Editing 2236: Is it Christmas?

{{comic
| number    = 2236
| date      = December 2, 2019
| title     = Is it Christmas?
| image     = is_it_christmas.png
| titletext = We've tested it on 30 different days and it hasn't gotten one wrong yet.
}}

==Explanation==
https://isitchristmas.com/ is a popular simplistic website that informs the visitor whether or not it's {{w|Christmas}}. Christmas is a holiday observed in many parts of the world on December 25 of each year. At the top on the tab of the site in the browser it says "Is it Christmas?" with a large '''NO''' printed if it is not December 25, and a '''YES''' if it is December 25. This website asks the user's browser for the date, and updates accordingly if it is indeed Christmas. In addition, isitchristmas.com gives the answer in the language of your region (i.e. for a visitor from Canada, the site gives the answer in English and French to account for Canada's bilingularity, and in most other countries just their word for No will be shown). Since the page uses the browsing computer's time setting, it is possible to easily check that the page works by changing the date on the computer used to access the page to see the text change to Yes if you are reading it on December 25. This also means that the page is only as correct as the time setting on the computer used to view the page (so in case of connection problems, you may check your computer's calendar instead).

Here [[Randall]] spoofs the website. He claims to have made a competitor to isitchristmas.com which nearly always correctly tells if it is Christmas. The joke is that the comic will always display a static image reading '''NO''', even on Christmas Day, and that the rare incorrect answer is rare enough to not cause any concern.

Randall lists a rounded calculation of 99.73% for the precision of his prediction of whether or not it is Christmas. This number is accurate with or without including leap year. An average year is 365.24 days, meaning that he is only wrong 1 out of 365.24 days. So only 1/365.24 ≈ 0.2738% of the days would the prediction be wrong, resulting in a correct reply rate of 99.726%, which he has rounded to 99.73%. Using or not using the leap year will give the same result to three decimal places. 

This precision rate is only true for a definition of Christmas which lasts only one day, regardless of which day that is (see trivia). For any definition of more than one day of Christmas, the error rate would be higher than 0.2737%. (If one considered the traditional {{w|Twelve Days of Christmas}} to all be Christmas, then Randall's website would be wrong on all 12 days, or 3.29% of the year.) However, in the US, where [[Randall]] lives, Christmas is usually defined as the single day of December 25th. 
 
Although Randall's claim on {{w|Accuracy and precision#In binary classification|accuracy}} is true, accuracy alone doesn't make a predictive device useful. In this case, the page {{w|False positives and false_negatives#false negative rate|miss rate}} or false negative rate, that is, the percent of positive condition days (it's Christmas) that are predicted by the comic not to be Christmas, is 100%. In other words, it misses all actual events of Christmas. 

When building a model for rare events, a common mistake is to ignore the implicit cost function built into the standard prediction accuracy validity statistic for binary events. Prediction accuracy (# correct guesses/total guesses) assumes that false positives and false negatives are equally bad.  Given the implicit cost function of this performance statistic, the best-performing model is commonly a persistence forecast model--i.e., the optimal prediction model returns the most common value whatever the model inputs are. It's probably a better choice to optimize a model using a performance statistic which relies on a cost function that penalizes missing correct prediction of rare events more than it penalizes missing correct prediction of common events.

In fact, in most settings where a single outcome is a lot more common than any other one, predicting always that most common outcome would yield very high accuracy without any usefulness. It isn't hard to find examples even more accurate than Randall's:

* A useless test for AIDS giving always negative results would have an accuracy about 99.95% when applied to a random human, and even more if used in countries with low prevalence of AIDS.
* A website saying "You are not the cartoonist Randall Munroe" would be right for 99.9999999857% of humans.
* [https://knowyourphrase.com/even-a-broken-clock-is-right-twice A stopped watch is accurate twice a day] while a running watch is almost never accurate (and oddly, is more frequently correct the faster/slower it runs).  A watch that runs backwards is right 4 times a day.  If you make it spin at thousands of rpm it is right multiple times per second.  (A better metric would be something like the root mean square of the time error -- it's acceptable for a watch to be a little off, as long as it's not off by too much.)

The title text is a "proof" that his service works. He claims to have tested this on 30 different days and confirmed that NO is the correct result. Any date except Christmas would result in a correct result, and the comic was the first to be released in December 2019, so unless the test had run for almost a year, he would not even have had a chance to test this on Christmas Day. Since this is a joke, the comic will of course not change the Yes on Christmas Day, because then it would be 100% accurate, as is the page the comic mocks.

Being right on most days, but not the one that mattered was also the subject of [[937: TornadoGuard]].

At the same time this Christmas comic came out, the [[Header text|header text]] was [[Header text#2019-12-02_-_Into_Science|changed]] to ask if there were someone that would like Randall's new book ''[[How To]]'' as a Christmas present.

==Transcript==
:'''<big><big><big>No*</big></big></big>'''
:<nowiki>*</nowiki>99.73% accurate

:[Caption below the panel:]
:xkcd.com presents a new "Is It Christmas" service to compete with isitchristmas.com

==Trivia==
*Randall usually makes [[:Category:Christmas|Christmas comics]] around Christmas Day, but this year he has made two comics mentioning Christmas already by the 2nd of December 2019. 
**The first came two comics before this with [[2234: How To Deliver Christmas Presents]]. 
**Only two times before has there been released any Christmas related comics so close to Christmas without being released in the few days around Christmas Day (22-26 of December). See the explanation for Christmas comics.
*The calendar used by most of the world for civil purposes is the Gregorian calendar, instituted by Pope Gregory XIII of the Roman Catholic Church in 1582.  However, most Eastern Orthodox churches continue to use the Julian calendar for the purpose of their holidays; December 25th in the Julian calendar is January 7th in the Gregorian calendar for years between 1900 and 2100, so that is the civil date when those countries observe Christmas.  The author of isitchristmas.com is [https://github.com/isitchristmas/web/issues/67#issuecomment-29585160 aware that this is the case], but has chosen to recognize a single date (December 25th in the Gregorian calendar) as Christmas for the sake of simplicity.

{{comic discussion}}

[[Category:Christmas]]
[[Category:Statistics]]