Being taught Statistics at University was a real eye opener on this...
this post was submitted on 09 Oct 2024
577 points (96.9% liked)
Science Memes
10652 readers
3169 users here now
Welcome to c/science_memes @ Mander.xyz!
A place for majestic STEMLORD peacocking, as well as memes about the realities of working in a lab.
Rules
- Don't throw mud. Behave like an intellectual and remember the human.
- Keep it rooted (on topic).
- No spam.
- Infographics welcome, get schooled.
Research Committee
Other Mander Communities
Science and Research
Biology and Life Sciences
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- !reptiles and [email protected]
Physical Sciences
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
- [email protected]
Humanities and Social Sciences
Practical and Applied Sciences
- !exercise-and [email protected]
- [email protected]
- !self [email protected]
- [email protected]
- [email protected]
- [email protected]
Memes
Miscellaneous
founded 2 years ago
MODERATORS
So...
- normal people are scared because they fall for the gambler's fallacy,
- mathematician is feeling fine because a 50% chance is a 50% chance,
- and the scientist is feeling extra fine because the experimental data shows that the surgery is actually safer than 50%
Did I get it right?
The mathematician is probably feeling fine because he is computing the conditional probability of survival (otherwise fuck no I am not taking a surgery that has a %50 chance of killing me, that is way too much).
Nope.
- gambler's fallacy
- 50 is 50
- gambler's fallacy
Gambler's fallacy goes both ways. Just because the last 20 were successful doesn't mean the actual success rate is higher.
EDIT: Or you might actually be right and whoever made this is wrong.
Gamblers fallacy does go both ways. There's also a thing in gambling, not part of the gamblers fallacy, more of a superstition thing, that there can be runs of, what is more or less luck. The gamblers fallacy would have you believe that after 20 successes, a failure is "due to happen". According to math, that's not the case, and in the event of something that requires skill to execute, almost nothing is just luck or statistics.
So the last one isn't so much the gamblers fallacy, if anything it would be the superstition that the run of successes will continue; however scientists will look at this more as a game of skill. While 50% of all patients who have the procedure do not survive, or whatever, the last 20 of this doctors patients have survived. Clearly their skill for the procedure is above average. Even from a statistics perspective the rate might be 50% but you're in the hands of a doctor pushing that number up to 50%, rather than dragging it down to 50%. So on all fronts, if you hear this, bluntly, you have an unknown risk level, somewhere between 50% and 0%.
If the surgeons been successful the last 20 times, then they're probably doing something different or much better at it than the pool of surgeons the data was pulled from.
You’re only looking at the last 20 and not all of the the surgeries. Imagine the surgeon has had 10020 surgeries and the first 10000 are an even 50/50 split, but the last 20 are all successful. Would you really think you have a higher chance because the last 20 were successful?
Yeah. The odds of 20 coinflips coming up heads is 1 / 2^20, 1 in 1048576. Meaning, it would take over one million tries of 20 coinflips to get one that comes up all heads on average. Our surgeon has done the surgery 10000 times, you say. That's 500 runs of 20 coinflips. The surgeon could do 2000 times the amount of surgery they've already done and not expect a given run of 20 surgeries to succeed. Granted, odds are slightly higher looking for a string of 20 successes in one run of 10000 than looking for one successful run of 20 in 500 runs of 20, but I don't believe those odds are higher by one or more orders of magnitude. I don't remember enough of my statistics class to calculate it though and I'm too lazy to look it up. In any case, either that last 20 is an INSANELY lucky run, or the surgeon has improved drastically, and it's far far farore likely the surgeon has improved or found a way to improve the procedure.
