Altair
Skip to content

Trading Analysis Report: MCD

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 12:57:57
  • 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 MCD at 245.28. Add nothing and place no resting bids until the Q3 2026 print date is confirmed and the results are read. Test existing size against an 8% gap to roughly 225.7; if that loss is not tolerable, sell the excess with limit orders at 246-250, and sell at market if those have not filled by the last session before the print. After a run-rate-or-better print, and only after a full post-reaction session closes back above the 10 EMA, open a first tranche of ¼ to ⅓ of a standard allocation near the 10 EMA, with one stop at least 2 post-reaction ATR away (today about 229.20, which is 2.7 ATR from 241). Further adds come only on closes above 250.18, then 257.49, each holding a retest; never exceed a standard allocation while the weekly trend is down, and exit on a daily close below 229.20, negative U.S. comps or a guidance cut.

Current Price: 245.28

Price Target: 250.4

Confidence: Medium

Time Horizon: 1-3 months

IV. Risk Management Team Decision

Aggressive Analyst

Aggressive Analyst: Neither of the other two analysts has spoken, so I'll take on the arguments I expect from them. The Hold is right, but I want to explain why it's an aggressive call and not a timid one.

Start with what the market has already done to MCD. The stock fell about 31% on a closing basis, from 334.50 to 230.88, while H1 2026 EPS grew about 6.3%, operating margin held near 47%, and H1 free cash flow rose about 18%. The multiple went from roughly 28x in February to about 20x trailing and under 18x forward. The stock has repriced from premium defensive compounder to something close to a broken story, while the earnings engine kept running. That is where asymmetry comes from. The 16.6M-share flush on 9/23 and the 11.4M-share reversal on 10/8 look like forced selling meeting real buyers. OBV has already retraced its late-September drop while price has recovered only about 74% of it. Money is coming back faster than the price shows.

Now the conservative case. I expect them to lead with the weekly and monthly SuperTrends being down, the falling 200 SMA at 282, and the 50 below the 200. All true, and the plan respects it by refusing to chase. But treat that as a reason to avoid positioning now and you'd only buy once the weekly stop at 265.86 breaks. That is about 8% above here and more than 10% above the 241 entry. By the time the higher timeframes confirm, most of the repair is gone. The monthly z-score of -1.78 and the 7-of-9 monthly buy setup say the longest timeframe is maturing to the downside. Trend followers will always be late, and the plan buys the pullback and manages risk with defined levels.

They'll also cite the insider selling, about $40.5M and no buys. But the sales came at $280 to $333, mostly after option exercises, and Erlinger's look like a scheduled plan. Nobody has sold since June, so no insider has sold near 245. The absence of buying is not confirmation of anything. It's a non-signal, and I won't build a thesis on it in either direction.

They'll point to the slowing U.S. comps, the affordability backlash, and the 97% odds of no Fed cuts. The comp slowdown is real, and Q2 revenue growth slowed to +3.7%. But "no cuts in 2026" at 97% is consensus, not news. The stock has de-rated into that backdrop, and an 8% recession probability means no demand collapse is priced in. The negative headlines are everywhere in the sentiment data, which tells me the bad news is crowded. A mixed print into a 17.8x forward multiple and a 3.1% yield could be a relief event. I'd also flag that the 1.15x dividend coverage and the PEG are rough derived numbers, not hard facts. I'm not leaning on the unverified Citi $295 target either, because it's retail hearsay.

The neutral analyst will say Hold is obviously fine and stop there. I'd push on that. Hold is correct because the daily TD 9, the +1.77 z-score, price pressed against the 246.41 upper band, and a flattening MACD histogram say the bounce wants a rest. The aggressive move is to let the market come to us. The 10 EMA at 241.24 and the Bollinger midline at 236.70 sit right where the 2x ATR stop reference lands, and the daily SuperTrend at 232.28 lines up with the 3x ATR stop. From 241 with the 229.20 stop, risk is about 11.80. A move to 257.49 pays about 1.4R, 265.86 about 2.1R, and a return toward the falling 200 SMA about 3.5R. That's a strong payoff for a low-beta name with a 47% operating margin.

