Statistical jokes

Most of these jokes were posted to Usenet news groups. Peo­ple who read such things col­lected them and put them on their web sites. I have shame­lessly bor­rowed them, edited them and posted them here for the light relief of other statisticians.

The Biologist, the Statistician, the Mathematician, and the Computer Scientist

A biol­o­gist, a sta­tis­ti­cian, a math­e­mati­cian, and a com­puter sci­en­tist are on a photo-​​safari in Africa. They drive out into the savan­nah in their jeep, stop, and scour the hori­zon with their binoculars.

The biol­o­gist: “Look! There’s a herd of zebras! And there, in the mid­dle: a white zebra! It’s fan­tas­tic! There are white zebras! We’ll be famous!”

The sta­tis­ti­cian: “It’s not sig­nif­i­cant. We only know there’s one white zebra.”

The math­e­mati­cian: “Actu­ally, we know there exists a zebra which is white on one side.”

The com­puter sci­en­tist: “Oh no! A spe­cial case!”

The Physicist, the Chemist, and the Statistician

Three pro­fes­sors (a physi­cist, a chemist, and a sta­tis­ti­cian) are called in to see their dean. Just as they arrive the dean is called out of his office, leav­ing the three pro­fes­sors there. The pro­fes­sors see with alarm that there is a fire in the wastebasket.

The physi­cist says, “I know what to do! We must cool down the mate­ri­als until their tem­per­a­ture is lower than the igni­tion tem­per­a­ture and then the fire will go out.”

The chemist says, “No! No! I know what to do! We must cut off the sup­ply of oxy­gen so that the fire will go out due to lack of one of the reactants.”

While the physi­cist and chemist debate what course to take, they both are alarmed to see the sta­tis­ti­cian run­ning around the room start­ing other fires. They both scream, “What are you doing?”

To which the sta­tis­ti­cian replies, “Try­ing to get an ade­quate sam­ple size.”


  • Sta­tis­tics means never hav­ing to say you’re certain.
  • Sta­tis­tics is the art of never hav­ing to say you’re wrong.
  • Vari­ance is what any two sta­tis­ti­cians are at.
  • 97.3% of all sta­tis­tics are made up.
  • It’s like the tale of the road­side mer­chant who was asked to explain how he could sell rab­bit sand­wiches so cheap. “Well,” he explained, “I have to put some horse-​​meat in too. But I mix them 50:50. One horse, one rab­bit.” (Dar­rel Huff, How to Lie with Sta­tis­tics)
  • Are sta­tis­ti­cians normal?
  • Smok­ing is a lead­ing cause of statistics.
  • 43% of all sta­tis­tics are worthless.
  • 3 out of 4 Amer­i­cans make up 75% of the population.
  • Death is 99 per cent fatal to lab­o­ra­tory rats.
  • A sta­tis­ti­cian is a per­son who draws a math­e­mat­i­cally pre­cise line from an unwar­ranted assump­tion to a fore­gone conclusion.
  • A sta­tis­ti­cian can have his head in an oven and his feet in ice, and he will say that on the aver­age he feels fine.
  • 80% of all sta­tis­tics quoted to prove a point are made up on the spot.
  • Fett’s Law: Never repli­cate a suc­cess­ful experiment.
  • Clas­si­fi­ca­tion of math­e­mat­i­cal prob­lems as lin­ear and non­lin­ear is like clas­si­fi­ca­tion of the Uni­verse as bananas and non-​​bananas.
  • A law of con­ser­va­tion of dif­fi­cul­ties: there is no easy way to prove a deep result.
  • This is a one line proof…if we start suf­fi­ciently far to the left.”
  • The prob­lems for the exam will be sim­i­lar to the dis­cussed in the class. Of course, the num­bers will be dif­fer­ent. But not all of them. Pi will still be 3.14159… ”

A Greater Than Average Number of Legs

The great major­ity of peo­ple have more than the aver­age num­ber of legs. Amongst the 57 mil­lion peo­ple in Britain there are prob­a­bly 5,000 peo­ple who have only one leg. There­fore the aver­age num­ber of legs is

(5000 x 1 + 56,995,000 x 2)/57,000,000 = 1.9999123.

Since most peo­ple have two legs…

The Man who Counts the Number of People at Public Gatherings

You’ve prob­a­bly seen his head­lines, “Two mil­lion flock to see Pope”, “200 arrested as police find ounce of cannabis”, “Britain #3 bil­lion in debt.” You prob­a­bly won­dered who was respon­si­ble for pro­duc­ing such well rounded-​​up fig­ures. What you didn’t know was that it was all the work of one man, Rounder-​​Up to the media, John Wheeler. But how is he able to go on turn­ing out
such spot-​​on sta­tis­tics? How can he be so accu­rate all the time?

