risk control. Selling down to about 30 to 35% of a standard allocation turns a 15% gap from a 15-point hit into roughly 4 to 5 points.

The fundamentals back that up. Gross margin went from 78.9% a year ago to 70.7%. Cost of revenue grew 61.5% against revenue growth of 24%. Q2 operating income was $162M, down 55%, and H1 operating income is down about 18%. Net income exceeded operating income by $136M, so the bottom line flatters the earnings power. FCF was flat at about $2.0B in H1 despite 23% growth, and liabilities jumped about $6.5B in one quarter. Debt is around $8.5B against $2.5B of cash, with a 0.70 current ratio. The stock trades at 87.5x GAAP earnings, and the 28x forward multiple rests on a non-GAAP EPS I can't verify.

The neutral view will say the weekly SuperTrend is up, price is above every moving average, RSI is 56, and nothing is stretched. There are three problems with that. First, the weekly stop is at 103.13, about 26% below, so that signal won't change until after the damage is done. Second, ADX is 18.5, and the technical report itself says trend-following signals are less reliable in that state. In a range you sell the top of the range, and leaning on the MA stack is leaning on the weakest tool available. Third, the weakening evidence is real: OBV has stepped down from 586.8M to about 522M, MACD has plateaued, and the daily TD-9 is complete. "Nothing is stretched" isn't an edge for holding. It just means mean-reversion has nothing to offer in either direction.

They'll also say the market already knew the Q2 margin collapse and still bought the stock. That's true, and it's my argument. The stock is up about 52% from the 91.94 low on 7/23, so a clean Q3 is substantially priced in. Meanwhile the macro is moving against a long-duration name. Polymarket gives another hike by December 10 an 82% chance and puts Oct 29 at 24% and rising. Zero cuts in 2026 is 97%. CPI is running about 4.6% annualized over three months, oil is around $103, and the VIX at 15 looks complacent. On sentiment, StockTwits shows 16 bullish tags and zero bearish, but that comes from about nine authors, one of whom wrote 7 of the 16. The tone is "bears rn" and "170+ after ER." That's a one-day sample, so I wouldn't lean on it hard, but when the crowd is positioned that one-sidedly, the surprise risk skews down.

The conservative view will ask why keep any stub, and why not sell it all before a binary event. Because the aggressive way to take risk is to take the right risk. A full exit means being right twice, on the exit and the re-entry. The 7/22-23 volume spike lined up with the 91.94 low and was followed by a 61% rally, so events cut both ways here. The franchise is intact, revenue is accelerating at +24%, and the CEO bought $3M of stock near $105. A 30 to 35% stub keeps real participation if the print re-rates the stock. It's also not blind. The plan commits in advance to exit if the daily close breaks 126.81, or after the print if GM is below 70%, OM below 9.5%, or 9-month OCF is down more than 5%. I'd rather hold a small position with a kill switch than a zero with no way back in.

I'm not claiming certainty. The earnings date is unconfirmed, I don't have the acquisition debt terms, SBC or cRPO, and I don't know anyone's actual holdings. That's why the guidance is relative to a standard allocation, why it drops to about 25% if the print lands within a session of the FOMC, and why flat accounts shouldn't initiate at 139.98. I also can't read the Citi target revision, so I'm not counting it either way.

The aggressive move is to sell into the 140 to 148 strength, finish on any push toward 147 to 148, and keep the dry powder. If the pullback comes, the 130 to 137 zone, where the 10 EMA, Bollinger midline, 50 SMA and lower band cluster, is where I'd want to be a buyer, and only after the earnings date is confirmed. The setup pays about 6% to be long and risks 9 to 15% on the event, so I'd take the trim. FINAL TRANSACTION PROPOSAL: SELL. Aggressive Analyst: Before I defend anything, I want to say what I'm giving up, because some of my plan was wrong.

Neutral is right about the 147-148 zone. The 8/31 bar closed at 147.99, right at its high, so that's a high-water mark, not a rejection. What I have is two failed intraday pushes in about six weeks, on 9/15 and 10/12. That's a reference level, not resistance, and a plan that needs price to get there will walk into the print still holding shares it meant to sell. So the last tranche gets a date, not a price. Look up the earnings date first, since it's a fact and not a forecast. Whatever is above target gets sold by the close of the session before the print.

