Xenia posted $0.45 in Adjusted FFO per diluted share for Q4 2025, a 15.4% year-over-year increase on $265.6 million in revenue. The Street expected $0.04 EPS. They delivered $0.07. Same-Property RevPAR grew 4.5% to $176.45. None of that is the interesting number.

The interesting number is 44%. That's non-rooms revenue as a share of total revenue. Food and beverage alone grew 13.4% for the full year. In an industry where most lodging REITs generate 70-80% of revenue from rooms, Xenia is running a fundamentally different mix. A 44% non-rooms contribution means the per-occupied-room economics look nothing like a typical upper-upscale portfolio. It also means the cost structure looks nothing like one. F&B at 13.4% growth requires bodies... servers, cooks, banquet staff. Wages and benefits are guided to grow roughly 6% in 2026. That's the tension hiding inside an otherwise clean earnings print.

The 2026 guide tells the real story. Same-Property RevPAR growth of 1.5% to 4.5% against a 4.5% increase in operating expenses. At the midpoint, that's 3% RevPAR growth versus 4.5% expense growth. Run the flow-through math on that spread and you get margin compression unless non-rooms revenue fills the gap. Management is explicitly betting it will. Adjusted FFO per share is guided to $1.89 at the midpoint, roughly 7% above 2025. That 7% FFO growth on 3% RevPAR growth implies the non-rooms engine does all the heavy lifting. It's a plausible thesis. It's also a thesis that breaks if group demand softens or if F&B labor costs accelerate past 6%.

Capital allocation is where the discipline shows. The Fairmont Dallas disposition at $111 million avoided an estimated $80 million in near-term CapEx and generated an 11.3% unlevered IRR. That's a sell decision that most REITs wouldn't make because the asset looks fine on a trailing NOI basis. But trailing NOI doesn't capture the CapEx cliff. Xenia looked at the forward capital requirement, compared it to the disposition proceeds, and chose liquidity. They also repurchased 9.4 million shares at a weighted-average price of $12.87 while the stock now trades near $16. The buyback math works (so far). The $25 million land acquisition under the Hyatt Regency Santa Clara to eliminate lease renewal risk is the kind of quiet, unsexy move that adds real long-term value and never makes a headline.

One thing to watch. Director Barry Bloom sold 151,909 shares on February 26 at $15.73, reducing his position by 90.89%. Insider sales have a thousand innocent explanations (diversification, tax planning, estate planning). A 91% reduction in position two days after an earnings beat has fewer innocent explanations than a 10% trim. I'm not drawing a conclusion. I'm noting the data point. Check again when Q1 results hit May 1.