Loss Aversion

Context & Mechanics

Definition and origin

Loss aversion is a core building block of prospect theory (Kahneman/Tversky, 1979): people evaluate outcomes relative to a reference point, and the pain of a loss empirically weighs about 2 to 2.5 times as much as the pleasure of an equal-sized gain. A $1,000 loss thus feels roughly as intense as a $2,000 to $2,500 gain – an asymmetry that systematically overrides rational expected-value calculation.

Effects in trading

The best-known consequence is the disposition effect: losing positions are held (“as long as I don't sell, it's not a real loss”), winning positions closed early (“lock in the profit before it disappears”). In the journal this shows as a distorted R-distribution: many small winners around +0.5R, single large losers at −2R or worse – the opposite of what positive expectancy needs. Loss aversion is also behind moved stop-loss levels (not wanting to realise the loss), behind the win-back impulse of revenge trading and behind the excessive fixation on the win rate – frequent small wins feel better than they are statistically.

Counter-strategies

The bias does not disappear through knowledge – it can only be bypassed structurally: define exits before entry (stop and target as a bracket order), evaluate results in R instead of currency, make decisions based on the trading plan instead of feeling, and regularly check one's R-distribution for disposition patterns. Framing helps too: treating the stop as an “insurance premium” makes realising losses easier.

Why it matters for traders

Loss aversion is not a character flaw but standard equipment of the human brain – what matters is whether the process neutralises it. In the GlanWick journal the R-distribution makes disposition patterns visible – GlanWick is a training and simulation tool and not a prop firm itself.

Execution Example

A trader on a $100,000 account ($1,000 risk per trade) plans a trade with a stop at −1R and a target at +2R. The position first runs to +1R, then falls back and approaches the stop.

Plan: −$1,000 stop, +$2,000 target – positive expectancy at a 40 % hit rate.
Loss aversion in action: near the stop the level is moved by “just a few points” – the realised loss would hurt, the paper loss feels provisional.
Result: the market keeps going – exit only at −$2,100 (−2.1R) instead of −$1,000.
Journal picture after 50 trades: average winner +0.6R (locked in early), average loser −1.4R (stops moved) – the bias has turned a planned +2R/−1R strategy into a losing one; bracket orders would have enforced the plan mechanically.

Execution Risk & Errors

Moving stops to avoid having to realise the loss

Closing winners early and letting losers run (disposition effect)

Classifying paper losses as “not real yet”

Evaluating results in currency instead of R and thereby deciding emotionally

Never checking one's R-distribution for disposition patterns

Frequently Asked

How heavily do losses weigh compared to gains?

Empirical prospect-theory studies find factors around 2 to 2.5: a $1,000 loss feels about as intense as a $2,000 to $2,500 gain.

What is the disposition effect?

The tendency, following from loss aversion, to hold losing positions too long and close winning positions too early – measurable in an R-distribution with small winners and large losers.

Can I train away loss aversion?

Hardly the bias itself – it is deeply anchored neurologically. It can be bypassed structurally: fix exits in advance, use bracket orders, think in R and let the plan decide instead of the feeling.

How do I recognise loss aversion in my journal?

By moved stops (losses below −1R), early-closed winners (many +0.5R instead of the planned +2R) and a win rate higher than profitability would suggest.