Where I'd lean harder than the plan is the execution after the print. A third of a standard allocation is a fine first step, but if Q3 comes in mixed or better and guidance holds, I'd want the plan to scale quickly to a full allocation. The real risk is gapping through 241 on good news and sitting out the repair, and the plan should accept that cost without hand-wringing. The stops are already defined: cut half below 232.28, exit below 229.20.

Finally, I wouldn't ignore the data holes. We don't have the Q3 date, the comp number, guidance, debt, or interest coverage, and the 10/21 bar may be partial. That is exactly why the plan sizes small before the print and big only after the evidence arrives. Cutting risk on a quality franchise at the low would be selling the bottom of a de-rating. Adding before the print would be gambling on a number we haven't seen. Holding with a loaded limit plan, tight exits, and the willingness to press on a good print is the high-reward path. Aggressive Analyst: Let me give back what I should first, because you each landed a real hit.

Neutral is right that my R-multiples were measured from a 241 fill that only exists after the print. Conservative's were measured from 245 that nobody proposes to buy. Neutral is also right that "crowded bad news" was an overreach. A 5.2 at low confidence, with Reddit missing and 9 tagged posts out of 30, is close to no reading, not a washout. And the forward multiple is softer than I treated it. If the $13.75 is calendar 2026, then using the report's derived Q4 of about $3.03, the second half needs EPS growth of roughly 23%. So I'll drop "under 18x" and use the trailing figure instead. That is about 19.9x against roughly 28x in February, still about eight turns of compression while EPS grew.

That leads to Conservative's estimate-cut math, which I think cuts the other way. Even if a mixed print trims growth to 6% and forward EPS to about $13.05, the multiple moves from 17.8x to 18.8x. That is one turn, against the eight the market has already taken. I'm not calling this cheap, since the PEG of 2.57 says it isn't. My claim is narrower: the stock has already paid a lot for deceleration, so a mixed print has less left to take away than it did in February. The stops handle the case where I'm wrong.

On gap risk, Conservative's "size for the gap" is right for existing exposure, and I won't argue it. But for new money the plan already solves it. Nothing goes in before the print, so the earnings gap, and the kind of 4.8% day we saw on 9/23, happens while we're not adding. That is the case for being flat into the print.

Conservative's contradiction charge also misreads me. I never proposed buying a gap to 255 with a 229 stop. The limit sits at 241 or below, and "accept the gap cost" means don't chase. Neutral's fix is right, though: any tranche bought higher should carry a stop set against its own entry, not inherit 229.20.

On the limit filling "when the market is moving against you," that is what a pullback limit is for. Conservative says the short-term base case is a pullback toward the 10 EMA and the midline, because of the daily TD 9 and the flattening MACD. If you believe that, you want the order waiting there after a clean print, not chasing 246 to 250. His argument supports the entry.

On trimming existing exposure to half, I side with Neutral and add a point. The TD 9 and the +1.77 z-score sit on a bar with 2.57M shares against a normal 4 to 6M, so they're provisional. Selling half on that, with no exit trigger fired, means you may sell inside the 246 to 250 supply zone and then rebuy higher if the print is fine. The plan caps new money at half because we're blind. Existing money was underwritten earlier, and selling and rebuying costs more than not buying.

On the monthly 7-of-9, I wasn't using it for timing. It says the longest timeframe is maturing, not that a low is in. On macro, I agree rate relief isn't coming and the 8% recession market is thin. My point was that the multiple has already de-rated into no cuts, so it doesn't need relief, only no new shove. On FCF, Neutral is right that H1 against H1 compares four reported quarters, so it's firmer than the 1.15x coverage. I'll treat both as directional: cash flow is growing and the cushion is thin.

Where I still disagree is the ceiling. Conservative wants the half cap until the weekly SuperTrend turns. Neutral wants it lifted on a close above 257.49. Both gates sit in the supply zone, so most of the position would be built above 250. Conservative's own math shows what that costs: buying near 255 pays roughly 11 points to the weekly stop against 26 of risk. The weekly stop is also a moving target. By construction a SuperTrend ratchets down while the trend is down, so 265.86 can fall toward price over time. That is my inference from how the indicator works, not something the report states.

Most of what the half cap guards against, namely the Q3 date, debt, coverage, comps, guidance and the FCF reconciliation, gets answered by the release and the filing, not by price. So lift the cap when those checks come back clean, and let price confirmation govern only the adds above the pullback zone.

