2236: Is it Christmas?

 Is it Christmas? Title text: We've tested it on 30 different days and it hasn't gotten one wrong yet.

Explanation

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 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 accuracy is true, accuracy alone doesn't make a predictive device useful. In this case, the page 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.
• 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 to 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 xkcd Header text was changed to ask if there were someone that would like Randall's new book How To as a Christmas present.

Transcript

No*
*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 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 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.

Discussion

according Wayback Machine, the site still says NO even on Christmas. --valepert (talk) 21:56, 2 December 2019 (UTC)

Hmmm, maybe it will be fixed this year. I imagine everybody spammed the guy on twitter when it didnt work last year. 172.68.132.155 22:06, 2 December 2019 (UTC)
It uses the `new Date()` constructor to check whether it is Christmas, which uses your system time. Thus, the Wayback won't get anything, but changing your system time will. Kay? That's right, Jacky720 just signed this (talk | contribs) 22:14, 2 December 2019 (UTC)

Okay so looking at isitchristmas.com, there's a comment at the top of the page saying to look at the console, but I don't see anything in there, anyone know what that might be about? Also at the bottom of the html file, the bottommost <script> tag looks like it might contain code for a chat client? I don't know JS so I'm unsure, but I tried changing all the related "false" values to "true" that looked relevant and nothing happened, so idk. Maybe someone else can figure it out. 172.68.132.155 22:03, 2 December 2019 (UTC)

(I'm the dev.) Check back again closer to Christmas, on 12/23 or 12/24. Konklone (talk) 00:55, 3 December 2019 (UTC)
I mean, I kind of find that suspect. 172.68.47.234 01:39, 3 December 2019 (UTC)
seconded. that's a suspicious lot of javascript just to display yes or no, even with globalisation. --141.101.107.144 10:07, 3 December 2019 (UTC)
There are some other features besides globalization, but they only take effect closer to Christmas. Konklone (talk) 17:24, 3 December 2019 (UTC)

Well, one way to put it is that the overall probability of a false negative is 0.27%, which doesn't seem too bad, but the conditional probability of a false negative given that it's Christmas is 100%, which is horrid. --IByte (talk) 22:37, 2 December 2019 (UTC)

Of course it depends on your definition of "It's Christmas". The figure is right if you just mean Christmas Day and ignore the Julian vs Gregorian issue, but not you subscribe to "the 12 days of Christmas" = Christmas ... Just saying ;-) RIIW - Ponder it (talk) 23:53, 2 December 2019 (UTC)

Every year, people ask me about including multiple definitions of Christmas, but I think for my own sanity it's going to keep looking just at 12/25. ;) Konklone (talk) 00:57, 3 December 2019 (UTC)
There's an assumption here that people will only want to celebrate Christmas once a year. Might be different for someone living in a community with multiple religions who likes the holiday, or wants an excuse to take an additional day off work! --Marcus Rowland (talk) 11:08, 3 December 2019 (UTC)
But that would not be Christmas. Christmas is the day the Bible states Jesus was born (or at least if it is in the bible at all, the day Christians choose to claim Jesus was born). Thus only one day can be Christmas and no matter how you choose to celebrate and when or how many days, still only the 25th of December will be Christmas! --Kynde (talk) 12:41, 3 December 2019 (UTC)
Not only Bible doesn't say anything about when was Jesus born, the first Christmas celebration happened in year 336 and the date was likely chosen to match Roman Festival of Saturnalia. -- Hkmaly (talk) 23:49, 3 December 2019 (UTC)
I'm Catholic, Christmas day is one day a year, but Christmas (the liturgical season) ends on The Baptism of the Lord, which varies but can be as late as January 13th. 172.69.63.91 15:19, 3 December 2019 (UTC)
There are at least two Christmas days, as noted previously - the Western one on 25th December, and the Eastern Orthodox version which is based on an older calendar and is now in January according to the Western calendar, because their calendar has slipped compared to ours. But if someone wants to celebrate both there is really no reason why not - they don't have to be followers of the religion to want to take a holiday. Or if you want another reason, in communities with followers of both versions of Christianity, there will be days when the followers of one or another version are celebrating Christmas and businesses etc. are closed, which will affect everyone regardless of which religion they follow, if any. --Marcus Rowland (talk) 15:33, 3 December 2019 (UTC)

