# Talk:Pareto principle

WikiProject Statistics (Rated C-class, High-importance)

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

C  This article has been rated as C-Class on the quality scale.
High  This article has been rated as High-importance on the importance scale.
This article has been mentioned by a media organization: Dale King (December 25, 2006). "The Pareto Principle". American Chronicle. (details)

This really needs an example.

## Proper Quote? Cite?

This page states "It was named after the Italian economist Vilfredo Pareto, who observed that 80% of income in Italy was received by 20% of the Italian population."

The Vilfredo Pareto page states "In 1906 he made the famous observation that twenty percent of the population owned eighty percent of the property in Italy, later generalised by Joseph M. Juran and others into the so-called Pareto principle (also termed the 80-20 rule) and generalised further to the concept of a Pareto distribution."

So which is it, income or property?

I can no where see the so called Pareto's law: ${\displaystyle N=A\cdot x^{n}}$ Where N is the number of individuals with an income over x. A and n are constants. Pareto found in all populations he studied that n=-1.5. Is 80/20 derived from this or are Pareto's law and Pareto's principle entirely separate things? —Preceding unsigned comment added by Laelele (talkcontribs) 12:11, 8 January 2010 (UTC)

I'll add something on the relationship between the 80/20 law and the identity N = Axn Michael Hardy (talk) 19:32, 15 September 2010 (UTC)

## Explanation

I don't know much about business, but I think the examples of misuse need brief explanations. It's not clear to me at all that the application to employees is either untestable or useless to the firm (if those are the reasons it's a misuse), and I also don't understand why the application to advertising is useless. —JerryFriedman 19:41, 24 Jan 2005 (UTC)

For the benefit of the many, let's elaborate on "A clear misuse of Juran's assumption (...)":
In manufacturing, systems are designed so that everybody contributes a pre-defined amount of work, especially in synchronous settings such as assembly lines. Nobody could argue for applying the Pareto principle there.
In agriculture and mining, individual contributions are sometimes measurable (e.g. baskets of apples picked per shift). A for-profit organization seeks out the better workers, so the bottom part of the distribution is not be represented at all. Family and community operations often find a way to gainfully employ persons with less strength (e.g. 11-year-old girls) or with disabilities (e.g. blind people), but their number is not large enough to warrant a Pareto distribution.
Which brings us to the only economic system many readers have experience of: The services sector.
There is a place for "stars" in fields like moviemaking and advertising. In these fields, moneymaking ability is often modeled as 80:20, and even as 90:10 or 99:1 (think of the distribution of pay for fashion models).
In other service areas, such as software development, it has been argued (and many influential minds such as Frederick Brooks and Joel Spolsky have devised methods for measuring and quantifying this) that a top designer, programmer, or QA person is roughly ten times as effective as an average one.
It can be argued that this observation does not carry to other areas of computing such as system administration.
Finally, the difference between top waiters (or shoe salespeople, mail delivery people...) and average ones may not warrant Pareto analysis at all.
elpincha 00:41, 4 Feb 2005 (UTC)

## Recursion

O.K. I may be missing something here but the examples of how the set is recursive don't seem consistent to me. The idea of a 64:4 rule makes sense in a linear fashion (80 - 16):(20 - 16) but that doesn't seem to pan with a 51.2:0.8 rule. The 64:4 rule also seems to suggest that the remaining top 16% of the work accounts for the remaining top 16% of the output which is inconsistent with this principle. It seems to me that a 40:10 rule (10% of effort yields 40% of results) would make more sense (80/2):(40/2) (or 80/n:20/n)

