---
title: Mehbark’s First Law 
description:
    Any curve will leave someone disappointed.
    A 1% curve will disappoint people 2% short of their desired grade, and a 10% curve will disappoint people 11% short of their desired grade.
    Even a 100% curve will disappoint some people because every score is now meaninglessly high.
tags: post,short,law
date: 2026-05-26 23:02:29 -4
---

export const gpa = <abbr title="Grade Point Average">GPA</abbr>;
export const evs = <abbr title="Effort Values">EVs</abbr>;

<p>{props.description}</p>

## Why
[Non-linear consequences](https://www.youtube.com/watch?v=a7anye0FwcY).
Point thresholds matter a lot in academics; the obvious example is {gpa}.
For the grading scale that determines {gpa}, shown below, the only points that make a difference are 60, 70, 80, and 90.

<iframe src="https://www.desmos.com/calculator/1gnsgvqw3a?embed" title="Graph showing a piecewise function that only increases at 60, 70, 80, and 90. The derivative everywhere else is zero." class="desmos" width="500" height="500" style={{border: "1px solid #ccc"}} frameBorder="0"></iframe>

Only getting to the next “step” is valuable, so being just below is frustrating—*even if you were farther away before your grade was curved*.
Objective outcomes aside, a high B feels worse than a low A even though we know they can be arbitrarily close.

Height perception functions the same way.
6′0″ is not a milestone in centimeters (182.88), and 180cm is not a milestone in feet (≈5′11″).
In an alternate history where the meter was different (maybe a different Earth circumference), more or fewer people would meet that threshold.

## More examples
Grades are the most obvious example (hence the use of the word “curve”), but anywhere there are breakpoints, this effect might apply.

A <span class="work-title">Slay the Spire 2</span> player [suggested](https://old.reddit.com/r/slaythespire/comments/1tnjfjp/just_let_me_sell_potions_at_the_shop/)
that it should be possible to sell potions at shops.
Being a few gold short of a valuable purchase is frustrating and often not strategically interesting (gold rewards are random within a range),
but this proposed solution would create situations where you are just shy even after selling your potions.

Competitive Pokémon has a truly staggering number of relevant breakpoints.
Investing {evs} in a stat lets you survive, kill, or outspeed more often, but it is always at the cost of another stat, which means giving up surviving, killing, or outspeeding in other situations.

I feel like I am missing some obvious example.