Votes are not players: what toplists actually measure

September 20, 2026

Comprehensive Status Tracker, Emulation History & Development Directory

← All posts

analysis

By WoW Tracker Editorial ·

The top three World of Warcraft servers on xtremetop100 read 8,028, 8,027 and 7,990 votes. Three servers, separated by 38 votes out of 8,000, on a list that resets monthly. That clustering is not what organic preference looks like. It is what organized voting looks like.

The paid layer, stated plainly

The toplist's own advertise page sells Gold Membership: featured banner, gold title and badge, priority visibility, and a bonus +100 votes monthly after the reset. The top of the current chart is wall-to-wall gold badges. This is not an accusation of fraud; it is the documented business model. But it means the leaderboard confounds two variables (player enthusiasm and marketing spend) into a single number.

Add the mechanics: one vote per visitor per 12 hours, full reset monthly. That system does not reward the most-played server. It rewards the server with the most organized Discord willing to click twice a day in a sprint window.

Votes vs heads, with our receipts

Our tracker checks populations, not votes, and the two series barely correlate:

None of this makes toplists useless. They are excellent discovery feeds: roughly two-thirds of the serious top-50 overlaps our tracker, and it is how we found GemstoneWoW; they are terrible censuses. Read them for names, never for numbers.

How we read them

Votes answer "who campaigned best this month." Population answers "who will be online Tuesday night"; our census uses ladders, counters, API telemetry and Discord presence, each labeled with what it actually is. When a server publishes its own ladder, like GemstoneWoW and Centurion do, that number beats any vote total ever printed.

SOURCES: xtremetop100 WoW toplist and advertise page (Sep 20); GemstoneWoW public ladder; tracker census methodology.