Data and decisions
Rage quits: how to investigate players leaving matches
Players tend to leave just after losing a tower. Should the tower go, does the opponent's advantage spiral, or can nobody see a way back? The data locates a question; it does not yet give you the answer.
Key idea
A departure pattern tells you where to investigate; observing the game lets you propose a cause and test it.
A departure does not explain what happened on its own. Someone may leave a match because of frustration, a poor connection or circumstances unrelated to the game. When many departures cluster at the same point, the pattern becomes a useful clue for design.
Find the breaking point
A drop in players following the loss of a tower, after an opponent obtains a particular item or at the end of an especially uneven round may indicate that players perceive that moment as irreversible. Analytics and observation can identify when the departure happens, although they cannot establish each person's individual reason with certainty.
This kind of data does not dictate a solution. If departures increase after losing a tower, removing towers will not necessarily improve the game. The associated reward may generate too much advantage, the change may open up the map in an oppressive way or comeback options may exist that the game communicates poorly.
From metric to hypothesis
The next step is to return to the game. It helps to observe matches, examine the state of both teams before and after the event and ask which options the losing players perceive. The data defines an area to investigate; analysing the system makes it possible to formulate and test explanations.
One frequent cause is the snowball effect: an initial advantage produces resources, control or information that makes creating another advantage easier. When the player can no longer imagine a move that could change the match, quitting begins before the game is closed.
Losing without disconnecting from the match
A competitive game does not need to guarantee constant comebacks. It needs to preserve valuable decisions even for players at a disadvantage. It can do this through alternative objectives, windows of risk, recovery tools or clearer communication of the available opportunities. The goal is for defeat to remain a match worth thinking about and acting in, rather than a passive wait for the final result.
Apply it to your game
- Locate the event around which departures cluster. Where possible, distinguish technical disconnections and other circumstances before attributing them to design.
- Observe matches around that moment. Review the advantage, the available options and what players believe they can still achieve.
- Form a specific hypothesis and change one factor to test it. Compare equivalent matches and combine data with observations before drawing conclusions.
When an advantage becomes irreversible, review how to preserve comeback options. To turn comments into testable explanations, use the approach to interpreting playtests.