Vivenso Product 06 — a tool for one player community, not a product
WoS
Every cycle the same argument, in chat, from screenshots and memory: why we lost the last battle, who was holding which buff, and what time everyone has to be online. WoS is where one state in Whiteout Survival — the mobile strategy game its players shorten to WoS — writes that down instead.
Why it exists
Everyone remembers the fight differently.
A state in this game fights as a group, on a schedule, and the group has the same rows over and over: who is actually strong enough to lead a hit, what decided the last defeat, who was holding which buff, when everyone has to arrive. All of it settled from memory, from screenshots pasted into a chat app, and from whoever sounded most certain.
So WoS keeps the record. Battles go in with their stat panels attached. Members are ranked by a weighted score over that record rather than by the one power number the game shows. The schedule and the state calendar sit in one place, and the planning that used to live in somebody’s private spreadsheet is shared.
Written for players rather than administrators: seven languages, and a sign-in a member can get through on a phone without an email account.
Unofficial, fan-made, and not endorsed by anyone
WoS is an unofficial tool made by and for the players of one in-game state. It is not affiliated with, endorsed, sponsored or approved by the publisher of Whiteout Survival. Whiteout Survival and the in-game names the app refers to are the trademarks and copyright of their respective owners; nothing here claims any right in them. The name WoS is the players’ own shorthand, not a brand of this company.
The model
How the ranking works, and where it gets things wrong.
It is a weighted score over the twelve combat statistics, with troop quantity scaled non-linearly. The first version was a plain weighted sum and it fell apart on lopsided armies, which is why it was replaced.
Four ways it is known to be wrong, each written down next to the score with what the app does about it:
- Lopsided compositions. A near-absent troop type still carries weight in the arithmetic while being irrelevant in the fight. Those observations are flagged rather than scored quietly.
- Unknown buff state. Buffs move results more than raw strength does, so buffed and unbuffed observations are read as separate populations instead of averaged into one ranking.
- Where the fight happened. Field and garrison engagements account for losses differently, so the reported loss and the permanent loss are not the same figure and are not treated as one.
- Baselines nobody confirmed. Automatically generated baselines may already include buffs. They carry a flag saying they are not canonical, and the flag is checked before they are used.
The same goes for the conclusions. The strongest pattern in the logged battles so far is that buff state decides more fights than stat advantage does — but the sample is far too small to call that a finding, so it is carried as a hypothesis to keep testing.
Honest state
Running for its community. Parts of it are still placeholders.
It gets used every cycle, and it is unfinished in places you can see from inside it.
-
The battle record and the combat ranking
Observed engagements with their stat panels and special bonuses, and a weighted score over them that ranks members on what the record supports rather than on the power number the game displays.
In use -
Shared planning: the schedule, the leaderboards and the calendar
A submission and review flow for the recurring state-versus-state cycle, with a staff side for review and locking, plus power leaderboards and the state calendar.
In use -
Seven languages and a sign-in built for players
English, Polish, Turkish, Spanish, German, Portuguese and Arabic, chosen before sign-in from the device language so a non-English player is not met with an English wall. Members identify themselves by in-game name; only staff accounts use email.
In use -
The pet levelling planner
A cost model for a long, expensive in-game progression, built from community reference data and corrected against measured outcomes when the two disagreed. Owner-gated while the model is still being checked.
Being validated -
Rally timing, new baseline entry, state rules
Placeholders in the live app. They are reachable, they say what they will be, and they do nothing yet — which is stated in the app rather than hidden behind a broken screen.
Not built -
Any public or commercial version
There is none, nothing here is for sale, and there is nothing to sign up for. It was written for one group of players, one of whom wrote it.
Not a product
Limits, stated
What it does not try to be.
- Not connected to the game. It reads nothing from the publisher and automates nothing inside the game. Everything in it was entered or observed by a person.
- Not a general-purpose tool. It is shaped around one state, its alliances and its calendar. Nothing was generalised for an audience that does not exist.
- Not a predictor. The model ranks and explains observed engagements. It does not tell anyone they will win.
On data
No player name, roster, battle record, screenshot or in-game figure from the app appears on this site. What is described above is purpose, method and honest state.
Built by Vivenso Labs