Concretely, using Neutral's "define it before you see it" test (margin near 46 to 47%, no guidance cut, nothing ugly on debt or coverage, and the daily SuperTrend at 232.28 holding after the reaction), I'd buy a third at 241. Then I'd put a second third near the 236.70 midline, with the stop at 229.20. A blended entry of about 238.85 puts risk near 9.65 points, below the 11.80 from 241 alone. From there 257.49 pays about 1.9R, 265.86 about 2.8R, and the 200 SMA a bit over 4R at today's level, less as it keeps falling. The final third waits for closes above 250.18 that hold a retest, with its own stop. If the stock never pulls back, I don't get filled and I accept that, which is the cost Neutral asks me to own.

Conservative closes by saying that if I'm wrong, I'm full-sized in a falling-200-day, lower-high stock. In this plan nothing is sized before the print. The position gets halved on a close below 232.28 and exited below 229.20, which is roughly a 4% loss from the blended entry. Missing a base also has a price, and a staircase that buys mostly above 250 pays it every time. I'm backing the Hold, because on current evidence it's the right call before the print. The difference is that I want the order book ready for the pullback after it. Aggressive Analyst: I'll take the hits first, because most of what Conservative and Neutral said about my ladder is right.

The drop from 11.80 to 9.65 points of risk was a per-share comfort number. The position was twice as big, so dollar risk to the 229.20 stop rose about 64%. The 236.70 slice also sat about one ATR above my own cut-half trigger at 232.28. A rule set that buys at 236.70 and trims at 232.28 argues with itself, so I'm dropping that bid. If I ever want a second bid below the first, it has to come out of the same risk budget, not on top of it.

I'll also drop "underwritten earlier." Neutral's right that it's sunk-cost thinking. The test for existing size is whether an 8% gap to the mid-220s is tolerable. If it isn't, trim the excess with limit sells in the 246 to 250 zone. The dividend hook is out too. About $1.92 a quarter doesn't matter against a stock that fell 4.8% in a day.

On the multiple, "eight turns" leaned on a February peak, and the report has no multiple history. A better anchor is that the stock was around 287 a year ago and is 245 now, down about 15% while trailing EPS is higher. Neutral's rough estimate of four to five turns of compression is fair. I'm not calling that a cushion, because an unexplained decline is a risk. My narrower claim stands: a mixed print has less left to take away than it did in February, and the stops cover me if I'm wrong.

The selection point is Conservative's best one against a limit at 241, but it overreaches. The bid isn't blind. It only goes live after the checks come back clean, and a close below 232.28 means I'm not in. A soft tape that holds the 10 EMA zone and 232.28 is information. The alternative isn't free of selection either. Buying strength into 250 to 257 means buying where everyone who bought near the 250.35 close on 9/22 is hoping to get out, with the weekly stop only 8 to 16 points overhead. The technical report calls a pullback to the 10 EMA or midline the base case. I'd rather pay for the expected move with a defined exit than for the unexpected extension at a worse price.

On gap risk, I take Neutral's point that fat tails cut both ways. Conservative's 1.4R only applies to the losers that gap, and the real control is sizing to the gapped loss, which is cheaper on a third than on a full position.

That leaves two small disagreements.

The first is what lifts the half cap. The plan set it at half because of the Q3 date, debt, coverage, and the FCF reconciliation, and those are all information gaps. The release and filing close them. So I'd move from a third toward half on clean checks plus a daily close holding above 232.28. Conservative is right that clean information isn't market acceptance, which is why I keep the price condition and don't jump straight to standard. Price gates should govern everything beyond half, including the step to two-thirds on a close above 250.18 and the step to standard above 257.49. That's about a sixth of a standard position of difference between Neutral's staircase and mine, and I'd rather say so than dress it up.

The second is the shape of the staircase. Neutral said the later tranches pay about 1.5R to the weekly stop and under 1R to the 257 shelf, so they buy confirmation and not payoff. If that's right, equal slices up the staircase are the wrong shape. The asymmetry sits in the first tranche, so on clean checks it should be a third and not a default quarter, and later tranches should be smaller. If the pullback never comes and the stock reclaims 250.18 and holds a retest, a quarter there is fine. I'll own the cost of missing the pullback.

