A community for a webcomic of romance, sarcasm, math, and language.
Title text:
The plural of anecdote may not be data, but the singular of data is anecdote.
Transcript:
Transcript will show once itβs been added to explainxkcd.com
Source: https://xkcd.com/3262/
The reaction to sports pseudo-stats is what really separates casual viewers from real fans. It's the only way to raise stakes on otherwise forgettable games.
"This team is on a 5 game win streak": π₯±
"This player has never lost an away game in June": π―πΏ
Here on Austrian TV, there was an excellent example of this just yesterday during the match against Argentina, where the commentary helpfully told us at some random point in the second half that in the last four world cup matches Austria played in, they scored a goal during extra time, the implication being that that would probably happen again now (it didn't)...
The last four world cup matches Austria played in were... one in 2026 and three in 1998.
Cricket stats are getting so stupidly specific it has become a meme at this point.
It basically boils down to something like "Most runs scored by an Indian batsman on a Tuesday while batting first in overcast conditions.
Reminds me of that show Air Crash Investigations, you might know it as Mayday.
They did the episode on the collision of two 747s on Tenerife fairly early in the show's run, so they'd shot the "worst aviation accident in history" wad. And yet the format demanded they quantify the subject's exact place in history. So they start going "It was the worst aviation accident involving an American-manufactured plane flown by a non-American crew in American airspace to occur on a summer Tuesday."
P-hacking is the academically problematic practice of attempting to come up with a question for which the data offers a significant p-value (probability value), as opposed to correct scientific analysis in which a question is formulated clearly and then answered with data.
It took a while to parse this comic, but with the explanation it's probably much easier to understand for anyone who doesn't know what P-hacking is.
One thing you can use p-hacking for is that if you want to prove vaccines are bad, give a bunch of kids vaccines and measure 20 different vital indicators. Then theorise that the vital indicator which got worse was caused by the vaccines.
Reduce the sample size by increasing qualifying parameters until you find a dataset that matches your hypothesis in such a way that the research grant will be approved.
Sometimes even worse, which is to collect a raft of data testing one hypothesis, and then realize it all came up empty, and so go looking for any data you can form a new hypothesis from that matches the data you already have.
Thanks for that. I'd never heard the term before.
It sounds a little subjective though? Are there features that can be used to quantity how "P-Hacky" something is?
I feel like a sports state of "a team tends to lose if thier top scoring player in the first quarter is injured before the end of the first half" has a lot of specific weirdness, but my intuition drives that this specifically could be a very legitimate observation.
How do you draw the line?
Usually p hacking doesn't come from 1 constraint, especially a well explained one, but instead comes from adding a couple or completely unexplained constraints (like a team losing more if their coaches wife is in one section of the stands or another) because at that point it's decreasing the number of samples (times you have as a reference) to force a significant result.
So usually for sports p hacking is stats about 1 team only, rather than a general stat about the sport. Preferably a restriction on the other team, then a follow up game based restriction so it seems plausible to the viewer.
p hacking
you what
I'm normally not a praying dude, but if you're up there, save us Jungkook
Ok, that helps. I think you're saying the issue arises when the set of constraints limit the observed events to a number too small to draw appropriate conclusions from.
I'm hesitant to shy away from "bizzare" constraints. If there are enough data points for that scenario to draw some statistical correlation... then that just is the reality even if we can't explain it (yet).
If the coaches wife sits in a different section for 20% of the games, and they disproportionately lose when she sits there... that's the correlation.
Could be she sits in a further away section if she's pissed after a fight with her husband the night before, which is a signal the coach also had a bad night, and is fatigued and unfocused during the game now.
But yeah, you need enough observed instances.
This kind of data recognition is wasted on sports. I'd kill to have a commentary recognizing a player's top deck:whiff ratio during a mid game dig in commander. I don't care about optimized turns to win, I wanna see storm/cascade failure rate and scoop on sight %.
This is Legitimately my favorite thing about baseball. The wild stats they come up with.
Baseball stats are so unnecessary but God damn do I love them.
Can you give some examples (I never watched Baseball)
Not a perfect example, but stuff like this: Tony Gwynn did not strike out in back to back games during the 1992 season
If you watch a minor league baseball game, the announcers will discuss the game and give out hilariously specific information relevant to the game. When a batter comes up, they'll talk about the last time he encountered this pitcher, or how he does against left handed pitchers in general, or how this pitcher does against this team, or how this batter does later in the week. All of it could be relevant, but it becomes hyper specific.
BTW, I mention minor league baseball because you can watch it for free, as opposed to major league baseball, which costs a crapton of money, but you get the same experience in the major leagues. If you want to try a game, download the milb app and check it out. https://play.google.com/store/apps/details?id=com.bamnetworks.mobile.android.gameday.milb
That actually sounds kinda funny.
It's a blast! Some people take it super seriously, but I prefer it when teams lean into the fun of the game.
Iβve complained about this for years. This is only in America btw. In other countries they just watch and donβt care about the statistics.
This is only in America btw.
Literally comments all over this thread talking about Austria and Australian Cricket and Argentina.
Must be that it is only countries that begin with A and the stats are just working through alphabetically.
That's not entirely true, us Germans love statistics but more like goal to attempt rate or something like this. Received passes, etc. Not whatever they're trying to do here.
Sorry, what I said was an exaggeration, but yeah keeping track of shots on goal, passes, etc are just basic parts of the game. Essentially the same as keeping score. Americans donβt do that. They literally keep track of exactly stuff like the xkcd shows. Itβs not even an exaggeration.
Here in the UK we get a bunch of stats every game, but its pretty much the same stats every time with no fuckery. Shots on target (that would have gone in if it wasnt for that pesky keeper) and stuff.
You can count on F1 commentators (mostly Crofty) to pull out some obscure stats though
Exactly. I don't care if they could have saved it if their mother was born in August, statistically speaking, or if they had a goldfish at seven.
Yeah. This makes it even less enjoyable to watch
I can lose money betting on things of which my ancestors could never dream.