We can’t,” admits Wheeler blithely. “Frankly, after the first mil­lion we stop count­ing, and round it up to the next mil­lion. I don’t know if you’ve ever counted a papal flock, but, not only do they look a bit the same, they also don’t keep still, what with all the bow­ing and cross­ing themselves.”

The only way you could do it accu­rately is by tak­ing an aer­ial pho­to­graph of the crowd and hand­ing it to the com­puter to work out. But then you’d get a head­line say­ing, ‘1,678,163 [sic] flock to see Pope, not includ­ing 35,467 who couldn’t see him,’ and, believe me, nobody wants that sort of headline.”

The art of big fig­ures, avers Wheeler, lies in psy­chol­ogy, not sta­tis­tics. The pub­lic like a fig­ure it can admire. It likes mil­lion­aires, and million-​​sellers, and cen­turies at cricket, so Wheeler’s inter­na­tional agency gives them the fig­ures it wants, which involves not only round­ing up but round­ing down.

In the old days peo­ple used to deal with crowds on the Isle of Wight principle–you know, they’d say that every day the pop­u­la­tion of the world increased by the num­ber of peo­ple who could stand upright on the Isle of Wight, or the rain-​​forests were being decreased by an area the size of Rut­land. This meant noth­ing. Most peo­ple had never been to the Isle of Wight for a start, and
even if they had, they only had a vision of lots of Chi­nese stand­ing in the grounds of the Cowes Yacht Club. And the Rut­land com­par­i­son was so use­less that they were dri­ven to abol­ish Rut­land to get rid of it.

No, what peo­ple want is a few good mil­lions. A hun­dred mil­lion, if pos­si­ble. One of our inven­tions was street value, for instance. In the old days they used to say that police had dis­cov­ered drugs in a quan­tity large enough to get all of Rut­land stoned for a fort­night. *We* started say­ing that the drugs had a street value of #10 mil­lion. Absolutely mean­ing­less, but peo­ple under­stand it better.”

Some­times they do get the fig­ures spot on. “250,000 flock to see Royal two,” was one of his recent head­lines, and although the 250,000 was a rounded-​​up fig­ure, the two was quite cor­rect. In his pala­tial office he sits sur­rounded by relics of past headlines–a million-​​year-​​old fos­sil, a #500,000 Manet, a pho­to­graph of the Sul­tan of Brunei’s #10,000,000 house–but pride of place goes
to a pair of shoes framed on the wall.

Why the shoes? Because they cost me #39.99. They serve as a reminder of mankind’s other great urge, to have stu­pid odd fig­ures. Strange, isn’t it? They want mass demos of exactly half a mil­lion, but they also want their gramo­phone records to go round at thirty-​​three-​​and-​​a-​​third, forty-​​five and seventy-​​eight rpm. We have stayed in busi­ness by remem­ber­ing that below a cer­tain level peo­ple want odd­ity. They don’t want a rocket cost­ing #299 mil­lion and 99p, and they don’t want a radio cost­ing exactly #50.”

How does he explain the times when the fig­ures clash–when, for exam­ple, the organ­is­ers of a demo claim 250,000 but the police put it nearer 100,000?

We pro­vide both sets of fig­ures; the fig­ures the organ­is­ers want, and the fig­ures the police want. The pub­lic believe both. If we gave the true fig­ure, about 167,890, nobody would believe it because it doesn’t sound believable.”

John Wheeler’s name has never become well-​​known, as he is a shy fig­ure, but his firm has an annual turnover of #3 mil­lion and his eye for the right fig­ure has made him a rich man. His great­est plea­sure, how­ever, comes from the peo­ple he meets in the count­ing game.

Exactly two bil­lion, to be precise.”

(Miles King­ton, writ­ing in The Observer, Novem­ber 3, 1986)

Final Exam

A sta­tis­tics major was com­pletely hung over the day of his final exam. It was a true/​false test, so he decided to flip a coin for the answers. The sta­tis­tics pro­fes­sor watched the stu­dent the entire two hours as he was flip­ping the coin… writ­ing the answer… flip­ping the coin… writ­ing the answer. At the end of the two hours, every­one else had left the final except for the one stu­dent. The pro­fes­sor walks up to his desk and inter­rupts the stu­dent, say­ing, “Lis­ten, I have seen that you did not study for this sta­tis­tics test, you didn’t even open the exam. If you are just
flip­ping a coin for your answer, what is tak­ing you so long?”