The Pareto principle is *not* a ratio and can not be reliably manipulated as though it were a ratio to produce equivalent formulations. To avoid the "ratio" analogy, I'll make up a new operator ":::" to represent the Pareto principle. If J is the consequences percentage and K is the causes percentage, and J:::K is the Pareto principle stating "J of the consequences stem from K of the causes", then J:::K does not necessarily imply that (J-16):::(K-16), nor does it imply that (J/n):::(K/n).
However, it is claimed that the Pareto principle can be reduced recursively.
Let f(X, N) = (X/100)*f(X, N-1), for N > 1; f(X, 1) = X.
J:::K implies f(J,N):::f(K,N), for N >= 1.
Also the principle can be "expanded".
Let g(X, N) = ((100 - g(X, N-1)) * (X/100))+g(X, N-1) for N > 1; g(X, N) = X.
J:::K implies g(J, N):::g(K, N).
Using 80:::20: 80% of the 80% of consequences stem from 20% of 20% of the causes. This multiplies out as 80% * 80% consequences from 20% * 20% causes, which gives 64% of the consequences from 4% of the causes.
We can express this reduction using the recursion function 'f(X, N)' for "reduction" given above where X is 80 and the reduction step N is 2:
f(80, 2):::f(20, 2)
= [(80/100)*f(80,1):::(20/100)*f(20,1)]
= [0.8 * 80 ::: 0.2 * 20]
= 64:::4.
The next recursive step is: f(80, 3):::f(20, 3)
= [(80/100)*f(80,2):::(20/100)*f(20,2)]
= [0.8*64:::0.2*4]
= 51.2:::0.8.
The "expansion" of the 80:::20 rule is:
g(80, 2):::g(20, 2)
= [((100 - g(80, 1)) * (80/100)) + g(80, 1):::((100 - g(20, 1)) * (20/100)) + g(20, 1)]
= [((100 - 80) * 0.8 + 80:::((100 - 20) * 0.2 + 20]
= 96:::36.
Expanding again: g(80, 3):::g(20, 3)
= [((100 - g(80, 2)) * (80/100)) + g(80, 2):::((100 - g(20, 2)) * (20/100)) + g(20, 2)]
= [((100 - 96) * 0.8) + 96:::((100 - 36) * 0.2) + 36]
= 99.2:::48.8.
68.100.145.9 16:20, 21 November 2005 (UTC)

80-20 results in 64-4, but there are two 16-16 correlations: one representing the lower top, one representing the upper bottom, for a 'fair' section of 32% owning 32% of the wealth: 64 + 4 + 32 = 100. This perhaps explains the strange shape of the graph below, where the straight line section is a displaced 'fair' bit, joining the very poor with the very wealthy. Where I'm curious is (a) whether this 32% also follows the 80-20 rule and (b) how 'fair' this 32% is, given that it's only a function of applying the recursion once: apply it twice, and you get a much larger 'fair' section, apply it 'half a time' 1-(0.8^1.5 + 0.2^1.5) and it's smaller. SleekWeasel (talk) 15:07, 30 November 2013 (UTC)

I added some explanations on this recursion thing to the article. They'd help me to understand it; I don't know about others... But without some sort of explanation, I believe the paragraph should be removed entirely from the article.--Niels Ø 15:52, 15 February 2006 (UTC)

And then I thought some more about it, and removed the paragraph entirely. It appears to be unreferenced, perhaps original research, and I don't think it's "true" in the same sense the 80-20 rule as such is true, viz. (1) that it is often cited in various situations, (2) it is actually not far from being correct in many of those cases, and (3) even when it is far from being accurate, it may point to a fact that needs to be taken into consideration. Here's what I removed (ncluding my own attempts to explain):

The principle can be viewed as recursive, and may be applied not only to the top 20% of causes; thus there would be a "64-4" rule (64% of the consequences stem from 4% of the causes), and a "51.2-0.8" rule, and so on (80% = 4/5 or four fifths, 64% = (4/5)2 or four fifths of four fifths, 51.2% = (4/5)3; 20% = 1/5, 4% = (1/5)2, 0.8% = (1/5)3). In the opposite direction, Tipton Cole has observed that the Pareto Principle applies to the residue of its first application, yielding a "96-36" rule. If 20% of the causes are responsible for 80% of the consequences, the other 80% of the causes are only responsible for the remaining 20% of the consequences. Applying the principle recursively, the 64% least consequential causes are responsible for only 4% of the consequences, and hence the 36% most consequential causes are responsible for 96% of the causes.

Among my reasons to discard it is the following figure, where I think the curve found by the recursive principle is just too weird-looking. (The other curve is a hyperbola fitted through (0,0), (80%,20%), and (100%,100%) - it may not be a lot more "correct" in any particular situation, but at least it looks less strange.)

