Robux is the one thing a game directory can never see. There is no public field anywhere in Roblox's own data for "how much money has been spent inside this experience", and the BloxBus directory does not invent one. What follows is therefore not a spending ranking. It is an honest map of the surfaces where spending happens, built only from figures the directory actually records for its 76 fully-synced games.
That distinction matters, because the two get conflated constantly. A game with 95,925,241 favourites has not necessarily earned more money than a game with 24,667,599 favourites. Favourites are free to give. But favourites are also the clearest voluntary signal a player can leave, and the games that collect the most of them are the games where cosmetic purchases, game passes and private servers have the largest audience to sell into. That is the useful claim, and it is the one this article will make.
Every fully-synced entry in the directory carries the same set of counters: total visits, current concurrent players, favourites, upvotes and downvotes, plus links to whatever supporting pages the site has published for that title. Across 76 games that adds up to 681,196,761,148 lifetime visits, 3,281,405 people playing at the moment of the sync, 808,813,313 favourites, 222,859,364 upvotes and 25,767,849 downvotes.
Not one of those five counters is a purchase. Visits are a launch count. Concurrent players are a snapshot taken at one instant. Favourites are a bookmark. Upvotes and downvotes are an opinion poll with an approval rate of 89.6% across the whole directory. Robux appears nowhere in that list, and any article that tells you otherwise is guessing. What you can do instead is read each counter as a proxy for a different part of the spending surface, and then be explicit that it is a proxy.
Of everything the directory records, favourites cost the player the most deliberate effort. Visits happen accidentally - a friend sends a link, a front-page tile loads. Favourites require opening a game's page and pressing a button for the specific purpose of returning later. That makes the favourite count the best available stand-in for "how many people have decided this experience is worth their time", which is the precondition for ever spending anything in it.
| Game | Favourites | Live now | Total visits | Approval |
|---|---|---|---|---|
| RIVALS | 95,925,241 | 101,373 | 18,965,357,690 | 93.7% |
| Brookhaven 🏡RP | 30,683,521 | 255,765 | 88,909,886,510 | 85.9% |
| [🌗] Steal a Brainrot | 29,630,004 | 280,486 | 74,163,901,688 | 85.1% |
| Murder Mystery 2 | 24,667,599 | 263,963 | 31,834,248,204 | 90.8% |
| Blox Fruits | 20,128,281 | 108,380 | 65,103,824,447 | 92.3% |
RIVALS is the clearest illustration. It holds 95,925,241 favourites - 11.9% of every favourite in the directory - while sitting at 101,373 concurrent players and 18,965,357,690 lifetime visits. A game with that many bookmarks relative to its visit count is a game people intend to come back to, and returning players are the ones who buy. Compare it with Steal An Egg, which leads the directory in concurrent players at 768,030 but has collected only 8,725,312 favourites - roughly 11 times fewer. Huge live traffic and comparatively modest bookmarking usually means discovery-driven play rather than a committed audience, and the monetisation surface looks very different in those two cases.
The simulator category in the directory contains several titles where this gap shows up sharply, because simulator loops are built around long sessions and incremental progression - the exact shape that both attracts favourites and supports paid upgrades.
Upvotes and downvotes are free too, but unlike favourites they carry a direction. Across the whole directory players have cast 248,627,213 votes and 89.6% of them were positive. Individual games move around that average in ways worth reading before you spend anything.
Blox Fruits carries 12,792,614 upvotes against 1,068,456 downvotes, a 92.3% approval rate on one of the largest vote totals in the directory at 65,103,824,447 visits. [🌗] Steal a Brainrot sits lower at 85.1% (15,482,293 up, 2,703,651 down). Neither number tells you what is being sold, but a large, polarised vote count is a reasonable signal to look at the in-game shop carefully before committing anything, and to check whether the complaints are about the monetisation rather than the gameplay.
Everything above is inference. One Robux cost leaves a footprint the directory can see directly: private servers. 11 games currently have a dedicated private-server page published on the site - [🌗] Steal a Brainrot, Blox Fruits, Brookhaven 🏡RP, Murder Mystery 2 and others - and a private server is by definition a paid, recurring subscription rather than a one-off cosmetic. Of the 76 games with complete statistics, that means 14% expose at least one visible subscription-style purchase, while the rest either do not sell one or have no page tracking it.
If you want to know what a game will actually cost you per month, start there rather than with the cosmetic catalogue. The private server page for Brookhaven is the kind of page that spells out what the paid tier unlocks, and the equivalent page for Murder Mystery 2 shows that two games in the same genre can treat private servers completely differently.
Promotional codes are the counterpart to all of this. 22 of the 76 fully-synced games have any code list at all, the rest have none, and across every page the directory currently renders 54 working codes and 0 retired ones. Blox Fruits is the outlier with 24 working codes on its own dedicated code list for Blox Fruits.
Read those two facts together. A game can publish a long free-code list and still sell heavily through game passes; a game can publish nothing free and sell nothing at all. Code count is not generosity and it is not a measure of spending. It is simply the free end of a catalogue you should inspect end to end.
Start from the approval rate rather than the headline player count - an 89% directory average means a game at 90.8% like Murder Mystery 2 is broadly liked, while a game near the bottom of that range deserves scrutiny. Check whether a private-server page exists, because that is a monthly charge rather than a single purchase. Read the code list to see what is given away free. Then check the Rivals guide and similar genre guides for what is actually locked behind payment. Finally, remember that none of these figures record purchases: they tell you where the spending surface is, not how much money crosses it.
That is the honest limit of what a public statistics panel can tell you about Robux. For a fuller walkthrough of what the currency buys and does not buy, see our explainer on what Robux actually buys.