An asymmetry most players feel
Win a ranked match and you gain some rating. Lose and you give up a similar amount. In a simple system the two are about equal in size. Yet most players report that the loss stings more than the win pleases. A drop of 20 points can ruin an evening, while a gain of 20 is quickly forgotten.
Behavioural economists call this loss aversion. It is one of the central ideas in prospect theory, developed by Daniel Kahneman and Amos Tversky, and it describes a tendency for losses to be weighted more heavily than equivalent gains. Researchers have often estimated that the weighting is roughly two to one, though the figure varies by person, context and study, and the finding has been refined and debated over the years. The safest way to say it is that losses typically loom larger than gains of the same size.
Why ranked play is a natural fit
Ranked ladders put loss aversion on display for several reasons.
- Ratings are visible and numeric. A number that goes up and down makes every game a small gain or loss in plain sight.
- Reference points are obvious. A rank boundary, a season peak, or the rating you started the day with all act as anchors. Falling below an anchor feels like a loss, even if the underlying skill is unchanged.
- Ratings feel earned. Because a rating represents effort, losing points can feel like losing something you own, which is stronger than missing out on something you did not have.
- Feedback is immediate. Every match ends with a number moving.
The effect is also tied to identity. For many players, rank signals status, so a drop is not only a change in a number.
A simple illustration
Take a ladder where a win gives 25 points and a loss costs 25. If a loss feels about twice as heavy as an equal win (an assumption, since the true ratio varies), then a win feels like +25 and a loss like -50. At a 50 percent win rate the average felt result per game is (25 − 50) ÷ 2 = −12.5. So a player who is exactly breaking even in actual rating can feel as if they are losing ground. This is why long sessions of even results often feel bad.
The figures are for illustration only. Real systems vary the points by opponent strength, and the feeling is not measured on any exact scale. The point is that an even record can feel like a negative one.
How it changes behaviour
Loss aversion does not push everyone in the same direction. It tends to produce several distinct patterns.
Overcaution near a boundary. When a player is close to a rank threshold, a loss feels especially costly, and they may play too safely. They avoid contesting objectives, refuse reasonable trades, or decline to commit when the play is positive in expectation. Safe play is not always right, and over-protecting a lead can hand the initiative to the opponent.
Avoiding the queue. Some players stop playing when they are on a good rating, to “lock in” a rank. Others avoid queueing at all after a loss. In a system where ratings reflect long-run skill, sitting on a rating does not make it more real, but it can feel safer.
Risk seeking to get back to even. Prospect theory also describes a pattern in which people facing a loss become willing to take gambles they would otherwise avoid, to try to reach the reference point. After a drop, a player may try a high-variance strategy or queue repeatedly to recover. This is closely related to the sunk cost trap, and it can build into emotional tilt.
Blaming the system. When losses feel disproportionate, it is tempting to conclude the matchmaking is unfair. Sometimes there are genuine system quirks, but the feeling alone is not evidence. Part of it is the asymmetry itself. Looking at the sample size of your results is a better test than feeling.
What the ladder is really telling you
Ratings are an estimate of skill, and they come with noise. In rating systems built on ideas like Elo ratings, each result nudges the estimate, and the estimate moves up and down around the true level even for a player whose skill does not change. Short-term swings are expected, and rating volatility is a normal part of the process. Seen this way, a 20-point loss is not a verdict. It is a small, noisy data point.
A useful reframe is to think in terms of the whole distribution of your results. Over a month, a player of steady skill will see their rating wander. A single dip tells you little, and reading it as a loss of something owned adds an emotional cost that the number does not justify.
Practical ways to reduce the distortion
None of these remove loss aversion, which is a general feature of human judgement, but they can reduce its effect on play.
- Decide your risk approach before the match. Choose plays based on the situation, not on how close you are to a boundary. “What would I do if the rating were irrelevant?” is a useful question.
- Use a long horizon. Judge a week or a season rather than a night. A longer view makes individual losses less prominent.
- Track something other than the headline number. Measures of execution and decision quality, such as those from stat tracking, give you progress to see even when the rating is flat.
- Avoid correcting a loss with more games. If you notice the urge to recover points immediately, that is the cue to pause.
- Do not hide the rating, but do not stare at it. Looking after every game invites the comparison to the anchor. Checking at planned times is calmer.
A note on evidence
Loss aversion is well documented in general decision research, though its size and how widely it applies are still debated. Evidence about ranked play specifically is thin. The patterns described here are plausible applications, not proven rules, and individuals differ a lot. The value of the idea is as a prompt: when a decision feels strangely heavy, check whether the weight comes from the situation or only from the fear of losing a number.