--Niels Ø 13:31, 16 February 2006 (UTC)

And then Michael Hardy re-added a brief version of it. It seems that it's OK one way (64-4 rule), but perhaps not the other way (96-36 rule).--Niels Ø 19:57, 18 February 2006 (UTC)

Regarding "Thus, the 80-20 rule would imply that 64% of wealth is held by 4% of the people; however, one cannot reverse this and say that 4% of the wealth is held by the poorer 64% of the people.", I call bogus that 64:::4 doesn't apply equally to the richest 4% and poorest 64% owning 64% and 4% respectively: the population/wealth table at the start of the page seems to correlate with precisely that: 20% have 82.70%, and the third, fourth, and fifth 20%s add up to about 5.5%. The second 20% owning 11.75% is broadly in line with 16% owning 16% of the wealth, correlating with a 'fair' share. SleekWeasel (talk) 13:52, 30 November 2013 (UTC)

## Examples of misuse

Is there a reason to remove the examples of "misuse"? If not, I'll re-add them. Mnbf9rca 01:29, 18 February 2006 (UTC)

## Software engineering

I think this section is correctly to be stated, "The first 90% of the code accounts for the first 90% of the development time. The remaining 10% of the code accounts for the other 90% of the development time" - see 90-90 Rule 198.49.180.40 22:11, 13 November 2006 (UTC)

Perhaps this is in reference to content that has since been removed, but the canonical software engineering example of this priciple is that "80% of your time programming is spent on 20% of the code". Michael.Urban (talk) 15:09, 7 November 2008 (UTC)
Ninety-ninety_rule SleekWeasel (talk) 15:37, 30 November 2013 (UTC)

## Self-reference removed.

I have removed the following from the examples section: "* 10% of wikipedia users make 80% of all edits"

It seemed like needless self-reference, especially when I clicked on it and was taken to what appeared to be a powerpoint outline for a talk given at some conference by Jimmy Wales. ~ ONUnicorn(Talk|Contribs)problem solving 20:41, 31 January 2007 (UTC)

## Second Paragraph quite confusing

"The Pareto principle is only tangentially related to Pareto efficiency, which was also introduced by the same economist, Vilfredo Pareto. Pareto developed both concepts in the context of the distribution of income and wealth among the population."

but the first paragraph says that the principle was developed by Joseph M. Juran and named after pareto. The second paragraph should be changed to reflect this. I would do it, but I don't know enough about this topic. —The preceding unsigned comment was added by Rebent (talkcontribs) 16:28, 4 May 2007 (UTC).

## Mathematical notes

I see that a comment has been added saying Note, however, that that sometimes adding up to 100 is indeed meaningful. This appears to directly contradict my assertion in the previous paragraph. My point is that there is no reason why the two numbers should add up to 100 as they are percentages of two different quantities. If they do, it is purely a coincidence. Perhaps I am missing something? DaveApter 09:21, 18 May 2007 (UTC)

I don't know anything about maths but what you say makes sense. However, this is wikipedia, so what do the sources say?--Rebent 16:17, 18 May 2007 (UTC)

That last paragraph does have a valid point to make, but it's making it in a confused and inefficient way. I shall return... Michael Hardy 21:16, 18 May 2007 (UTC)

Mathematically, where something is shared among a sufficiently large set of participants, there must be a number k between 50 and 100 such that k% is taken by (100 − k)% of the participants. k may vary from 50 (in the case of equal distribution) to nearly 100 (when a tiny number of participants account for almost all of the resource).

Why the need for 100 here? In the distribution table displayed, it shows a ratio of 82.7 : 20 which does not add up nicely to 100. So why is this in here? What does it say, and how is it not just contributing to confusing the subject at hand?

Also, why is this in the introduction as well as the Mathematical notes? If it is useful, is it useful there?

tylermac (talk) 18:38, 21 October 2009 (UTC)

This section really doesn't make sense. — Preceding unsigned comment added by 174.56.105.36 (talk) 04:37, 24 September 2014 (UTC)

## Optimization Phrases?

In computer science the Pareto principle can be applied to optimization phrases

Shouldn't it say:

In computer science the Pareto principle can be applied to optimization phases —Preceding unsigned comment added by Sfandino (talkcontribs) 08:13, 12 May 2008 (UTC)

## Statistics

Folks, look. The 80-20 rule is to alert people in the workforce that many problems, many advantages, etc., are caused by a few things, and that, therefore, those few things may be more important to find than others. It's just a convenience. Isaac Newton didn't discover it, it isn't one of the laws of the universe.

This sentence in the current article boggles the imagination: "The 80 and 20 measure different things and do not need to add up to 100."

I'm removing it. I don't know who I'm talking to, so this is difficult. Please listen and bear with me. By definition, 100% means complete certainty, everything, all the facts, every single statistical sample. Things that are measured with different measurement standards cannot be compared -- that defeats the whole idea of being comprehensive.

Saying something such as "80% of our crop were apples, and 80% were oranges" isn't so much as totally invalid, it's just a misconception of how statistics work. What the person might mean, in this case, is that half of the crop were oranges, and half were apples.

