The gambler's fallacy
Вставка
- Опубліковано 27 січ 2025
- Flip a coin five times, and if you get five heads, you may begin to expect the next flip to land on tails. The "gambler's fallacy" doesn't just affect bets at a casino. Chicago Booth's Kelly Shue describes how this bias can impact decisions with important consequences. Read the story: chibooth.biz/1N....
Would writing a multiple choice test, and answering C three times in a row, then assuming the next answer is LESS likely to be C, an example of Gamblers Fallacy?
Based on the idea that the test was made by another human being we can assume that they were subject to the gamblers fallacy, and avoided long sequences of C. In that case answering the test accordingly, would not be fallacious. If the position of the answers were randomly generated by a computer or something, then yes you would be displaying the gamblers fallacy.
No, that is Reflexivity Theory. If you assume C again, that is Gamblers Fallacy.
@@Whorhin
I took a psych class in high school, and the teacher gave a pop test (not pop quiz, lol) on the first day of class, and legitimately every. single. answer *in a row* out of 30 or 40 questions was C. He wanted to see who would get caught up in intuitions of probability and change their answers and who would stick with their gut about the correct answers
No, because tests' answers are not random distributions.
Thats education fallacy😂
Though Prof. Shue correctly defines the gambler's fallacy the instances of biased judicial decision making are not examples of it. Judges similarly make worse decisions in the afternoon than in the morning. Those kinds of bias are the result of fatigue, and not due to a failure to comprehend basic probability or a random approach to judicial decision making.
There is one exception. For an exaple the sequence is red red black red black black and you double it ten times from an exaple 1 dollar to 51,2. Or in dice with better odds 49,5%, low low high low high high. Then you simulate 2-4 first attempts and if they miss you start on 1 dollar. Then it works.
1:45 she said the judge was more likely to grant access to the current one if s/he denied the second one twice when she meant to say opposite 😂😂😂
Past: Denied
Current: Granted
I think you are confused. She was talking about a lucky individual that is being granted due to the previous series of denial in the past.
@@kynikoi_6867 no hes right
I did not know that this fallacy is called the Gambler since Kenny Rogers sang that song. Good to know about flipping the coin whether it's heads or tails. : 0
Probability of heads or tails is 50/50 right? Due to this being the case, the recorded history of flips will show over an unknown length that a completely balanced coins results came out to be 50/50. This is just how it works. So for example, the 20 heads in a row that seemingly throws this out of balance WILL be made up for at some point if you recorded spins from that point on. If you also recorded thousands of spins right up to the existence of the 20 in a row, it would also show 50/50. It will balance out to 50/50 statistically over time AT ONE POINT over and over again based off a starting point recording any results before, after, or during current spins. This is an ever-flowing concept that's constant in the universe regardless if you have recorded history or not. Each flip is an independent event yet collectively, the results do show to be accumulative by nature of probability otherwise it wouldn't be 50/50 because long-term results would show a bias. 50/50 implies that the 2 given possible outcomes each account for 50% of the total 100% of the events guaranteed to come next. The 20 heads in a row does change the probability of the next spin just to a completely unknown degree in my opinion. If 20 heads came and you had infinite money, you could safely bet on tails over and over until you win, knowing that each loss means one closer to the win when is does hit tails because it will and it cant hit heads forever. Mathematically it cant because the results wouldn't be 50/50. Events have a probability to hit in a succession or order, for instance heads heads tails heads, that exact order has a probability to occur and this can be said for anything you can think of for things TO occur but the moment you mention the probability of things NOT hitting again after a completely abnormal amount of times mathematically, all of a sudden no math applies, and you're just an idiot believing a fallacy because after 10 spins you started betting the opposite. To be clear I highly recommend not playing anything that offers "50/50" because at the casino there is no such thing, its all slightly less. Roulette or even baccarat isn't 50/50 because of the zero and double zero and the banker commission. Betting against a streak is a good way to lose your money but I don't think its because you simply bet the opposite after witnessing a mathematical abnormality. You had a 50/50 chance of being right when you walked up there to place the bet either way, but after the abnormality it would be unlikely it would produce the exact same string of events again right afterwards, you can still be wrong but doubtful you'd see it again right after. If you put you money on seeing it again right after you'd end up a losing player still. Here's an idea, stop playing anything that's "50/50". Its a dirty game sometimes.
