Balance
Which Weapons and Items Do Players Really Use?
Compare how often each option was picked with how often it was on offer. Pick counts settle fast, win rates do not. Low use has three common causes.
Win Rates and Balance: Is It Really Too Strong?
A win rate means something only with its range. Check each side against a target, widen the ranges for the number of sides, and ignore a few games.
Game Economy: Sources, Sinks and What Piles Up
Add up what comes in, what goes out and what piles up, hour by hour. If coins pile up with nothing to buy, add things worth buying before you cut rewards.
Win rate check
Your AI works out how often each weapon, class or side wins and how sure that is given how often it was played, and answers too strong, too weak, fits, or cannot tell yet. It says how many more plays would settle it, and shows how often each option gets picked when it is on offer.
Economy check
Your AI adds up what comes into your game's economy, what goes out and what piles up, per hour of play and over a session, and shows where the coins come from and what they are spent on. It also catches money that appears from nowhere, which is usually a bug.
One class wins 52 percent, so is it too strong or just lucky in its matchups?
In a game where classes counter each other, an average win rate hides the pairs that are really unfair, and games of a class against itself blur it further. This entry shows which pairings to look at and which classes truly dominate.
How much damage does a build deal in a minute with cooldowns and mana?
Listing each ability's damage tells you nothing about a fight, because cooldowns and a limited resource decide what can be cast. A short simulation of the best sensible play gives a single number per build to compare.
Which place to farm pays far more than the rest?
Players soon find the one spot that gives the most for their time, and then do nothing else. Calculating the expected reward per minute for each spot, from kill speed and drop tables, finds it before they do.
Players are swimming in gold: where does it come from and what takes it away?
When everyone is rich, or no one can afford anything, the cause is almost never one big thing. Listing each source and each drain with its size shows what to turn down or up, and whether the balance is heading anywhere dangerous.
A character wins 54 percent of matches: overpowered, or just a lucky streak?
Every character will look a bit too strong or too weak in some sample. Before you change anything, ask whether the number is likely to be chance, and remember that checking many characters at once makes false alarms more likely.
Find the items that cost too little or too much for what they do
When a shop has dozens of items it is hard to see which ones are a bargain and which are a rip-off. Drawing the line that the prices roughly follow shows the ones that sit far off it, so you only have to look at those.
Which character builds do players drop soon after choosing them?
A build that is picked a lot can still be a disappointment, because players change away from it a few levels later. This entry shows how to measure the drop-out per build fairly, without being fooled by players who simply stopped playing.