I worked with a GM once at a two-property operation... one hotel was a cash machine, the other was a project. Every Monday morning, the owner would look at the combined P&L and feel pretty good about life. And every Monday morning, that GM knew the truth: the strong property was masking the fact that the weaker one was slowly eating itself. The combined number was a lie they both agreed to believe.

That's what I see when I look at Las Vegas Sands right now. Marina Bay Sands in Singapore just threw off $788 million in adjusted property EBITDA in a single quarter. A 53% operating margin. From one building. That is a staggering number... roughly $8.6 million a day in EBITDA from one integrated resort. Meanwhile, the five Macao properties collectively generated $633 million in EBITDA, with margins that actually compressed year-over-year... 29.9% versus 31.3% a year ago. Five properties generating less EBITDA than one, with shrinking margins. And the CEO says the plan in Macao is to spend more aggressively on customer incentives to chase market share. That's not a margin improvement strategy. That's a volume play dressed up in margin language, and anyone who's ever run a hotel knows the difference.

Here's what's really happening. Macao's gross gaming revenue has plateaued around $28 billion annually. The junket business that used to drive premium play is essentially gone. Online gambling is siphoning off the casual customer. Visitor spending habits have fundamentally changed. So Sands is doing what operators always do when the market shifts... they're buying revenue with promotional spending. That can work. I've seen it work. But let's not pretend that "being more aggressive with customer incentives" is a path to higher percentage margins anytime soon. The CEO has essentially said as much publicly... the bet is that higher absolute EBITDA (more total dollars) justifies thinner percentage margins. That's a legitimate strategy. Just don't call it a margin improvement story when the margins are moving in the wrong direction.

The Singapore side is fascinating for a different reason. Sands is pouring $8 billion into expanding MBS... a fourth tower, 570 luxury suites, a 15,000-seat arena, more convention space. They took on a $9 billion loan to finance it. The property already runs at 94-99% occupancy with full exhibition halls, so the demand case is real. But here's the thing that should make every operator's antenna go up: when you take a property running at 53% margins and you add $8 billion in development cost, your breakeven math changes dramatically. The expanded MBS will need to generate substantially more revenue just to maintain its current return profile once that debt service kicks in. The current MBS is the single most profitable hotel-casino asset on the planet. The expanded MBS is an $8 billion bet that lightning strikes the same spot twice but bigger. History suggests that's harder than it sounds.

What I'm watching is the gap between the narrative and the numbers. The consolidated picture looks great... $3.59 billion in revenue, up 25%. The stock took an 8.6% hit after earnings anyway because Wall Street did what Wall Street does... it looked past the Singapore headline and found the Macao margin compression underneath. LVS repurchased $740 million of stock in the quarter while carrying $16 billion in weighted average debt. They're simultaneously expanding, buying back shares, paying dividends, and trying to fix a market (Macao) where the structural dynamics have fundamentally shifted. Something in that equation eventually has to give. The question is whether Singapore can keep throwing off enough cash to fund everything else. Right now, it can. The word "right now" is doing a lot of heavy lifting in that sentence.

LVS reported $3.59 billion in Q1 2026 net revenue, up 25.3% year-over-year, with consolidated adjusted property EBITDA of $1.42 billion. EPS came in at $0.85 against estimates of $0.76 to $0.78. Singapore delivered $788 million in property EBITDA on a 53% margin. Macao contributed $633 million, up 18%-plus. Those are the numbers every analyst led with. They're not the numbers I'd lead with.

The number I'd lead with is $5.24 billion. That's what LVS has spent repurchasing its own stock since Q4 2023, retiring 109 million shares at an average price of $47.95. In Q1 2026 alone, they bought back $740 million at $56.64 weighted average. They've eliminated 14.3% of their outstanding float in roughly two years. Meanwhile, Q1 capex came in at $194 million against an expected $336 million. A company spending nearly four times more on buybacks than on capital expenditures in a quarter is making a statement about where it sees the better risk-adjusted return... and it's not in bricks and mortar right now.

That calculus gets more interesting when you decompose the balance sheet. $3.33 billion in unrestricted cash against $15.57 billion in total debt. Net leverage is elevated. The $8 billion Marina Bay Sands expansion won't generate revenue until 2031. Macao property refreshes (starting with room product at one of their flagship properties, targeting completion by end of 2027) will, as CEO Patrick Dumont acknowledged, "naturally increase expenses" and "continue to negatively impact margins" near-term. So you have a company carrying significant debt, committing to multi-year capital programs on two continents, absorbing near-term margin compression from reinvestment... and simultaneously buying back stock at the most aggressive pace in its history. The implied conviction is that the stock at $56 is still cheap relative to what these assets will produce at stabilization.

The Singapore story is straightforward. $788 million EBITDA on a 53% margin in a market projecting record tourism receipts of S$31-32.5 billion in 2026 with 17-18 million arrivals. That's a mature, high-performing asset in a structurally supply-constrained market (Singapore has exactly two integrated resort licenses). The expansion adds capacity into proven demand. Macao is the variable. Analyst projections for 2026 GGR growth range from 3% to 6%, mass and slot driven, with total GGR still 10-15% below pre-pandemic levels due to VIP regulatory constraints. LVS is targeting $700 million in quarterly Macao EBITDA "over time" (a phrase I've learned to stress-test). Current run rate is $633 million. Closing that $67 million gap while margins compress from reinvestment requires meaningful revenue growth. The mass market share hit 25.7% in Q1, strongest since Q1 2024. That trajectory matters more than the absolute number.

The question for anyone analyzing LVS as a proxy for Asian gaming recovery: is the buyback pace sustainable if the Macao margin story takes longer than projected? $740 million per quarter in repurchases plus $194 million in capex plus debt service against a cash position that, while substantial, isn't infinite. If Singapore stays at current levels and Macao grows 5% annually, the math works. If there's a demand shock (regulatory, macro, geopolitical), the company is buying back stock at $56 that it may wish it hadn't. I've analyzed portfolios where management's conviction in buybacks turned out to be correct and portfolios where it turned out to be expensive. The difference is almost always whether the underlying asset thesis holds through a stress scenario... and LVS hasn't been stress-tested at this leverage level with this capex commitment yet.