How do we know Randall's version is wrong on Christmas? Has anyone set their system clock to Dec 25 and checked it? Barmar (talk) 02:14, 3 December 2019 (UTC)

Have tested; Randall's displayed "no" as per main image. isitchritmas displayed "yes" as per explanation. RedHillian (talk) 02:44, 3 December 2019 (UTC)
It may be Randall uses Server-Time, and not client-time like isitchristmas.com - Also, in the official Calendars, there are two Christmas days, first and second day of Christmas. I think Randall went only for the gifting day, which is different in many countries anyway. My kids get there presents on Christmas eve (24 December), their friends mostly on 6 December (Sint Nicolas) (or the evening before). 141.101.77.62 07:45, 3 December 2019 (UTC)
Based on his 99.73% accuracy comment, I assume the comic is a static image, and will always show "NO", even on Christmas day. I think that is the whole joke, that his comic is correct 364 (+ leap days) of the year, when it is not Christmas. 172.69.34.158 08:23, 3 December 2019 (UTC)
Of course this is a static image, that is the whole point. ;-) In Demark we also have the is it Friday which at least is easier to check if it works, as it changes from No (Nej) to Yes (Ja) once a week. As mentioned above in Denmark the isitchristmas answers in Danish with a Nej. --Kynde (talk) 08:34, 3 December 2019 (UTC)
Yes, that is clearly what the comic implies - that it is static and will be wrong on Christmas. But to be fair, it would very much be Randall's style to instead change this comic to say "YES" on christmas, and then ALSO to change it again so that it is wrong on some other unexpected day... :) Praxiq (talk) 05:23, 4 December 2019 (UTC)
Ummm, yeah, it ALWAYS says "No". LOL! Which for 364 days out of 365 is true. I haven't checked the math, but presumably 364 divided by 365 gets the percentage he states. It seems to me he's making fun of how silly a concept that website is. :) NiceGuy1 (talk) 05:15, 5 January 2020 (UTC)

About "when is christmas". Christmas consists of two days, which makes the calculation of predicted correctness too optimistic. (99,45% for 363/365 and for 364/366; 99,38% for 363/365.25)141.101.77.62 07:45, 3 December 2019 (UTC)

No only the 25th of December is called Christmas Day. All other days are thus not Christmas, but just days you choose to celebrate that the 25th is Christmas! --Kynde (talk) 12:41, 3 December 2019 (UTC)
Ah, learned something today. In Germany and several neighbouring countries Christmas has two days, a first and a second Christmas day (25th and 26th of December). As Christ was born in the night, both days around that night get celebrated. Stupidly I assumed this would be the case in English speaking countries too. So, the UK and a few former colonies have only one day of Christmas, and a "Boxing Day". Thanks for making me spill my time on wikipedia and other sources. ;) 172.69.54.39 08:10, 4 December 2019 (UTC)
Actually, there's no Boxing Day in the United States, they refer to it as a Canadian "holiday" (speaking as a Canadian). It rather surprised me that they don't have the term, though we don't "celebrate" it in any way (except stores having Boxing Day Sales), it's just a name for the day after Christmas, like the opposite of Christmas Eve. Now I'm surprised to hear some places celebrate over 2 days. (As usual when such cultural differences come up, it's safest to assume the meaning is with how things are in the States, as that's where Randall lives. Not always true, but usually). NiceGuy1 (talk) 05:32, 5 January 2020 (UTC)

"In the western world celebrated on 25th" is not correct. For UK,US and Canada and probably a lot others it might be true, german and I think spanish speaking countries not.