The stu­dent replies bit­terly (as he is still flip­ping the coin), “Shhh! I am check­ing my answers!”

The Ten Commandments of Statistical Inference

  1. Thou shalt not hunt sta­tis­ti­cal infer­ence with a shotgun.
  2. Thou shalt not enter the val­ley of the meth­ods of infer­ence with­out an exper­i­men­tal design.
  3. Thou shalt not make sta­tis­ti­cal infer­ence in the absence of a model.
  4. Thou shalt honor the assump­tions of thy model.
  5. Thy shalt not adul­ter­ate thy model to obtain sig­nif­i­cant results.
  6. Thy shalt not covet thy col­leagues’ data.
  7. Thy shalt not bear false wit­ness against thy con­trol group.
  8. Thou shalt not wor­ship the 0.05 sig­nif­i­cance level.
  9. Thy shalt not apply large sam­ple approx­i­ma­tion in vain.
  10. Thou shalt not infer causal rela­tion­ships from sta­tis­ti­cal significance.

      Q: Why is it that the more accu­racy you demand from an inter­po­la­tion func­tion, the more expen­sive it becomes to compute?

A: That’s the Law of Spline Demand.

An alge­braic limerick:

    \[\int_1^{\sqrt[3]{3}} t^2dt \times\cos(3\pi/9) = \log(\sqrt[3]{e})\]

Which, of course, trans­lates to:

Inte­gral t-​​squared dt
from 1 to the cube root of 3
times the cosine
of three pi over 9
equals log of the cube root of ‘e’.

And it’s cor­rect, too.

The remain­ing lim­er­icks are taken from Mark Carter’s page which has since disappeared.

A math­e­mati­cian con­fided
That the Möbius band is one-​​sided
And you’ll get quite a laugh
If you cut one in half
’Cause it stays in one piece when divided.

A math­e­mati­cian named Klein
Thought the Möbius band was divine
Said he: If you glue
The edges of two
You’ll get a weird bot­tles like mine.

There was a young fel­low named Fisk,
A swords­man, exceed­ingly brisk.
So fast was his action,
The Lorentz con­trac­tion
Reduced his rapier to a disk.

Tis a favorite project of mine
A new value of pi to assign.
I would fix it at 3
For it’s sim­pler, you see,
Than 3 point 1 4 1 5 9

If inside a cir­cle a line
Hits the cen­ter and goes spine to spine
And the line’s length is “d“
the cir­cum­fer­ence will be
d times 3.14159

A chal­lenge for many long ages
Had baf­fled the savants and sages.
Yet at last came the light:
Seems old Fer­mat was right–
To the mar­gin add 200 pages.

If (1+x) (real close to 1)
Is raised to the power of 1
Over x, you will find
Here’s the value defined:

This poem was writ­ten by John Saxon (an author of math textbooks).

    \[((12 + 144 + 20 + (3 * \sqrt{4})) / 7) + (5 * 11) = 9^2 + 0\]

A Dozen, a Gross and a Score,
plus three times the square root of four,
divided by seven,
plus five times eleven,
equals nine squared and not a bit more.

In arc­tic and trop­i­cal climes,
the inte­gers, addi­tion, and times,
taken (mod p) will yield
a full finite field,
as p ranges over the primes.

A grad­u­ate stu­dent from Trin­ity
Com­puted the cube of infin­ity;
But it gave him the fid­gets
To write down all those dig­its,
So he dropped math and took up divinity.

A con­jec­ture both deep and pro­found
Is whether the cir­cle is round;
In a paper by Erdös,
writ­ten in Kur­dish,
A coun­terex­am­ple is found.

(Note: Erdös is pro­nounced “Air-​​dish”)

Three jolly sailors from Blaydon-​​on-​​Tyne,
They went to sea in a bot­tle by Klein,
Since the sea was inside the hull,
The scenery seen was exceed­ingly dull.

By Chris Boyd (see com­ments below).

    \[4 + (6! - 0.5(12^2 + (403 + 1))) = 2(15^2)\]

Four plus the dif­fer­ence between
The fac­to­r­ial of six and the mean
Of twelve squared and four
Hun­dred three (plus one more)
Equals dou­ble the square of fifteen.

Inter­est­ing The­o­rem:
All pos­i­tive inte­gers are inter­est­ing.
Assume the con­trary. Then there is a low­est non-​​interesting pos­i­tive inte­ger. But, hey, that’s pretty inter­est­ing! A contradiction.