The 80-20 rule is simply an example whose purpose is to alert people to identify things that are important, and not to "sweat the small stuff". It is not at all important to the point, if in some circumstances, let's say, 81% of the sales come from 20% of your customers.

There's no mathematical need to make all studies conform to the 80-20 rule.

And...please, please...there's no need to explain why the 80-20 rule "does not work". The 80-20 rule isn't absolute...it's not because the 80 and the 20 are measuring different things. —Preceding unsigned comment added by 24.130.14.170 (talk) 04:34, 4 June 2008 (UTC)

The comments above are silly. Obviously 80 + 20 = 100, but you can also have an "80/10" rule, e.g. a population in which income is so distributed that the highest-income 10% of the people have 80% of the income. Note that 80 + 10 ≠ 100. Michael Hardy (talk) 05:05, 4 June 2008 (UTC)
The editor who wrote the above comment, that "The 80 and 20 measure different things and do not need to add up to 100." is silly does not seem to understand what is being asserted by it. The 80/20 principle is a shorthand statement of a more generalised power-law relationship. The 80% measures one variable (eg sales) and the 20% measures another (eg salespersons). Something like 80% of your sales will come from something like 20% of your sales team. It may be that 80% of the sales actually come from 10% of the team, or it may be that 80% of the sales come from 30% of the team. Either would be a valid instance of the Pareto principle. But 80/20 is a useful headline. DaveApter (talk) 17:15, 16 October 2008 (UTC)

## The opening paragraphs needs a re-write

The first paragraph reads as follows

The Pareto principle (also known as the 80-20 rule, "Haddad's Theroem", the law of the vital few and the principle of factor sparsity) states that, for many events, 80% of the effects come from 20% of the causes. Business management thinker Joseph M. Juran suggested the principle and named it after Italian economist Vilfredo Pareto, who observed that 80% of income in Italy went to 20% of the population. It is a common rule of thumb in business; e.g., "80% of your sales comes from 20% of your clients."

It is worthy of note that some applications of the Pareto principle appeal to a pseudo-scientific "law of nature" to bolster non-quantifiable or non-verifiable assertions that are "painted with a broad brush".[citation needed] The fact that hedges such as the 90/10, 70/30, and 95/5 "rules" exist is sufficient evidence of the non-exactness of the Pareto principle. On the other hand, there is adequate evidence that "clumping" of factors does occur in most phenomena.[citation needed]

I agree with you - the second paragraph was incohherent, and I have deleted it. The inro is now a bit sparse and could probably do with some expansion. DaveApter (talk) 13:08, 22 October 2008 (UTC)
And to answer your question the "fact" tag (which appears as [citation needed]) is inserted by editors to draw attention to assertions in an article that may be correct, but are not substaniated by references to a reliable source. This gives other editors the opportunity to find such sources, and if they are not after some reasonable time period, the assertion may be deleted. DaveApter (talk) 13:16, 22 October 2008 (UTC)

Books on management have unanimously preached the message of rethinking priorities, and that 80% of your time, you're probably putting into areas of low return. Also, they preach not to worry about the sum of small things, but just focus on the core--i.e. what you're in business for, what you do best. Sentriclecub (talk) 07:56, 31 July 2008 (UTC)

## 80-20 rule ??

For my part, I've ever heard of a "20-80 rule" so far.

• May be this is not very important. But the starting point of organisation is to aim at better comfort and efficiency, isn'it ? So, admitted that the basical form is "x-y rule", the fact that Pareto allowes us to organize better refers to the y/x ratio, describing the increasing effect of a right distribution as "four times more than invested". And the corresponding ratio is 20-80, not 80-20, that warns us not to forget that decreasing effect(s) do follow the phenomenon. Is that correct ? 2rh (talk) 13:22, 22 December 2008 (UTC)
• Unless the correct form in english is "y-x rule" Is it ??? 2rh (talk) 13:24, 22 December 2008 (UTC)
• I may be entirely wrong with all this (french part also reports about a "80-20 rule") Sorry 2rh (talk) 15:40, 22 December 2008 (UTC)

## Pareto principle

For my part, i've heard that "Pareto's principle is "not to get a linear distribution", for a linea distribution is impossible to organize (source : ENISE 1974). As a consequence of it, Pareto's principle can't be reduced to on of its ratios, even if on of them is very popular. The second most popular ratio of a Pareto distribution beeing "50-95" (still admitted - question above - that the correct form of a ration in english is "x-y", not "y-x" 2rh (talk) 13:35, 22 December 2008 (UTC)

## Microsoft and software bugs

Microsoft also noted that by fixing the top 20% of the most reported bugs, 80% of the users would not encounter any bugs.