Conservative is right about my gap math. I used open gaps, and the close-to-close moves were bigger: 17.7% on 4/23, 11.5% on 12/15, and 9.95% on 1/29. At 17.7%, a 30 to 35% stub costs about 5.3 to 6.2 points of a standard allocation, not the 4 to 5 I said. I'll size off that number. Conservative is also right that my numeric post-print triggers are thin. A gross margin line of 70% when the trough is 70.7% means a flat quarter passes. I'd judge the print on guidance, sequential gross margin, and the acquisition's contribution instead. And I'm dropping the StockTwits sample, the TD-9, the insider trades in both directions, and macro as a sizing input. The market rallied 52% through hot CPI, so hawkishness is old news. I also agree the VIX at 15 says nothing about NOW's own implied volatility, and that I shouldn't buy the 130 to 137 zone before the print. Knowing the date doesn't remove the gap.

Now the pushback. Conservative, your case has an inconsistency. You say you can't forecast the print, but your cost-benefit uses an 8% up-gap against an 18% down-gap. If you can't forecast, the up-gap can be as big as the down-gap. This tape printed +14.4% on 5/29. If you can forecast, you have a directional view, so own it. "Paying 8% for information" also isn't cheap, because you pay the entire up-gap for it. Neutral had it right that "opportunity cost versus realized loss" is accounting. A point forgone costs the same as a point lost, and re-entering after a clean print means buying higher. Your 10 to 15% also gets the worst of both. It's too small to matter if I'm wrong, and I still pay both trades and any tax on the way.

Neutral, you say my stub pulls two ways. I think the 147-148 ceiling and the re-rate case describe different paths. The ceiling applies to drift into the print. On the event path, a clean print can gap straight through 148. What I do have is a lean on skew, not a call on direction. In July the Q2 margin collapse printed and the stock made its 91.94 low that same week, then ran 61% on the close. The market forgave it. Now you pay 52% more for the same forgiveness, on a 28x forward multiple that rests on EPS I can't verify, with GAAP at 87x. In July a clean print was a relief, and now it's the minimum. That's a lean, not a forecast, and it's why the stub is a hedge against being wrong.

So I'll meet you at the top of your range. The target is about 25% of a standard allocation by the close before the print, and 20% if it lands within a session of the October 29 FOMC. At 25%, an April repeat costs about 4.4 points, and a clean 8% gap up earns about 2. That's survivable for any account sized sensibly in the first place, and accounts that can't absorb it should already be at zero. I won't go below that, since a stub that small is a full exit with extra friction.

On defined risk, I like it, but the structure matters. A collar sells the 148 to 150 call, which is exactly the re-rate upside the stub exists to keep. If I'm holding for that, a put spread is the cleaner expression. A collar only makes sense if the puts are expensive and I'm willing to give up the up-gap. Nobody has the chain, so check the prices.

The execution is this. Sell the first tranche now into 140 to 143, where the upper Bollinger band sits. Sell the next on any push toward 145 to 148. Sell the remainder above target by the session before the print. Keep the 126.81 closing stop for drift, like the 6/1 to 6/25 slide that took three and a half weeks. Size handles gaps and the stop handles drift. After the numbers, re-add on a retest of 130 to 137, or on a close above 143 and then 148 with ADX through 20 to 25 and OBV above 531.7 to 534.9M. If the guide cuts margin or free cash flow expectations, exit the stub. If it gaps up, don't chase.

All of this is relative to a standard allocation. I don't have the date, the option prices, your holdings, or your cost basis, and a low basis changes what trimming costs. Flat accounts should wait.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: We've converged on almost everything, so I'll say what I'm taking from each of you and then argue the one number we still disagree on. The sale is time-bound, not price-bound. Nobody builds in the 130 to 137 zone before the print. The 147 entry, the "triple rejection" wording, the hard margin triggers, StockTwits, TD-9, insider trades and macro all come out of the trader's note.