Everything else we've converged on. Add nothing before the print and place no resting bids. Run the gap test on existing exposure. Define "clean" in advance: operating margin near 46 to 47%, no guidance cut, nothing ugly on debt, coverage, or the financing outflow, and U.S. comps not materially worse than the headlines imply. Anchor the post-print bid to the post-reaction 10 EMA and daily SuperTrend, not to this week's numbers. Give each tranche its own stop under its own retest low, and cap the total at standard until the higher timeframes change.

Look up the Q3 date first. McDonald's has historically reported somewhere between late October and early November, but that's my memory and not in our data. It decides whether "before the print" means days or weeks, and a stock that's pulled back to the 10 EMA two weeks before earnings is a much harder bid to leave unfilled.

I still back Hold. The plan stays flat while the key numbers are unseen, and it commits when the information arrives, not weeks later when price has done the confirming and the payoff is down to 1R. That timing is where the staircase and the ladder differ, and it's why I'd front-load the one tranche that has real asymmetry. Aggressive Analyst: Conservative, Neutral, you've talked me out of most of my ladder, so I'll say what I'm giving up, what I'm keeping, and the one thing I'd still fight for.

Conservative is right that a clean test built on operating margin doesn't discriminate. Margin sat at 45 to 47% all the way down and the stock still lost 31%. His EPS arithmetic holds, and I'll push it a step further. By the report's own quarters, 2025 EPS was about 11.95. Run-rate growth of 6.3% gives roughly 12.70 for 2026, which puts the stock near 19.3x, not 17.8x. So on a run-rate print the stock isn't cheap, and I'm done using valuation as the cushion. My case rests on defined exits and on what the tape tells us after the print, not on a low multiple.

I'll take Neutral's definition of clean as "no worse than run rate." A run-rate print earns a quarter. A result above run rate with a reaction that holds earns a third. That settles quarter versus third.

You're also both right about 232.28. It's where the plan cuts half, so using it as the gate to lift a cap is circular. And Conservative's 0.8R to the first shelf is correct, so I retract "real asymmetry." What's left of the first tranche's case is narrower. The stop is cheap relative to the entry, and it's bought into weakness after the information arrives, not into the 250 to 257 supply zone. The 229.20 stop is about 2.7 ATR from 241, already past Conservative's two-ATR floor. Neutral's amendment is right too: if the open-risk budget gets tight, shrink the add and don't tighten the stop.

I like the open-risk cap more than you might expect. Paying for adds by trailing earlier stops up is pyramiding, which is the aggressive trader's discipline: add to what's working, never to what's falling. Once the first tranche is validated by a close above 250.18 and a retest, its risk is mostly released and can fund the next tranche. I'd set that size by the gapped-loss test, not a fixed fraction. And if the stock never gives a retest, I accept missing it.

The premise check is the best contribution here, and I agree it goes first. My memory says late October and Conservative's says early October, so neither counts, and a calendar lookup settles it. If the print is already behind us, that helps the plan. The blind-spot reason for the half cap disappears, you read the release, and the first bid goes at the post-reaction 10 EMA now, with no waiting.

The 9/23 homework matters for the same reason. My "forced selling meeting real buyers" line was an inference, and I'll own that. If peers and the index fell that day, it gains some weight. If it was company-specific, "clean" has to answer that specific problem, and the first tranche drops to a quarter and waits for 250.18. I'd rather say now what would change my mind than discover it afterward.

I'd also rank the homework. The Q3 date and the 9/23 news change the plan. The 10-Q debt and coverage check and the FCF reconciliation are veto checks. I expect them to come back fine for a company like this, but that's an expectation, so run them. Re-running the daily readings once the 10/21 bar is final is routine. My worry is that this plan now has so many gates it can't be wrong, and so can't make money either. The point of the work is to act the day it's done.

Hold stands. Before the print, add nothing and place no bids. Existing size passes the 8% gap test or gets trimmed with limit sells in the 246 to 250 zone. What I'd still fight for is putting the first tranche in at the post-reaction pullback on the day the checks are complete, not after several more closes, and sizing it to the reaction. The cost of being wrong is modest. A third of a standard position stopped at 229.20 loses about 4.9% on that third, under 2% of a standard allocation, and roughly 2% even if it gaps to the mid-220s. The cost of the fully confirmed version is also known: buying near 251 with a stop around 243, and paying about 10 points of entry price for the shelf having cleared. Neutral says that buys probability, and I agree it buys some. I'd rather own both tranches, the cheap one first and the confirmed one second, than only the expensive one. Aggressive Analyst: Conservative, Neutral, most of what you added this round is right, so I'll take those points first and then say where I still disagree.