Only the first part of the sentence is true and is what the source says. I think someone has over-interpreted the article. I'm changing the second half of the sentence. BTW, I bet that that most users will finally find some bug, especially in Microsoft's software :) Trosmisiek (talk) 10:47, 25 January 2009 (UTC)

## Criticisms

This so called principle seemed stupid to me the first time I read about it. After a bit of thought it seemed even stupider. I wrote to my Economics lecturer asking what he thought about it. This is what he had to say. (I think it would be a good idea to create a formal criticism section.)

I had not heard of this 'Principle' before. It looks to me to be either (a) vacuous or (b) idiotically false.
Just to be clear about what the Principle is saying, let's express it a bit more clearly. For a particular effect E, there is a set of possible causes C1, C2 ... C10. Of this set, 20% of them (say, C1 and C2) bring about 80% of the occurrences of E. Thus the remaining 80% of the set of causes (C3 to C10) account for the remaining 20% of the occurrences of E.
First the vacuity charge: 80% of WHAT is due to 20% of all the possible causes of it? One must answer the question of WHAT effect is being referred to before one can even begin to make judgements about its truth. The proposition is 'empty' until it is 'filled in' with this content. As it stands, it seems to just be saying that "80% of something is caused by 20% of all the possible things that could cause that 'something'". That's just as useless as saying "something causes something", except with some percentages added. Because the percentages don't refer to anything specifically, the statement is not really 'saying' anything. That is, it is not really true OR false as it stands.
Second, the falsity charge: if the Principle is claimed to be empirically true of ANY particular effect one cares to name, then it seems obvious to me that it is empirically false, as per your examples ... and the squillions more one could think of. (This incidentally is partly due to the inherent vagueness of the Principle: one can choose any particular set of causes and effects one wants.) Here's another example. The Principle says that 80% of earthquakes are due to only 20% of all the possible causes of earthquakes. Since tectonic plate movements in fact make up 100% of all the possible causes of 100% of earthquakes, the Principle is false in this case, and therefore false as a universal claim too.
But wait! A defender of the Principle might come back with this: "Ah, but the Principle is referring to POSSIBLE causes of earthquakes, and you really only referred to the ACTUAL past causes of them. It might be that of all the POSSIBLE causes, tectonic plate movements only make up 20% of THEM." Reply: but that must mean there are only four other POSSIBLE causes of earthquakes besides tectonic plate movements. Now the defender has a problem. They must come up with another four ... and ONLY four ... other possible causes of earthquakes. But more serious than that is the following question: What logical reason could one have to believing there MUST be four and ONLY four other causes? Why not three, or two, or one, or twenty, or fifty other possible causes? There is no logical reason for thinking it must be four and only four, and so there is no logical reason for thinking the Principle is true.
As for the last ditch line of defence, namely, the 'spooky cosmic coincidence factor' where the 80-20 split shows up all over the place ... this signifies nothing. One can find any 'split' one likes if one looks hard enough. And in economics, there is certainly nothing pervasive about it. It all depends on how you define what you are measuring. For example, the split between those not in poverty and those in poverty in Australia in 2005 was 80%-20%. Ooooo! But that is only if we set the poverty line to be 60% of median income. If we set it at 50% of median income, the split becomes 87%-13%. Still spooky? What about if we look at the poverty rates for given families over a 5 year period. Then it becomes 97%-3%. Spooky result gone. What about income inequality measures? Wikipedia refers to the spooky result of 83% of world income being distributed to 20% of the world's richest people. What if we were referring to Australia, however? Then we find it is 39% of national income going to 20% of the richest Australians. Spooky result gone. Everything depends on what is measured and how measures are defined. There is nothing magically real about the 80-20 split. I suspect this sort of thing is a function of the human brain's inclination to 'look' for patterns once they have been told there is one.
Of course the reason for this stupid idea, and the dodging and weaving required in defending it, is that it involves taking a numerical claim about one thing (Pareto's observation of the distribution of land, which is not a causal claim at all) and using those numbers in a totally unrelated context (causation). It is like noticing that, say, "approximately 95% of men who do the HSC increase their chances of employment by 5%", and then going on to formulate the Squire Principle: "approximately 95% of people who subscribe to the Pareto Principle are not taking out of their arses only 5% of the time." One has nothing in common with the other, except for the percentages. —Preceding unsigned comment added by 144.140.131.68 (talk) 05:36, 9 October 2009 (UTC)
What was the economics lecturere's name? Michael Hardy (talk) 00:56, 22 April 2010 (UTC)