The difference between this and the gambler's fallacy being it's not using past results to justify the next, we're using past results to make a hypothesis that the conditions have changed. If you have a slot machine with a 1% chance to pay out, going 200 or 300 pulls without winning wouldn't really be unusual. If it didn't happen for enough tries that you reach the same odds as the surgeon hitting that 20 coinflips run, I'd say your 1% chance slot machine is broken or faulty -- you're so far out of the normal distribution your data points is in a different zip code
I'm pretty sure somewhere your math is going wrong, but that's not really relevant because I don't think the hard probabilities are particularly relevant to my point anyway. We've established that if the surgeon does 10000 surgeries with 50% success but then does 20 with 100% success then that's not luck, that's skill. Just for a second I'm going to bend the original statement.
Instead of the last 20 being 100% the surgeon some point has done 20 successful surgeries in a row. Let's say he did 5000 surgeries with 50% success, then did 20 with 100% success and then did the next 5000 with 50% success. Would you still call that skill or is it now luck? I think it would be misleading to call it skill because their success rate didn't change after the 20 surgery streak.
But when we put those 20 to the end it becomes skill? So just because we don't know the success rate of future surgeries we're supposed to believe he's better than 50%? Call me a skeptic but it doesn't really matter how probable or improbable those 20 surgeries are, I wouldn't consider that an indication of skill. If someone flipped tails 20 times in a row I wouldn't go "wow, what a skilled coin flipper"
Depending on what you're treating, 50% sounds pretty good.
I remember when I went for my last surgery and I was signing all the consent forms, my doctor was emphasising the 17% chance of this known lifelong complication, and the increased 4% chance of general anaesthesia fatality (compared to 1 in 10,000 for general public).
My mum was freaking out because when she had the same surgery she'd been seen much earlier in the disease process, she wasn't expecting such a "high" risk of complications in my care.
But all I was hearing is that there's an over 80% chance it will be a success. Considering how limited and painful my life was by the thing we were treating, it was all no brainier, I liked those odds. Plus my condition is diagnosed 1 in 100,000 people, so how much data could my surgeon really have on the rate of risk, the sample size would be laughable.
Still the best decision of my life, my surgeon rolled his skilled dice, I had zero complications (other than slow wound healing but we expected and prepared for that). I threw my crutches in the trash 2 years later, and ran for the first time in my life at 27 years old after being told at 6 years old that I'd be a full time wheelchair user by 30.
the gambler's fallacy is the opposite of what applies to #1
"is the belief that, if an event (whose occurrences are independent and identically distributed) has occurred less frequently than expected, it is more likely to happen again in the future (or vice versa)." -per wikipedia
#2 is an optimist? A glass half full type of guy maybe.
#3 i'd guess is inferring that the statistics are based on an even distribution where the failures are disproportionately made up of by the same select few surgeons. or maybe that's #2 and the scientist actually know the theory of how the procedure works in addition to what #2 knows about statistics and distributions.
1 in 100,000 not 10,000 (anaesthesia deaths)
Well either it's 1 in 10,0000 in my country, or the department of health, my surgeon and GP are all missing a zero... which isn't more likely because there's no way our risk would be that much higher than the UK, and the NHS public health resources is 100,000.
I can't load that page so assuming it just says 1 in 10,000. That does seem like an insane difference and actually quite a high risk imo given how many people undergo surgery.
That's awesome. I'm glad everything went so well. Here's to a healthy and long life! Even the idea of going under is terrifying to me. You definitely had some courage with that attitude and that's really admirable.
It helps that I'd had several previous surgeries, so I'd had practice at keeping my cool. Plus my surgeon was in discussion with me for months with multiple consultations to really understand the issue, I don't think I could have possibly been in safer hands.
Gamblers fallacy or law of large numbers...
law of large numbers does not imply gambler's fallacy
10 points to Gryffindor
Yay!
So mathematicians and scientists cannot be normal people?
No, never.
That’s mean
That's median
That's mode
That's standard deviation
Until vastly larger numbers of people get trained in Advanced Mathematics and Degree Level Scientific disciplines, the human norm will never be anywhere near Mathematicians and Scientists.
It's quite literally abnormal to be a Scientist or Mathematician.
Then again, so is being Homeless or an Olympic-level Athlete.
no.
view more: next ›