No no. Even when you celebrate Christmas Eve on December 24th (as also in Denmark) we call the 25th Christmas Day. Even if you also celebrate the 2nd Christmas day, there is only one day a year that it is Christmas, and that is the 25th of December. That is when Christians claim Jesus was born. Of course I celebrate the Winter solstice (and accept that I do this a few days off, since that is when I can have the day off.) In Denmark we count the days up till the 24th and celebrate in the evening, (as the Vikings always did, because the new day began the sun set.) We do not look at the 12 days after. --Kynde (talk) 08:34, 3 December 2019 (UTC)
The word "celebrated" or "to celebrate", means (according to Lexico/Oxford): "Acknowledge (a significant or happy day or event) with a social gathering or enjoyable activity." This activity is not set as a single day for the western world. E.g. in Germany it is acknowledged with social gatherings and enjoyable activities for at least 3 days: Christmas eve (24th), First Christmas day (25th) and second christmas day (26th December). Additionally not all the western world (is that even properly defined?) does acknowledge the birth of christ (significant/happy event) by social gatherings or enjoyable activities (e.g. you, as you stated yourself.)--Lupo (talk) 09:59, 3 December 2019 (UTC)
The word celebrate is not present in the comic. That is about the question if today is Christmas. When you choose to celebrate Christmas has nothing to do with what day is Christmas, and also other religious holidays has nothing to do with Christmas at all. Do not mix up different issues here. This comic is only wrong on 25th of December which is Christmas day. All other days you choose to celebrate Christmas is not Christmas Day! And thus it is not Christmas! --Kynde (talk) 12:41, 3 December 2019 (UTC)
So you also agree, that the statement in the explanation is wrong? I will change it.--Lupo (talk) 12:47, 3 December 2019 (UTC)
Ummm, no, you shouldn't twist people's words to pretend they agree with you, sorry. Kynde was clearly arguing the other side. Again, as stated above, whether it IS Christmas is about what DAY has the name Christmas Day. The very name uses the singular "Day", not "Days". What day(s) people use is irrelevant. I suspect you can check the math to see that Randall is referring to one singular day. It was more accurate before, you shouldn't have changed it, especially against the majority "vote" like that. Christmas Day is December 25th, one day, period. NiceGuy1 (talk) 05:58, 5 January 2020 (UTC)
As you can see, I made [this edit] after my comment here, in which I tried to explain the comic to international readers. I changed celebreated to observed (as Kynde has said the comic is not about celebrating, but about wether it is the holiday), and changed "western world" to the US. As you said yourself in other comments here, it differs a lot between the US, canada and places that "celebrate" over 2 days. So to me it is not twisting his words, but now, with a few weeks passed, I see my comment as a bit of passive agressive as well, and would like to apologize to Kynde. I hope the current version (in which I have introduced the paragraph that the comic is only true for 1-day definitions, and which also does not any longer make claims about "the western world") is acceptable to all. --Lupo (talk) 07:37, 6 January 2020 (UTC)
The XKCD page has the additional advantage of being equally accurate whether you follow Western or Eastern Orthodox calendars, or whatever calendar you choose! --Quantum7 (talk) 08:39, 3 December 2019 (UTC)
I didn't mean to stir up such a hornet's nest, but there is a possible distinction between the Season of Christmas (from Christmas Day to Twelfth Night) and just Christmas Day and I was referring to that. Nobody talks of Advent Day or Lent Day and Easter is clearly Good Friday, Easter Day and Easter Monday, even if the Eggs are meant for just Easter Day. Just saying ;-) RIIW - Ponder it (talk) 19:04, 3 December 2019 (UTC)
It's true that only the 25th is called Christmas Day in the U.S. (for example), but it's also true that the 24th is called Christmas Eve. I see no reason to assume that only the former can lay exclusive claim to a 24-hour "Christmas". If you want only a 24-hour period, I'd argue for a period cutting across both dates. (As a Minnesotan I may be culturally influenced by an apparent Scandinavian tradition.) 162.158.214.148 20:43, 4 December 2019 (UTC)

Reminds me of 937: TornadoGuard in subject matter.--Henke37 (talk) 09:47, 3 December 2019 (UTC)

I was mostly reminded of 1132: Frequentists vs. Bayesians. --Lupo (talk) 10:06, 3 December 2019 (UTC)

Is the comic closely enough related to be put into Category:Statistics?--Lupo (talk) 13:21, 3 December 2019 (UTC)

No. There is nothing statistic here. It is just 1/365 giving the error percentage. That the explanation goes into statistics to tell how bad this version of is it x-mas is, does not make the comics topic statistics --Kynde (talk) 21:01, 3 December 2019 (UTC)