No. No. No. The point is not that 80/20 are special. The point is that in many cases only a relatively small minority of causes are responsible for the majority of cases. 129.62.150.186 (talk) 22:42, 3 November 2009 (UTC)

Yes to the above comment. The Pareto principle is not meant to have mathematical rigor. It's a statistical rule of thumb, based on observation, that's worth keeping in mind when evaluating processes. In industrial applications, for example, causes of process irregularity are ranked in order of frequency, and the top contributors of error are worked on first. --SparkleCents (talk) 22:45, 21 April 2010 (UTC)

I think that we should have, if not a criticism section in the article, at least something pointing out that this is, to a great extent, vacuous. The 80/20 rule is, like a lot of management speak, just an empty sound bite (as even the defenders above seemingly admit). - 124.191.144.183 (talk) 12:53, 8 April 2013 (UTC)

I think for basically any event with no means adjusted input (which results in Gaussian distribution) only two types of outcomes are possible: one to one , or something eventually resembling the 80-20 rule. Take dice; 20% of the dice's face doesn't result in 80% of the reads. No. Instead it averages out to a one face one read outcome. Unless the dice is loaded, only then does it approaches that 80/20 rule. Another example would be to take the availability of women to men. On a blindfolded speed date event, each man will probably get a phone number. But introduce looks or wealth, the 80/20 becomes apparent. A 3rd example are soldiers; under normal conditions, one man takes out one man, so the larger army wins. But introduce an advantageous technology, it becomes another story. The introduction of a bias (especially one with a potential for becoming exponential or has a cascading effect)is what caused the difference. 80/20 doesn't apply to so many things; 20% of houses are not homes to 80% of the population. 20% of photons don't result in 80% of light. 20% of people don't cause 80% of speech. To sum it up I'd anecdotally proclaim the Pareto rule applies to 20% of input/outcome phenomena, if allowed a big margin of error.72.252.236.224 (talk) 08:51, 10 May 2014 (UTC)

I just made an account to add to this discussion. I am getting my doctorate in Economics right now, and I think this article makes some very strong claims without using any real statistical tools. This "rule of thumb" may be valid in many situations, but it is no more likely than a 70/30 or a 60/40 "rule."

I'll use the above poster's blind dating example. If we modify this so that Johnny Depp is one guy in the room and everyone else is either homeless, hungover, or smells of haggis, then I wouldn't exactly expect this "rule" to hold. To put it more rigorously, the proportion of the phone numbers that the men above the 80th percentile receive does not necessarily converge in probability to 0.2. The same rules of probability (and current social norms) would also hold for females giving out their numbers too. :)

Actually we can even go further: there is Lebesgue measure zero of this proportion converging to 0.2. This means that as the number of men and women goes to infinity that the probability that the men above the 80th percentile receive this proportion of the total phone numbers exactly is zero. This would hold as long as we aren't choosing the distributions of the preferences of men and women solely based on this rule holding, and as long as the parameters of each distribution have an infinite number of possibilities (like the mean and variance of a normal distribution).

It's fine if you guys are going for an article about something that holds sometimes and is portrayed as a buzzword for anything that would more accurately be called anything from a 60/40 rule to a 90/10 rule, but the wording of this article (to me at least) seems to imply that this will actually hold generally. Econom1 (talk) 06:00, 4 November 2014 (UTC)

## Adding up to 100%, and the fictious "k" value

As it has been discussed before, the numbers do NOT need to add up to 100%. The article is correct in stating:

There is no need for the two numbers to add up to 100%, as they are measures of different things, e.g., 'number of customers' vs 'amount spent'

However, this is contradicted by the following, which occurs both in the opening paragraph and is duplicated in the Mathematical Notes section:

Mathematically, where something is shared among a sufficiently large set of participants, there will always be a number k between 50 and 100 such that k% is taken by (100 − k)% of the participants; however, k may vary from 50 in the case of equal distribution (e.g. exactly 50% of the people take 50% of the resources) to nearly 100 in the case of a tiny number of participants taking almost all of the resources.

When that paragraph says k% is taken by (100 - k)%, it implies that the sum must be 100, since: (k) + (100 - k) = 100. This is the very misconception that the first quote clarifies. This imagined "k" value doesn't exist. Consequently, the rest of the discussion is rendered superfluous:

k may vary from 50 (in the case of equal distribution) to nearly 100 (when a tiny number of participants account for almost all of the resource). There is nothing special about the number 80% mathematically, but many real systems have k somewhere around this region of intermediate imbalance in distribution.