Neutral is right that 87x against 28x is a GAAP versus non-GAAP gap, so the real question is whether $5.01 survives the guide. Neutral is also right that I mixed two paths when I compared +6% with -15%. Conservative is right that 5/29 and 6/1 weren't earnings events, so I shouldn't have used them to argue the print is symmetric. Those pops show the stock can gap up without a catalyst, which is an argument about the stub's value this week, not about the report. And I'll admit 25% came from meeting Neutral at the top of the range, which is no more derived than 12%.

So here is the tolerance input nobody supplied. ATR is 5.71 on 139.98, about 4.1% a day. That's the routine range the caller already lives with on a full standard position. Put 4.1 into Conservative's formula and divide by 17.7, and you get about 23%. An April repeat then costs a bit over 4 points, roughly one ordinary day's range on the full position. At 12%, the worst day in the sample is about 2 points, half a normal day. ATR is a range, not an average loss, and an overnight gap gives you no chance to react, so this is a yardstick and not a proof. But it's a stated tolerance with a reason, and it lands at 20 to 25.

Conservative, on the lean. I do lean down, which is why I'm selling most of the position. But go from 12% to 25% and you're holding 13 more points of stub. If the print averages a 2% drop, which is a big lean for a binary event, those points cost about a quarter of a point. The real difference is variance, and you said your mandate takes the lower variance. That's a legitimate preference, but it isn't an analytical edge. Your information-timing and acquisition arguments, as Neutral said, point to zero, and nothing in them stops at 12. The acquisition is the best company-specific risk on the table, but neither of us can say which way the guide cuts.

On the stop, you said it's worth at most 2.3 points against 4.4 for the gap. But 2.3 is what the stop costs, not what it saves. The slide from 135.86 to 89.52 was 34%. At a 23% stub that's nearly 8 points unprotected and about 2 with the 126.81 closing stop. In this sample the biggest loss was drift, not a gap, and drift is the part a stop can reach.

Neutral, on execution, I'd go further. ADX is 18.5 and the z-scores are near zero, so the tape gives no timing edge. I'd sell on a schedule, in roughly equal slices across the sessions left. A resting limit at 143 to 146 for one slice is a free option, but it shouldn't be part of the plan. On the date, the last gap days were all Thursdays. July 22 to October 21 is 13 weeks, the same spacing as April 22 to July 22 and October 29 to January 28. So I'd plan for the 21st, with the 28th possible, and verify it. Today is the 15th and our data runs through the 14th close. That leaves five sessions including today, or four if you finish a session early, as Conservative wants. Being early costs a few sessions of exposure to a stock we want small. Being late means carrying the full position into the gap.

On defined risk, the formula is tolerance divided by the gap. A put spread that caps an April-sized move near the 200-day would let the same tolerance carry a bigger stub. Elevated pre-earnings premium may make it not worth it, and nobody has the chain.

The aggressive part of this trade is what you do with the proceeds. After the April gap, the close went from 84.78 to 135.86 by 6/1, about +60%, though it gave all of it back by 6/25. After July it went from 91.94 to 147.99, about +61%. After January it fell another 29% to 83.00. Three observations aren't a rule. But selling buys the ability to decide after the guide, and the 130 to 137 retest is where I'd want real size, once the numbers are out.

My number is about 20 to 25% of a standard allocation by the last close before the release, toward 20% if it lands next to the October 29 FOMC. A holder who can't take a 4-point bad day should go lower. I still lack the date, the option prices, the holdings and the cost basis, and flat accounts should wait.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: I'll start with what's settled. We're now arguing over five points of a standard allocation, and Neutral is right that expected value is flat across that band. If the extra stub earns about a quarter point on a 2% average drop, it's insurance and not an edge, so I'll stop presenting my number as one.

I'll take Neutral's haircut on my ATR anchor. I matched a range you can watch and react to against an overnight gap you can't, and the honest adjustment lands near 20%, not 23%. So 20% is my default for a silent caller, and 15% if the reaction day is the Oct 29 FOMC decision day. The note should lead with the formula: the loss the holder will accept on a standard position, divided by the larger of 17.7% and 1.5 times the options-implied move. If a put spread caps the loss, the divisor becomes the capped loss, which is the one case where a bigger stub is earned.

