Why is statistics so unintuitive?

[Anonymous No.9100415]

Like with math, once you understand a concept, it makes perfect sense and you can follow steps to prove it

But statistics is so unintuitive and confusing. Even after being explained a problem, most humans won't understand. Why?

https://www.youtube.com/watch?v=4Lb-6rxZxx0

[Anonymous No.9100421]

From pic related:

We are good intuitive grammarians — even quite small children intuit language rules. We can see that from mistakes. For example: “I maked it” rather than the irregular “I made it”.

In contrast those of us who have training and decades of experience in statistics often get statistical problems wrong initially.

Why should there be such a difference?

Our brains evolved for survival. We have a mind that is exquisitely tuned for finding things to eat and for avoiding being eaten. It is a horrible instrument for finding truth. If we want to get to the truth, we shouldn’t start from here.

A remarkable aspect of your mental life is that you are rarely stumped. … you often have answers to questions that you do not completely understand, relying on evidence that you can neither explain nor defend.

https://www.youtube.com/watch?v=KyM3d4gQGhM

[Anonymous No.9100428]

Mathematics is the practice of reasoning using known (or assumed) things.

Statistics is the practice of reasoning using data generated by unknown (or not completely known) processes.

Math is a purely deductive exercise: Its entire edifice is built using only a handful of axioms and the rules of deductive logic. If you accept these, then -- assuming there are no logical inconsistencies -- the entirety of mathematical knowledge is incontestable.

Stats also uses deduction (it has its own structured, beautiful theorems, such as CLT), but induction is just as important. At its heart, induction is about using data to provide evidence to support or refute some hypothesis, a claim about the real world (which is noisy and can't be summed up by axioms). There are no such claims in math.

I like to say that math is the logic of certainty, while statistics is the logic of uncertainty.

Statistics is also relatively new, and predicting the future via mathematical inference or probability is definitely not an innate ability in anything, whereas counting and adding is something almost all animals can do.

[Anonymous No.9100432]

A probabilist, a statistician and a mathematician sit in a cafe across from a parked sedan. They see five people get into the car. After a minute passes, six people exit the car. The probabilist says that the event is non measurable since the sigma algebra of the car accounts only for 5 seats. The statistician says that six people walking out qualifies as a significant statistic against the null hypothesis that there were no people inside the car to begin with. The mathematician says that if another person were to enter the car now, then there would be exactly zero people in the car.

[Anonymous No.9100437]

it's a pretty cool field

https://en.wikipedia.org/wiki/Secretary_problem

sometimes feels like magic

https://www.youtube.com/watch?v=ZWib5olGbQ0

[Anonymous No.9100464]

Honestly, I have no problem understanding statistics. I actually quite liking interpreting them. Maybe this is because of the ways I was raised

[Anonymous No.9100505]

>>9100432
why did he let 500 people look at his sock drawer?

[Anonymous No.9100678]

yeah it's a weird field

any statisticians here?

[Anonymous No.9100806]

>>9100415
>statistics is so unintuitive and confusing
Probability is what is unintuitive and confusing. Many proofs in statistics involve applying probability with some additional tricks and then the answer is trivially computed.

On the other hand, most of the fundamental results are so intuitively obvious that teachers or students just skip them. Then you get students that do statistics on intuition and are basically useless for anything more advanced.

Also, Monty Hall is probability, not statistics.

[Anonymous No.9100814]

>>9100806
you lost me at "probability"

[Anonymous No.9100820]

>>9100415
It took a surprisingly long time for probability and statistics to develop. As others said, on average, we're just really bad at it.
>https://en.wikipedia.org/wiki/Timeline_of_probability_and_statistics

[Anonymous No.9101003]

>>9100421
>that pretentious video
What's so irrational and complicated about that? It's simple Bayesian probability, you only have at your disposal the conditional information that she read fluently at age four, which correlates positively with intelligence on average (or so one intuits, as part of their prior distribution). All those hypothetical conditionals like anorexia and bad breakups are lower probability events that would only affect the distribution of intelligence if they were already known conditionals. But since you have no knowledge whether she's anorexic or not, you can only take the expected probability over all likely conditionals, by the law of total probability.