All of that is nonsensical, since there is no such thing as a "k" value. It appears that this entire paragraph was borne out of a misunderstanding of the subject. --70.26.246.197 (talk) 22:58, 2 December 2009 (UTC)

Your'e mistaken. It is true that they need not add up to 100%. But that is in no way contradicted by the assertion that they may be so chosen. Say the richest 1% of the people have 90% of the wealth. Then there would be some number k between 0.5 and 1, and some exponent n, such that kn = 0.01 and (1 − k)n = 0.99. I'll work out the values of k and n and post them here later..... Michael Hardy (talk) 07:10, 3 December 2009 (UTC)

## Other Applications section

I removed the following because it's anecdotal, without evidence or sources:

The Pareto Principle also applies to a variety of more mundane matters: one might guess approximately that we wear our 20% most favoured clothes about 80% of the time, perhaps we spend 80% of the time with 20% of our acquaintances, etc.

--SparkleCents (talk) 22:31, 21 April 2010 (UTC)

## Degree of Effort

A further extension of the Pareto Principal involves work effort. Generally with 20% of effort 80% of the results is achievable. It then requires another 80% of effort to deliver the further 20% of results. —Preceding unsigned comment added by 118.93.7.156 (talk) 08:14, 28 April 2011 (UTC)

Id compare this this to exams. That this statistic sorta holds true is only due to bell curve norming as an input for expected/predicted results. Tests are purposely made to get as many as possible to pass but allow as few as possible to shine. The test giver could have just as easily made the test a simple pass/fail.It all depends on how the effort or result is framed. In any case involving effort, I believe The law of diminishing returns takes precedence over Pareto.72.252.236.224 (talk) 09:13, 10 May 2014 (UTC)

## Forbes List

Just did some checks for a statistics class. I used the most recent top 10 Forbes ranking for the (i) world, (ii) the UK and (iii) Germany taking the total wealth of the top 10 as 100%. For the three cases the cumulative relative wealth distributions were very similar, but the 80:20 rule does not hold. Rather, it seems there is a "two third : one half rule", i.e. about two thirds of the total wealth of the top 10 richest are owned by the first five of them. Enlarging the data base (to the top 100, say) might possibly change things somewhat (if scale invariance does not hold), but I haven't checked. In any case, the reference to Forbes does not seem too valid. — Preceding unsigned comment added by HeinzT (talkcontribs) 13:19, 12 December 2012 (UTC)

Please note that reference 18 is not pointing to a live page. I've just removed it, but that leaves a citation missing. — Preceding unsigned comment added by 115.64.232.240 (talk) 23:57, 7 August 2013 (UTC)

## Scale invariance example

I removed the following passage because it is confusing and self-contradictory and possibly uses too small a sample size in the example. Certainly it could be reinstated, but let's fix the contradiction first, and perhaps discuss the sample size.

Due to the scale-invariant nature of the power law relationship, the relationship applies also to subsets of the income range. Even if we take the ten wealthiest individuals in the world, we see that the top three (Carlos Slim Helú, Warren Buffett, and Bill Gates) own as much as the next seven put together.[1] In this case, the rule does not apply since the top 20% (of the ten wealthiest individuals in the world) own about 50% of the wealth (of the ten wealthiest individuals in the world), and not 80% of it, as would be expected by the Pareto principle.

My problems with the material are:

• It says "Due to the scale-invariant nature of the power law relationship, the relationship applies also to subsets of the income range", then gives an example of this statement being not true (top ten richest, the top two (20%) do not have 80% of the total wealth found in the top ten -- not even close).
• Your sample size here is ten people. Pretty small. Is the implication that of the top five richest people, the richest one (top 20%) should have four times the wealth of the next four combined (80%) or the principle breaks down? Much more useful might be the breakdown of the top thousand richest, or something.

