MC · The Loss Scale

A 90-second instrument · 8 gambles · no real money

Would you turn down a fair bet?

Eight 50/50 chances in Romanian lei, all hypothetical. Answer accept or decline — at the end, the scale puts your fear on the pan: your λ coefficient, how many times heavier a loss feels than an equal gain. Most people land between ~1.3 and ~2.5; the classic lab median is 2.25.

  • 8 gambles
  • ~90 s your test
  • λ ≈ 2.25 classic median
  • 1 leu of real money at stake: zero

Your answers never leave the page. No account, no server, no financial advice — just an illustrative number about you. Built on prospect theory (Kahneman & Tversky, 1979; Tversky & Kahneman, 1992).

Why loss weighs more

The prospect-theory value function has two properties: it is steeper for losses than for gains (that is λ) and it shows diminishing sensitivity — concave for gains, convex for losses. Losing lei 100 hurts more than gaining lei 100 pleases; the gap between 1,000 and 1,100 matters less than the gap between 0 and 100.

gain loss λ ≈ 2.25× steeper
Illustrative sketch after Tversky & Kahneman (1992): α ≈ 0.88, λ ≈ 2.25. This page estimates only λ, via a linear approximation.

How your number is made — and where it breaks

The rule, in plain words. Accepting "50% win G, 50% lose L" says G outweighs λ × L, i.e. λ < G/L. Declining says the reverse: λ > G/L. The eight G/L ratios (1.20–2.50) squeeze your λ between the largest declined ratio and the smallest accepted ratio; the estimate is the geometric mean of the two ends, rounded to one decimal. Accept-everything yields 0.7, decline-everything yields 3.5 — flagged as an edge pattern.

Limits, next to the result, not in a footnote. The gambles are hypothetical — with no real stake people look braver, and modern hypothetical-stake estimates (1.25–1.45) run below the classic median (2.25). The approximation is linear: it ignores curvature (α) and probability weighting. Eight items give a noisy snapshot, not a diagnosis. λ is not financial advice and not a character test.

Sources

  1. Kahneman, D. & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica, 47(2), 263–291. doi:10.2307/1914185
  2. Tversky, A. & Kahneman, D. (1992). Advances in prospect theory. Journal of Risk and Uncertainty, 5, 297–323 — median λ = 2.25, median exponent 0.88. stable quote
  3. Tom, S. M. et al. (2007). The neural basis of loss aversion. Science, 315, 515–518 — the 50/50 accept/decline task behind your test. doi:10.1126/science.1134239
  4. Vieider et al. Prospect theory's loss aversion is robust to stake size. Judgment and Decision Making — λ 1.25–1.45 on hypothetical stimuli. article
  5. Brown, Imai, Vieider & Camerer (2024). Meta-analysis of Empirical Estimates of Loss Aversion — heterogeneity by method. record

The estimator runs in gambles.js — the same 8 answers always give the same λ.