Conservative, your weekly scaling doesn't beat the daily anchor. A gap is a one-day event, so the one-day range is the matched window. What the scaling does show is that the anchor is a choice, which I already said. The "too low costs a point, too high costs four" framing also can't set a default. Neutral caught the total-versus-increment problem, and there's a second issue. Any asymmetry argument built that way ends at zero, because holding nothing costs at most the up-gap and holding anything is exposed to the down-gap. You've already said your information-timing argument points to zero too. An argument that doesn't stop at 12 can't pick 15. A stated tolerance can, and that's what should move the number.

You're right about the stop, though. A closing stop fills at the next session, so budget 2.5 to 3 points of drift loss on a stub that size, not 2. That's another reason not to size the stub on the hope that the stop saves you. It handles a three-week slide like June's and does nothing for an overnight gap.

You both caught a contradiction on pace. I said the tape gives no timing edge and then asked for equal slices, which means waiting for nothing. If waiting earns nothing, sell early. I'd do about half the reduction by Friday's close, most of the rest Monday, and keep Tuesday as the spare. A resting limit at 143 to 146 on part of Monday's slice is a free option, provided Tuesday's close is a hard deadline either way. The case for planning on the 21st rests on our own price data as well as memory. The January and April gaps show up as Thursday open gaps, which is what a Wednesday after-close release looks like. So the deadline is the last close before the release, and I'd finish a session earlier. If the date is the 28th, the pace relaxes. Looking up the date takes two minutes, and I'd do it before anything else.

On re-entry, you've both shown that the 130 to 137 zone and my ADX and OBV conditions are the wrong things to hang it on. A January-sized gap opens near 128, an April-sized one near 115, and an 8% pop near 151, and none of those starts in my zone. So I'd re-add in steps on content: the guide, sequential gross margin, and the acquisition's contribution, with levels redrawn from post-gap bars. Where I'd push is speed. In both events where we know the gap's direction, the weeks after moved more than the gap day: 29% lower after January and 60% higher after April. July's gap sign isn't in our data, so that's two observations, not a rule. Still, a clean print may never hand us a pullback, so I wouldn't make the re-entry wait for one. If the guide holds sequential margin and the stock holds its gap for a few sessions, adding in steps is the bold part of this trade. Price confirmation should be the last step and not the first.

Neutral, I agree the proceeds sit in cash, but I'd rest that on optionality and not on macro. Moving into INTU or ADBE swaps NOW's event risk for the same rate sensitivity and defeats the purpose of the trim. But we all agreed hot CPI was priced through a 61% rally, so I wouldn't hold the first re-add for the 29th. Polymarket has a 74% no-change on that day, and waiting for it would bring the macro call back in through the side door.

So the plan is to sell down to 20% of a standard allocation, or 15% if the reaction day is the FOMC decision day. The formula moves that number once the options chain and the holder's tolerance are known. Front-load the selling and finish a session before the last close ahead of the release. Keep the 126.81 closing stop for drift and reset it from post-gap bars after the print. Judge the print on guidance, sequential gross margin, and the acquisition's contribution. Don't chase the open, hold cash in the meantime, and re-add in steps.

The timid options here are holding full size and trusting a stop that can't work against a gap, or going to zero and calling it prudence. This plan takes a view on skew before confirmation, sizes for the loss it can't escape, and keeps a stub and cash to act on the information when it arrives. I still don't have the earnings date, the option prices, the holder's actual position, or their cost basis, and a low basis changes what trimming costs. Flat accounts should wait.

FINAL TRANSACTION PROPOSAL: SELL Aggressive Analyst: I'll start with what I'm taking, because most of your last round is right.

