UCLA Housing Rush: How 17,500 Beds Vanished in Two Weeks

Tracking the rapid disappearance of UCLA housing availability during the 2025 enrollment period

Data analysis and visualization | February-March 2025

Housing sign-ups can be disorganized and stressful. Some students have aspirations to live in a specific housing arrangement, such as a university apartment, with their friends, only to find that all the options were taken by their sign-up time. To many, the housing sign-ups feel like a chaotic lottery.

Starting at 9 AM on February 18, 2025, UCLA opened its housing portal for the upcoming academic year. Within seven hours that first day, over 3,000 bed spaces disappeared. By the time the dust settled two weeks later, more than 70% of UCLA's 17,514 available housing spots had been claimed.

About the Data

The author scraped UCLA's housing availability spreadsheet every hour it was updated for the two weeks the portal was open. The geographical data was collected by manually labeling buildings on Google Earth, and data regarding amenities for each housing option was provided by the UCLA housing website.

The Critical 50% Threshold

The tipping point came faster than many students expected. Between 11 AM and noon on February 20—just two days after applications opened—UCLA housing crossed the effective 50% capacity mark. Students who hesitated past that Thursday morning found their options significantly narrowed.

The pace was relentless during those first four days. From February 18-21, availability plummeted from 17,514 beds to just 7,254—a loss of nearly 10,300 spaces at an average rate of more than 100 beds per hour during business hours.

University Apartments: The Early Casualties

While on-campus residence halls offered students some breathing room, university apartments told a different story. Of the 12 apartment complexes analyzed, 10 reached 80% capacity within the first week, and 7 filled completely.

The fastest to fill:

• Landfair Vista Apartments: 80% full by 11 AM on Day 1, completely filled by 3 PM
• Landfair Apartments: Hit 80% capacity by 11 AM on Day 1, gone by 4 PM
• Glenrock Apartments: Reached 80% by 1 PM on Day 1, filled by 10 AM on Day 2

This pattern reveals a counterintuitive finding: the newest, most modern buildings were among the last to fill. Laurel, Tipuana, and Palo Verde—all built in 2022—took nearly a week to reach 80% capacity, while the oldest building, Landfair (built in 1956), filled within hours. Statistical analysis confirms this age paradox: a strong positive correlation (0.81) between building age and fill speed shows that newer apartments filled slower, not faster. This suggests students prioritized location over modern amenities—the older buildings happened to be closer to campus.

By contrast, only 5 of the 14 on-campus residence halls reached the 80% threshold during the tracked period, and none filled completely. Students targeting traditional dorms had significantly more time to deliberate.

Key Drivers of Demand: Distance vs. Amenities

For university apartments, distance from campus proved to be the dominant factor in filling speed. Statistical analysis revealed a moderate positive correlation (Spearman correlation: 0.56) between proximity to campus and how quickly apartments filled—meaning closer apartments filled faster. The pattern was clear: Landfair Vista, Landfair, and Glenrock apartments, all within a few blocks of campus, vanished almost immediately. Meanwhile, Laurel and Palo Verde, situated farther from the academic core, took nearly a week to reach 80% capacity.

However, initial analysis suggested other factors might be at play. Multivariable regression was used to isolate the independent effects of distance, building age, room density, and amenities like parking, air conditioning, and exercise rooms.

The Parking Factor

Parking initially appeared to be a major driver of demand, with a regression model showing an R² of approximately 0.56. However, when controlling for distance in a multivariable regression, parking's effect became statistically insignificant (p=0.67). This revealed that parking was actually a proxy for location—the buildings closest to campus happened to have parking, but students weren't chasing cars; they were chasing proximity. The apparent parking preference was simply correlation, not causation.

Interestingly, on-campus housing showed a weak negative correlation (-0.30) with distance to the campus centroid. This suggests that for residence halls, other factors—possibly social reputation, building amenities, or community culture—played a larger role than mere proximity to classes.

Fastest Selling Unit

The 2 Bd+Loft/6 Person configuration emerged as the "unicorn" unit, filling completely by 10 AM on Day 1—just one hour after the portal opened. This room type was the first to reach both 80% capacity and 100% capacity across all housing options, suggesting it struck an optimal balance of space, affordability, and location that made it irresistible to students.

Methodology

Data was collected hourly from the UCLA housing portal during the enrollment period from February 18 to March 4, 2025. The analysis tracked 17,514 total bed spaces across 14 on-campus housing locations and 12 university apartment complexes. Availability data was normalized to show percentage remaining to allow for fair comparison between facilities of different sizes.

To isolate the independent effects of various factors, multivariable ordinary least squares (OLS) regression was employed. This statistical technique allowed us to control for confounding variables—for example, determining whether parking truly influenced demand or was merely correlated with proximity to campus. By including distance, building age, room density, and amenities (parking, air conditioning, exercise rooms) as simultaneous predictors, the regression model revealed that distance was the primary driver, with other factors serving as proxies for location rather than independent preferences.