Averagarianism: Difference between revisions

Latest revision as of 09:15, 26 June 2024

Lies, Damn Lies and Statistics

Index: Click ᐅ to expand: Statistics Systems theory

Tell me more Sign up for our newsletter — or just get in touch: for ½ a weekly 🍺 you get to consult JC. Ask about it here.

Averagarianism
/ˈævᵊrɪʤeəriənɪzᵊm/ (n.)
The mistake of attributing an emergent mathematical property of a group to some or all the individual members of the group.

The folly of reasoning from the general to the particular with statistics.

For example, the Bill Gates on a bus paradox: the average wealth of 99 bankrupts and one billionaire is ten million dollars. But not one individual in the group has an income anywhere close to ten million dollars. The median wealth is zero. The median effectively discounts outliers either side, so is more likely to represent the “real consensus”.

See also the “average fighter pilot,” that Gladwellian character from any number of popular science books.

Lesson one: do not manage from the average to the particular.

Then there is the story — oft repeated at a microscale, sans doubte — of the global investment bank which addressed its gender pay gap by laterally recruiting a new general counsel for ten million dollars. Remaining victims of pay disparity remained unmoved, and undercompensated. (This is not to say, “don’t act to correct pay unfairness”; just “don’t do it by massaging the average”. Seek out and rectify, you know, actual pay unfairness. In the particular.)

Lesson two: definitely do not manage from the particular to the average.

Other reasons not to manage to the average

Rory Sutherland has an excellent snippet about the danger of managing toward averages. Among his reasons:

Find a niche

The “average” is where everyone else will target their product. The markets will be mature, barriers to entry high, demand inelastic and margins slim.

Go instead for the tails: have the average follow you, not the other way around.

To use a skiing metaphor, the best entertainment to be had is not on the groomed blue motorway with the poseurs, learners, and homicidal teenagers, but off piste. You just have to know how to ski. So, learn, or take up another hobby.

Don’t race to the bottom

Convergence on the same place everyone is converging isn’t good business, but a recipe for bankruptcy. It is a race to the bottom. As with evolution, the secret is to realise the process is a continuous drift from the unsatisfactory status quo to something else that doesn’t have that drawback, as opposed to a process converging on a consensus. The ecosystem is not seeking an equilibrium. It is perpetually seeking to escape it.

Averagarianism

Averagarianism that forces actually different people into generic categories. It imputes commonalities that don’t really exist. Sanding off contours and wonky borders to make everything regular simply because that suits the hyper-scaled prerogatives of commerce.

References

@@ Line 1: / Line 1: @@
-{{a|devil|}}{{author|Rory Sutherland}} has an excellent [https://youtu.be/UirCaM5kg9E snippet about the danger of managing toward averages]. Among his reasons:
+{{a|stats|}}{{dpn|/ˈævᵊrɪʤeəriənɪzᵊm/|n}}The mistake of attributing an [[emergent]] mathematical property of a group to some or all the individual members of the group.
-*The average — the top of the bell curve— is where everyone will be targeting their product, so existing markets will be mature, barriers to entry high, and margins will be the slimmest. Go for the tails, find the influencers and meet them drive your product into the mainstream. Have the average follow you, not the other way around.
+The folly of reasoning from the general to the particular with statistics.
+For example, the [[Bill Gates on a bus]] paradox: the  ''average'' wealth of 99 bankrupts and one billionaire is ten million dollars. But not one individual in the group has an income anywhere close to ten million dollars. The median wealth is zero. The median effectively discounts outliers either side, so is more likely to represent the “real consensus”.
+See also the “average fighter pilot,” that Gladwellian character from any number of popular science books.
+Lesson one: do not manage from the average to the particular.
+Then there is the story — oft repeated at a microscale, ''sans doubte'' — of the global investment bank which addressed its gender pay gap by laterally recruiting a new [[general counsel]] for ten million dollars. Remaining victims of pay disparity remained unmoved, and undercompensated. (This is not to say, “don’t act to correct pay unfairness”; just “don’t do it by massaging the average”. Seek out and rectify, you know, ''actual pay unfairness''. In the particular.)
+Lesson two: ''definitely'' do not manage from the particular to the average.
+===Other reasons not to manage to the average===
+{{author|Rory Sutherland}} has an excellent [https://youtu.be/UirCaM5kg9E snippet about the danger of managing toward averages]. Among his reasons:
+====Find a niche====
+The “average” is where everyone else will target their product. The markets will be mature, barriers to entry high, demand inelastic and margins slim.
+Go instead for the tails: have the average follow you, not the other way around.
+To use a skiing metaphor, the best entertainment to be had is not on the groomed blue motorway with the poseurs, learners, and homicidal teenagers, but [[off piste]]. You just have to know how to ski. So, learn, or take up another hobby.
+====Don’t race to the bottom====
+Convergence on the same place everyone is converging isn’t good business, but a recipe for ''bankruptcy''. It is a race to the bottom. As with [[evolution]], the secret is to realise the process is a continuous drift ''from'' the unsatisfactory status quo to something else that doesn’t have that drawback, as opposed to a process converging on a consensus. The ecosystem is ''not'' seeking an equilibrium. It is perpetually seeking to ''escape'' it.
+===Averagarianism===
+Averagarianism that forces actually different people into generic categories. It imputes commonalities that don’t really exist. Sanding off contours and wonky borders to make everything regular simply because that suits the hyper-scaled prerogatives of commerce.
-*Convergence on the same place everyone is converging isn’t good business, but a recipe for ''bankruptcy''. It is a race to the bottom. As with [[evolution]], the secret is to realise the process is a continuous drift ''from'' the unsatisfactory status quo to something else that doesn’t have that drawback, as opposed to a process converging on a consensus. The ecosystem is ''not'' seeking an equilibrium. It is perpetually seeking to ''escape'' it.
 {{sa}}
+*{{br|The End of Average: How to Succeed in a World That Values Sameness}}
+*[[Metis]]
 *[[Big data]]
+*[[Data modernism]]
 *[[Correlation]]
 *[[Parable of the squirrels]]
+{{Ref}}