Conservative, your convexity argument is the first one that picks a number for a caller who tells us nothing. On my own 2% average-drop assumption, thirteen extra points of stub are worth about a quarter point of expected value. Overshooting someone's tolerance is where they dump the rest, average down, or breach a mandate. Undershooting is cheap and overshooting is expensive, so the default belongs below the median guess. I'll take 15% for the silent caller and about 11 to 12% if the reaction day is the 29th. I'd go to 20% for someone who says they can sit through a 3.5-point bad day, or who is taxable with a low basis. The formula overrides all of it: tolerated loss divided by the larger of 17.7% and 1.5 times the implied move. The 15 versus 20 gap is under a point, so I'm done with it.

Neutral, you're right that I oversold the plan. If gap direction is a coin flip and the deterioration was public when the stock bottomed, this is variance reduction around an event nobody can forecast. The only lean is that after a 61% run, a clean print is the minimum. The note should say "trim for event risk, not a bearish call," so nobody sits out a clean re-rate. Your premium point on the put spread also stands. If insuring the extra stub costs anything near what it might earn, you've bought a smaller position at a worse price. Options are for a holder who wants participation and has the chain in front of them. The short strike has to sit well below 115, since an April repeat lands at 115.2, and the expiry has to run through the reaction day.

I'm also dropping my OBV and ADX re-add conditions. OBV is 9.6M shares from 531.7M, and the last two sessions traded about 13M each, so one up session clears it before any news. A reaction day of 55 to 84M shares moves OBV far more than that in either direction. ADX can't tell a rally from a collapse.

Now the places I'd push.

First, the stop. Conservative, your math is right. A January-sized close takes 139.98 to about 126, under 126.81. But the claim that it sells the stub before anyone has read the guide doesn't hold. If the release comes after the close, as the Thursday gap days suggest, the guide is public before the reaction session opens. A closing-stop breach that day sells the following morning. The real problem is that a mechanical level would substitute for a judgment we can make with the guide in hand. The fix is the same, though. The pre-print stop is cancelled at the release, the stub is judged on content, and a mechanical stop below the reaction-day low takes over.

Second, the mixed read. I'm with Neutral. You said rules should be written in advance so they aren't improvised after a 10% gap, and "sell if mixed" fails that test unless "mixed" is defined. The logic fails too. The stub is sized for event risk, and that risk is gone once the print is out. Selling ambiguity locks in whatever the gap was, and you've already conceded that a point forgone costs the same as a point lost. So write the three buckets now, measured against the Q2 run-rate: sequential gross margin, the free cash flow outlook, and the acquisition's contribution. Clearly worse on margin or FCF means exit. Clearly better means hold and don't chase. Anything else means hold with the reaction-day-low stop. A close below that low cancels every re-add not yet taken.

Third, the re-add, which is where the effort belongs. Neutral's three steps are right, and I'd sharpen two things. The January gap opened near 128, April near 115, and an 8% pop near 151. None starts in the 130 to 137 zone, and after an up-gap the pullback may never come. So define the price test from post-gap bars. After an up-gap, closes hold above the pre-release close. After a down-gap, no close below the reaction-day low. Step two also may not need to wait for the 10-Q. I believe the release and call usually carry cRPO and SBC, but that's from memory, not our tools, so check. This matters because the stock fell another 29% after January and rose 60% after April. Those are only two observations, but in both the weeks afterward moved far more than the gap day. The stub and the cash exist so we can act on the guide within days. On the 29th, I agree there's no add on the decision day itself. After that, the first add comes once the guide is read and the gap has held for a few sessions.

That leaves the following plan. The note leads with two lookups, the earnings date and the options chain. Then it gives the formula and the 15% default. Start selling today, with about half the reduction done by Friday's close, most of the rest Monday, and Tuesday the 20th as the spare. Plan around a Wednesday the 21st release until someone confirms the date. Keep the closing stop for drift until the release, then swap in the written post-print rules. Cash stays cash until the guide is read. The 147 entry, "triple rejection," TD-9, OBV leakage, the monthly SuperTrend and the hard margin triggers all come out of the note.

If anything here is bold, it's that size is set by the loss the holder can't avoid, and the payoff comes from how fast the cash goes back to work once the numbers are out. I still don't have the date, the option prices, the holdings or the cost basis, and flat accounts should wait.

FINAL TRANSACTION PROPOSAL: SELL