As of 2026-07-17, Twenty Twenty-Five leads the WordPress.org popularity ranking, followed by Hello Elementor and Astra. But Twenty Twenty-Five ships bundled with WordPress core, so part of its lead reflects bundling rather than deliberate choice — and that’s only the beginning of how strange this list really is.
If you’ve ever watched the WordPress.org “Popular” themes page from one day to the next, you’ve probably noticed odd patterns. A theme you’ve never heard of appears in the top 20. A few days later it’s gone. Sometimes five themes from the same author drop fifteen positions on the same day. Established themes with millions of installs sit comfortably at the top, but the order around them is volatile in a way that doesn’t map onto how widely those themes are actually used.
There’s a reason for that. The ranking isn’t measuring popularity in any everyday sense of the word. It’s measuring something much narrower — and the formula is simpler than most people assume.
The formula
Behind the “Popular” list is a single calculation, computed once a day and cached against each theme as a meta value:
popularity = active_installs / days_since_publishedThat’s it. Active installs divided by the number of days the theme has been in the directory. The result is a per-day rate — installs accrued, on average, per day of existence. The list is then sorted by that number, highest first.
This was confirmed publicly by Dion Hulse (dd32), one of the meta-team developers who maintains the wp.org infrastructure. The calculation itself lives in a private stats-generation script rather than the open-source theme directory plugin, but its use is visible in the public code: the popular browse query simply sorts by a precomputed _popularity meta value of type DECIMAL(20,10).
The two-week rule
There’s one additional filter that is in the open-source code, and it’s worth knowing about. From query-modifications.php:
Only include themes that have existed for at least 2 weeks into the popular listing. This avoids cases where a new theme skews our popularity algorithms.
A theme uploaded today won’t appear in the Popular list at all — not until day 15. The reason is mathematical: the formula divides by days_old, so a theme with 50 installs and 3 days of age would otherwise compute to a popularity score of 16.7 — higher than many established themes. The two-week floor prevents that kind of arithmetic accident.
The 5× gap hiding inside “1M+ installs”
Here is what the formula does to today’s top 10, recomputed from publicly available data:
The three green bars are all themes that wp.org displays as “1M+ active installs.” Their popularity scores are 1,703, 385, and 302. The bar of Twenty Twenty-Five is more than five times longer than Astra’s, despite both being in the same install bucket — and Astra is, by any normal measure, one of the most widely-used WordPress themes ever built. The only thing separating them is age.
The same pattern explains why a small, young theme like Hello Biz (100K+ installs, less than two years old) outscores OceanWP, which has five times the installs but a decade of age behind it.
What this actually measures
Once you’ve internalised the formula, a lot of the strangeness in the Popular list resolves into the same explanation: this isn’t a popularity ranking. It’s a growth-rate ranking. The brand-new theme and the established 50K-install theme are nearly tied. That’s why a small, recently-released theme with strong early adoption can briefly outrank themes that have been around for a decade. It’s also why such themes usually don’t stay there long: the denominator keeps growing, so unless installs grow at the same rate, the score declines every day.
The mega-themes at the top — Twenty Twenty-Five, Hello Elementor, Astra — sit on numerators large enough that no realistic denominator can dislodge them. That’s the floor of the list. The 20-to-100 range, on the other hand, is intensely sensitive to recent install-count adjustments.
Why default themes have a structural advantage
WordPress’s bundled themes (currently Twenty Twenty-Five, with Twenty Twenty-Four and Twenty Twenty-Three behind it) get a unique boost: every fresh WordPress installation activates a default theme, and those activations are recorded as active installs. The bigger WordPress’s installed base, the larger the install count, regardless of whether anyone chose the theme deliberately. Combined with the popularity formula, this gives default themes a near-permanent place at the top — until they’re rotated out and start ageing without fresh activations to feed the numerator.
This is not how plugins are ranked
It’s tempting to assume the Popular plugins list works the same way. It doesn’t. For plugins, the plugin directory code sorts by raw active-install count — no per-day division. A plugin that’s been around for ten years and has five million installs stays at the top.
The search interface on wp.org (typing a query into the plugin or theme search box) is a different system again — it uses Elasticsearch and combines multiple signals: download counts, support topics, reviews, the freshness of the listing, and so on. WordPress.com’s data team published a detailed write-up of the plugin search algorithm in 2017 that’s still mostly accurate. It’s a much more sophisticated system than the theme popularity ranking, which is, essentially, one ratio.
What this means for theme authors
If you’re publishing themes, the practical implication of the formula is that consistent adoption matters far more than peak installs. A theme that gains 50 users a week, every week, will rise on the Popular list as long as its install count keeps growing faster than its age. A theme that hits 10,000 installs and then plateaus will steadily lose ground as its denominator increases.
It also means new theme launches benefit disproportionately from a strong opening fortnight — that’s the window in which the score is most leveraged by a small numerator and a small denominator. Once a theme is in the top 50, retention requires sustained acquisition; it isn’t passive.
A useful current example: Online Newspaper, a news theme published by BlazeThemes on May 28, 2026. By late June — at around 700 active installs and roughly 26 days old — its popularity score worked out to about 27. That was enough to land it on the second page of the wp.org Popular list, sitting next to themes with tens of thousands more installs but a decade of accumulated age behind them. The 700 installs figure isn’t surprising for a brand-new theme nobody has heard of yet. The position is what’s surprising, and it’s the formula doing exactly what it was designed to do.
Whether the theme stays there depends entirely on whether new installs continue to accrue faster than the denominator grows. For most new themes they don’t — and the score decays a little every day until the theme drops back out of view. That’s why the second half of the Popular list looks so volatile from one week to the next: it’s mostly themes briefly catching this wave, then falling off it.
What this means for users
“Popular” on wordpress.org is not the same as “widely used.” It’s closer to “growing fastest right now, weighted by accumulated installs.” If you want a sense of which themes are actually live on the largest number of real websites, you need a different metric — for example, BuiltWith’s crawl-based theme detection across the public web, which often paints a very different picture (and includes commercial themes that aren’t in the directory at all).
We track both lenses on this site, day by day: the WordPress.org popularity ranking and the global real-world usage. The gap between them is sometimes the most interesting story of all.
Discover more from WP.MD
Subscribe to get the latest posts sent to your email.