Maybe what's intended is to say "Due to the scale-invariant nature of the power law relationship, the relationship applies also to subsets of the income range given sufficiently large sample size" or maybe what's intended is to say "However, the relationship does not apply to subsets of the income range" or maybe "Sometimes he relationship applies also to subsets of the income range and sometimes it doesn't" or something else. Not clear. Herostratus (talk) 21:02, 13 May 2014 (UTC)

## Really poor cite

Reference number two is a link to a website which is pretty poor quality. It's certainly not good enough to be considered a useful citation for WP. It should really link to the Pareto paper. — Preceding unsigned comment added by 82.132.224.8 (talk) 18:10, 12 June 2014 (UTC)

## How precise is the 80/20 rule

It's important to say how many significant digits are known in the 80-20 rule in all of the various cases. — Preceding unsigned comment added by 174.56.105.36 (talk) 04:34, 24 September 2014 (UTC)

## Scientific slow down? No.

The Pareto principle is sometimes cited as an argument for economic cost increase as a cause of theoretical science not progressing as fast today as it used to. However, the more generalized a theory is, the more predictions it makes. Since there is not an 1:1 relationship between scientific level and cost of further progress, generalizable theories have a greater chance of at least one of their possible evidence being cheap compared to fine-tuning of old theories. So if cost was the problem, fine-tuning would have suffered even more than breakthrough progress.2.69.217.169 (talk) 07:43, 7 February 2015 (UTC)

I'm sorry, but I fail to see the relevance of this to the article. And the "Science" section that contains a whole 1 sentence. How does this apply to 80/20? — Preceding unsigned comment added by 71.13.251.83 (talk) 13:19, 16 April 2016 (UTC)

## "is claimed" ?

The vague remark that something "is claimed" is a particularly awkward line to have no citation for. Distingué Traces (talk) 20:06, 23 February 2015 (UTC)

## It' not all 80-20

Three percent of forest fires cause 97% of the damage. http://www.cbc.ca/news/technology/alberta-wildfire-science-background-1.3565932

Five percent of ICU patients use 33% of Intensive Care resources. http://www.newswise.com/articles/extreme-icu-study-finds-5-of-patients-account-for-33-of-intensive-care-should-receive-special-focus — Preceding unsigned comment added by 69.157.67.111 (talk) 19:48, 6 May 2016 (UTC)

Ten percent of drinkers drink 60% of the alcohol. I've seen this in a few places. http://www.inc.com/jeff-haden/the-top-10-percent-drink-way-more-than-you-think.html — Preceding unsigned comment added by 69.157.67.111 (talk) 19:45, 6 May 2016 (UTC)

-> Comment by GHR1961 So in those cases, it's a lower bound, considering the corresponding Pareto index.

## Microtransaction whale

"The overwhelming majority of app revenue comes from gaming apps, which accounted for 85% of the $34.8 billion global app market in 2015, according to App Annie." http://www.cio.com/article/3003154/ios/free-to-play-ios-games-aim-their-harpoons-at-money-bloated-whales.html "That means that when apple generate$12b in app store revenue in 2014 that over \$8 billion of that came from in-app-purchases to win free games. And that most of that came from probably 15 million or so people worldwide (although no one will release actual statistics, that's a guess at 5% of active iPhones, it's probably close to 3 million at 1% of active users)."

https://medium.com/swlh/mobile-app-developers-are-suffering-a5636c57d576#.qjffniw15 "The app ecosystem has an extremely harsh power law where app adoption and monetization are heavily skewed towards the top few apps. It’s nowhere near 80/20. In fact, it appears to be more like 99% of the value is centralized to the top 0.01%. Let’s call it the app store 99/0.01 rule."

http://www.recode.net/2014/2/26/11623998/a-long-tail-of-whales-half-of-mobile-games-money-comes-from-0-15

"Broken out as a percentage of only players who pay anything at all, Swrve’s report still points to whales. As the chart below shows, the top 10 percent of that 1.5 percent of paying players produces a dramatic upswing in revenue from even the 20th percentile group."

This counts right? --99.245.28.74 (talk) 18:53, 15 July 2016 (UTC)

## Quick comment

In "Mathematical notes" section, first paragraph: "For example, it is a misuse to state a solution to a problem "fits the 80/20 rule" just because it fits 80% of the cases; it must also be that the solution requires only 20% of the resources that would be needed to solve all cases."

"...that would be needed to solve all cases.": Here "all cases" means 80% of the effects considering the "20% of the resources" written just before... right? — Preceding unsigned comment added by GHR1961 (talkcontribs) 17:24, 24 February 2017 (UTC)

## Keystone species

Is this basically what happens in biology with Keystone species, or is that not exactly the case for here? --2607:FEA8:3420:92B:F1CF:4798:33F3:228D (talk) 04:44, 6 April 2017 (UTC)

## Counting the peas in the pods.

In reference to the photo at the top of the article: Do 20% of the pods have 80% of the peas? To me they seem fairly equal but would it be original research for me to count the peas in those pods? 185.188.233.56 (talk) 08:31, 25 May 2017 (UTC)