Stream: social
Topic: Pareto Principle (or the basis of our 80/20 rule)
Jean Duteau (Oct 26 2021 at 15:22):
I was doing some statistical analysis work and came upon the Pareto distribution and the Pareto principle: "80% of outcomes are due to 20% of causes". It's not really the principle we use in FHIR - "80% of implementations in the world use this concept in this way" but it was interesting to read about it (if you're a math geek like I am). https://en.wikipedia.org/wiki/Pareto_principle
Rik Smithies (Oct 26 2021 at 15:42):
Is it really a thing? It has always struck me that it may just saying be "some things are uneven".
If things are even or there is a slight unevenness, e.g. 60/40, or if they are totally 100/0, then that is no surprise and you don't call it out.
If it's fairly uneven e.g. 70/30 or 80/20, then you make a rule and then only notice and cite the ones where things fit your pattern. I'd like to know if 80/20 has some special significance, but I doubt it.
David Pyke (Oct 26 2021 at 15:44):
It's a convention, and has direct applications in math and other places. However, it gets applied all over the place as a simplification rule, even where it may not actually be true in that case.
Rik Smithies (Oct 26 2021 at 16:03):
go on I will bite :-) Does it have direct application in maths in itself? Or is it just that power laws have application in maths, and this is saying that this one is special somehow and has a certain significance in the world (ignoring the cases where it doesn't...).
Rik Smithies (Oct 26 2021 at 16:05):
It reminds me of "third time lucky" or "these things happen in threes". If they happen in 1s, no big deal. If there are 2, no big deal. 3 starts to seem weird. 4 and 5 etc are less common (you either gave up trying by then or lots of different numbers happen). So you fixate on 3 and ignore other cases, and then things always seem to happen in 3s.
Jean Duteau (Oct 26 2021 at 16:19):
it actually is a thing. There is a specific statistical distribution that reflects the 80% of the results are covered by 20% of the causes.
David Pyke (Oct 26 2021 at 16:31):
Rik didn't read the article. Ask me how I know
Rik Smithies (Oct 26 2021 at 16:31):
because 80% of people don't read articles they comment on?
Rik Smithies (Oct 26 2021 at 16:32):
I only read 20% of it
Peter Jordan (Oct 26 2021 at 18:43):
Interesting, although some of it is clearly outdated - e.g. current World GDP Distribution is top 10% of adults holding 85% and WordPerfect is now housed in software museums.
David Pyke (Oct 26 2021 at 19:48):
I'm sure wikipedia would appreciate your edits
Peter Jordan (Oct 26 2021 at 23:12):
Probably not where the facts no longer support the theory. :grinning:
Martijn Verhoeven (Oct 27 2021 at 10:41):
I use WordPerfect still for finding rhyming words for my poems :joy:
John Silva (Oct 27 2021 at 16:05):
Hmm, speaking of math problems related to 3 --- Here's a TED talk on the "Simplest Math Problem No One Can Solve"
https://ed.ted.com/best_of_web/U1LBfyZ5
Jose Costa Teixeira (Oct 27 2021 at 18:45):
I have learned it years ago as just a rule of thumb roughly saying that you may expect to get 80% of the outcome with 20% of the effort. I thought it was commonplace.
Elliot Silver (Oct 27 2021 at 18:46):
Jose Costa Teixeira said:
I have learned it years ago as just a rule of thumb roughly saying that you may expect to get 80% of the outcome with 20% of the effort. I thought it was commonplace.
No, no, that one is "80% of the project takes 80% of the effort. The remaining 20% of the project takes the other 80% of the effort."
Jose Costa Teixeira (Oct 27 2021 at 18:50):
@Elliot Silver if I remember correctly that rule was developed in manufacturing, but indeed we also apply it in projects
Jose Costa Teixeira (Oct 27 2021 at 18:54):
I think Pareto's rule can be applied recursively. 80% of the benefit of Pareto's rule comes with the first 20% of the discussions about Pareto's rule. The other 80% only yield 20% of benefit :)
Jose Costa Teixeira (Oct 27 2021 at 18:58):
... I missed the joke, @Elliot Silver . I see it now.
For projects, Zeno's paradox also contains some truth ;)
Cooper Thompson (Oct 27 2021 at 20:01):
I feel like the Pareto Principle only applies about 80% of the time.
Last updated: Apr 12 2022 at 19:14 UTC
