Nubby's Number Factory
BANGER · Roguelike · 2025
Steam Deck PlayableThe verdict
BUY IT.
A BANGER at $4.99.
- Length
- 3.1h
- Value
- $1.61/hr
- Reviews
- 97% positive
Watch this gameOne email when it is worth buying. Your call on what that means.
Get one email when Nubby's Number Factory is worth buying.
One email per game. Unsubscribe in a click.
Minimum on the Steam store: 4 GB of memory. Windows only.
A plinko-style roguelike where you work at a number factory. Launch your spherical friend, Nubby, down a pegboard to make bigger and bigger numbers! But beware, if you fail to meet your number quota, the sun will explode.
What drives the score
How to read this
Every input gets a grade, S to D, read from the buyer's side. The score is multiplicative, so an input only counts as raising when it sits near the top of its own scale. Anything with real room left is holding the score back, because that room is the gap to 100. ▲ marks a lift, ▼ a drag or a dock, ? an input we do not have, • a fact this genre's score does not use. The exact weights stay secret.
The value story
It has never been cheaper than $2.49. That was December 2025. Its cuts run 50 percent off as a rule. There have been 9 in the 2 years tracked, about 10 weeks apart. Today it is at full price. The last sale was about 5 weeks ago, and on this game's rhythm the next one tends to arrive within 10 weeks.
Value puts it in the top 4 percent of the games it competes with. It is cheaper than 86 percent of its genre, and better reviewed than 99 percent. Among the games that feel like it, Balatro ($14.99, scores 95) beats it on value, while it beats 天外武林 (Traveler of Wuxia) ($24.99, scores 62).
13 of the 780 Steam-purchase reviews sampled here come from buyers who refunded the game, sampled from reviews written March 2025 to August 2026. That is fewer refunded reviewers than 8 in 10 paid games this site has sampled to the same depth, and among the 355 games in its genre sampled this deeply, 301 have a larger share. Inside the two-hour refund window, 95 percent of the 2,282 reviews written there were positive, against 97 percent overall, and those are 13 percent of everything written about it.
HowLongToBeat players log 3.1 hours as a typical playtime, and completionists log 32, 10 times as much.
The drop-off
24% start to 2% finishMost players never reach the end. On a 3.1-hour game, that's a lot you pay for and may never use. Steam global achievements
Players
731 players at peak in the last day.
The chart appears once this game has been counted more than once.Player counts via SteamSpy.
What players say
this game is perfect EXCEPT for the fact that after 27 hours it reset my highscore from somewhere in the millions, down to 56 thouwow. edit: mustve been the wind cus its back
i dont believe any other game has caused me to ask for a pregnancy test, but here we are
it sucks but i can't stop playing. entire game boils down to "get one of the 5 items that gets you infinite points or lose on round 35"
Player reviews from Steam: the most helpful all-time, plus the newest that clear the same quality bar, with the hours each player had put in when they wrote it. The mix of Recommended and Not recommended follows the game's Steam ratio. Rotates daily. Opinions are theirs.
No data upstream: critic score (Metacritic).
Cite this score
Embed the live badge, copy the line, or download the card.
Nubby's Number Factory scores 88/100 on the BANG Score (a BANGER). The BANG Score rates what a game is worth for the money. Via Steam BANGERS. https://steambangers.com/game/3191030/nubby-s-number-factoryThe BANG Score is one number for value, 0 to 100: price against how it reviews and how long it was built to be. It answers the money question and stops there; whether the game is for you is still your call.
Reviews set the ceiling and price moves you inside it, so a cheap game that reviews badly cannot climb. Ratings from few reviewers are shrunk toward the middle, so 100 percent from nine people does not outrank 96 percent from ninety thousand. Length comes from HowLongToBeat where it can be matched, measured against a per-genre floor so a five-hour narrative game is judged against other narrative games rather than against a sixty-hour RPG. Where no verified length exists the score says so and carries a dock, because the biggest input is then a guess. The full method, including where it is weak → Steam reviews as of Aug 3, price as of Sep 23.
A number here look wrong? Tell me what it got wrong and I will look at the inputs behind it.