Some other sites with the same idea: http://ismycomputeron.com https://shouldiblamecaching.com http://iscaliforniaonfire.com http://www.ismycomputeronfire.com/ Ahiijny (talk) 14:47, 3 December 2019 (UTC)

Should I panic yet? -boB (talk) 19:58, 4 December 2019 (UTC)

Also this flowchart: 1723: Meteorite Identification - I am currently to absent-minded to get a proper wording for the relation right. --Lupo (talk) 14:57, 3 December 2019 (UTC)

Not seeing the similarity. The flowchart he mocks there is actually a real flowchart, which helps you realize that the stone you find is not a meteorite. Because it never is... until it is. --Kynde (talk) 21:01, 3 December 2019 (UTC)
Both ask a question where the answer is known to usually be "no" and then go ahead, to proclaim "no" without further checks. --Lupo (talk) 08:37, 4 December 2019 (UTC)

Maybe it's just me, but I took this comic to be a mockery of a vastly over-complicated solution to a simple problem. I mean, have you looked at the source code for isitchristmas.com? Crazy! 172.69.63.195 20:47, 3 December 2019 (UTC)

I think he just mocks pages in general that tells you something you know. I do though enjoy to sometimes look at the, is it Friday page, but only when I know it is Friday, as it is always nice to think about the weekend is near, early Friday morning. ;-) --Kynde (talk) 21:01, 3 December 2019 (UTC)

The average number of days in a year in the Gregorian calendar is exactly 365.2425, not 365.25. Leap years are skipped in years divisible by 100 except in years divisible by 400. Programmerjake (talk) 18:12, 4 December 2019 (UTC)

I imagine this chap https://en.wikipedia.org/wiki/Andy_Park_(Mr._Christmas) will be very upset to find out NO days are Christmas!Daemonik (talk) 11:22, 5 December 2019 (UTC)

Interestingly, isitchristmas.com has a link to this XKCD at the top. Is this worth noting? 173.245.54.43 01:16, 26 December 2019 (UTC)

Am I the only one who again got adds in between paragraphs on this wiki? --Lupo (talk) 08:37, 4 December 2019 (UTC)

Nope, but I only get them occasionally. Elektrizikekswerk (talk) 08:53, 4 December 2019 (UTC)
Right now I'm seeing a lot of them in the discussion section that's transcluded on the main comic page, but not if I go directly to the Discussion Talk: page. See also Talk:2220: Imagine Going Back in Time/Ads (the previous discussion about this). Ahiijny (talk) 14:50, 4 December 2019 (UTC)
I've had to take a hiatus from XKCD so catching up now, I think this is my third comic with ads. And about every second paragraph, in bothe the explanation and here in the comments. NOT good. We may need to bring back the Ads comment section. :( NiceGuy1 (talk) 05:32, 5 January 2020 (UTC)

The Asterisk

One thing nobody has mentioned yet: the asterisk after "NO" is very distinctly drawn with 4 lines, for 8 points. The vertical points are longest; the horizontal points second-longest; and the diagonal points are shortest. This is a very popular way to represent the star of bethlehem, especially as a christmas tree topper - see https://imgur.com/a/B9det9a for lots of ornaments that look an awful lot like that asterisk! I'm sure it's deliberate, given the Christmas theme of the comic. (I intended to update the explanation to point this out, but as a new user I can't upload the image here, and I'm not sure if linking to external images is good etiquette. So instead I'm posting here, in case someone else wants to update the explanation. The linked imgur is a collage of my own creation, but it's made up of 6 copyrighted photos. They're marketing photos from Amazon, Wayfair, and other sites, of products for sale - I believe using them educationally as a representative sample of ornaments depicting the star of bethlehem should constitute fair use, but IANAL.)

Praxiq (talk) 05:18, 4 December 2019 (UTC)

Probably won't change

See the asterisk that says 99.73 accurate? It probably won't change, that's the joke. 172.69.22.164 20:15, 12 December 2019 (UTC)

Can confirm that it did not change. 173.245.54.43 01:16, 26 December 2019 (UTC)
Can someone else confirm? Because I just noticed, that someone made the transcript here change for that day, which I think is wrong.--Lupo (talk) 08:13, 6 January 2020 (UTC)