Difficulty
Learning paths through this topic
Pick a recorded opponent for a score duel that the player can beat about six times in ten
In duels against saved scores of other players, a match that is too easy or too hard feels pointless. This entry shows how to use the player's own recent scores to estimate the chance of winning against each candidate, ties included.
Is a level too long for the short sessions your players actually have?
A level can be perfectly balanced and still lose people because it does not fit into the minute they had. This entry shows which two numbers to put side by side so that length is judged against real habits and not against taste.
Find where a merge or match board fills up so far that players give up
On a board that fills as you play, the stress comes from running out of room, not from the move count. This entry shows how to measure how often and how long the board is nearly full, and whether players manage to recover from it.
Are failed attempts on a casual level close calls or hopeless runs?
Two levels can fail equally often and feel completely different: one is a frustrating near miss, the other a wall. This entry shows how to measure how close failures were, including levels with several goals where the obvious average lies.
Do players come back after running out of lives, or is the wait where they quit?
A waiting timer after the last life is a hard moment in a casual game. To judge it you must count who returned, but also be careful about people whose follow-up time has not run out yet. This entry shows the group you may and may not count.
What is a typical run in a score-based casual game when a few players score huge?
In endless games a few superb runs drag the average far from what most people experience. This entry shows which numbers to read instead and how to place new obstacles by how many players will see them.
How much time does a death cost, and where should I add a checkpoint?
A hard spot is not the same as an expensive spot. This entry shows how to measure what each death throws away, so you can see which save point is spaced badly and where one more would help most.
How do I get a fair typical clear time when a few players skip half the level?
A handful of impossibly fast runs drag the average of a level down and make every normal player look slow. This entry shows how to spot and set aside those runs with a rule that does not need a hand-picked limit.
When one side leads at half time, how often does the other side still win?
A match where the leader always wins is dull for the trailing side. This entry shows how to measure the chance of a comeback for different sizes of lead, in a way that works for matches with different scoring targets.
Do controller players die more than keyboard players, or do they just play other levels?
Comparing the death rates of two input devices is misleading when each group spends its time in different places. This entry shows how to put both on the same footing so the difference you see belongs to the device and not to the levels.
What really killed the player, if the last hit is often just a scratch?
Your death log probably blames whatever struck last, so harmless things look deadly and the true killers go unnoticed. This entry explains how to decide the cause from the moments before the end.
The map shows most deaths in the corridor, but is the corridor really dangerous?
A heat map of deaths mostly shows where players walk, not where they are in danger. A busy hallway lights up even when it is harmless, and a side room with a nasty trap hides. This entry shows what to divide by.