On timing, one full post-print close costs about a day, and I'll pay it. I was treating "the checks are done" as the trigger, but part of "clean" is whether 232.28 and the 10 EMA zone hold, and you can't know that intraday on reaction day. Neutral's single exit for new money is also correct. A cut-half trigger at 232.28 and an exit at 229.20 are 3.1 points apart, less than one ATR of 4.31, so for a fresh tranche they're one decision. I'd put one stop at structure, measure the two-ATR floor off the post-reaction ATR, and let a wider stop mean a smaller position. The staged cut stays for legacy exposure only. I also agree that beta of 0.45 shouldn't relax anything, since the stock lost 31% at that beta. The sizing language should be in loss units, because we can't see the caller's book. Run the gap test, and if an 8% drop to the mid-220s isn't tolerable, trim the excess with limit sells in the 246 to 250 zone. I'm also fine with the re-entry lock. After a stop-out, no new money until 250.18 is reclaimed and retested.

First disagreement: Conservative, I'd drop the hard $3.55 bar. Neutral is right that the market grades against consensus, and it's softer than that. We don't know what period the $13.75 covers. If it's calendar 2026, then with the report's derived Q4 of about $3.03, the second half needs about $7.65 against roughly $6.21 last year, which is 23% growth. Q3 would have to be near $3.90. If it's a forward twelve-month figure, $3.55 is only a loose proxy. A rule requiring a print to "clearly clear both numbers" could be one that no realistic print clears, and then the third-of-standard tier never fires. Use $3.37 and $7.34B to screen out bad prints, and let the reaction be the veto. A stock that's down 31% and holds 232.28 and the 10 EMA after a run-rate print is giving information no EPS threshold can.

Second, the re-entry lock makes the first tranche a one-shot, so don't make it a token. With a lock in place, an undersized first tranche pays the full cost of a locked-out regime and captures little if you're right. That's why I want a third when the print is above run rate and the reaction holds, and a quarter when it only meets run rate. The gap test caps both. Stopped at 229.20 from 241, a third of a standard allocation loses about 1.6% of the standard position. That's a cheap attempt, so make it a real one.

Third, a challenge to Neutral's cost-of-confirmation fix. A tranche near 251 with a stop around 243 is 8 points of risk, which is about 1.9 ATR at today's 4.31. That's under the two-ATR floor Conservative set, and I agree with that floor. So the confirmed tranche either gets a wider stop and a smaller size, or it has the same earnings-week whipsaw problem as an early entry. I think the staircase is still right, but the confirmed tranches are smaller and slower adds, not a substitute for the first one. The first tranche is the only one whose stop, about 2.7 ATR from entry, sits below structure instead of inside the 246 to 257 supply shelf.

I'd also state plainly what we give up. If the print is fine and the stock gaps to 255 or so and never returns to the 10 EMA, we're flat until a 250.18 retest, and then we hold a quarter. From 241 to the weekly stop at 265.86 is about 10%. From 251 it's about 6%. We forgo roughly four points of a ten-point move. I accept that price, and I'd rather say it now than rationalize it afterward.

Finally, the gates are only cheap if the homework gets done. The print date is the one fact that changes which branch we're on, and the 9/22 to 9/24 news check against peers and the index changes what "clean" has to answer. The 10-Q debt and coverage check and the FCF reconciliation are veto checks, so run them today. The date tells us how long we wait, not what we do. I said at the start that adding before the print is gambling on a number we haven't seen, and a distant print date doesn't change that.

So I'm at Hold on the same three live questions Neutral named. Which branch are we on, the print ahead or behind? Is existing size survivable at an 8% gap? Did the first post-reaction close hold 232.28 with the checks clean? If all three come back clean, we commit a real first tranche at the post-reaction pullback, with one exit and defined risk. We add only as price pays for it, and we stay capped at standard until the weekly trend turns. The edge here isn't being early. It's having every number defined before the information arrives, so we can act the day it does.