It becomes 50-50 over a huge number of sets.
Only apply if you don't have money to play many games, or the game was rigged from the start, otherwise math about chance is completely useless. Tossing 4 coins at once is not different from tossing 4 coins individually, the chance of getting expected result won't improve at all.
@@namvo3013 There's no way you're delusional enough to think that right?
Tossing 4 coins is astronomically different than tossing one if your goal is to eventually land on heads for example.
This is easily understood with an ab absurdum.
Let's say someone has a gun to your head and says, I'm going to flip a coin 700 million times in a row. If it eventually lands on tails you get to live.
Do you know why I wouldn't be nervous at all?
Inversely, let's say the person with said gun claims you get one coin toss. Heads you live, tails you don't live. How nervous would you be?
Would you, with a straight face, decline the 700 million attempts? I mean after all, getting the same consecutive result 700 million times in a row is pretty much the same as flipping a coin once right?
At some point you have to concede your way of thinking runs into a mathematical brick wall. 700 million and 4 have different likelihoods for the same reason.
Either the probability increases, or it doesn't.
The fact the odds eventually balance to 50/50 is evidence of what I'm saying, because getting heads 5 times in a row for example would never be canceled out unless the probability of tails increased due to this consecutive result of heads.
@@Ribcut Your 1 coin toss 700000000 times don't have better chance than 700000000 coins toss 1 time
@@namvo3013 I agree. Both of those examples are identical. It would be impossible for either experiment to not yield tails an absurd number of times.
The goal is to eventually land on tails, and this goal becomes more likely the greater number of attempts.
It doesn't matter if they're in order or all at once.
Reminds me of the group, always going for short term gains and losing long term in massive amounts. As if doing the same thing will hit that royal flush when they've gotten nothing but junk so far. Every percentage and probability approaches mean or average over time, however if that average or mean isn't possible from the equation it's a different story.
Land the coin on its side, never let them know your next move.
Does regression to the mean start to kick in if you got a 100,000 streak of heads?
Yeah at that point that shits cooked or something
I don’t understand, the odds of two coin toss being the same face isn’t 50%. For example if you do 10 heads, the chance of an 11 streak cannot be 50%. Maths isn’t my strong suit can someone please explain?
The chance of an 11 streak of heads/ tails isn't 50% (it's 1/2^11), but the chance of you getting a heads on the 11th flip after a 10 head streak is 50%. I think of it more of where you currently are in time. Imagine I ask you to flip heads 10 times in a row, you'd have to be very lucky to achieve that. On the other hand if you've already flipped heads 9 times in a row and only then I ask you to flip heads one more time, at that point in time you don't have to be that lucky to flip one more heads cause it's 50% chance. Sure after flipping the 10th heads and you look back of all the flips you'd think you're very lucky, but during the 10th flip you don't have to be as lucky as you are at the 1st flip to achieve the 10th heads as compared to 10 heads. I don't know if I explained that sufficiently hahahaha hopefully I did.
@@oceannlim This is exactly what I was trying to understand. Thank you for putting it so clearly.
If you've flipped a coin 10 times and it has landed on heads 10 times and continues to do so, the odds probably aren't 50/50.
Really confused how that analogy (the math of which was not explained) ties into authorities making less than fully informed decisions on what decides if an applicant should pass.
+Dustin O'Daffer Yes it is. Probability can't be viewed in the context of a small sample. Probability being 50% means that as we approach infinity in the number of tosses we make, we can expect the probability to close in towards 50%.
On your second question, they explain that even taking into account all other factors, there seems to be a bias or heuristic at play that unduely influences decision making. If you want to understand the math, you should read the paper, this video was of course not meant to go into all the details of the paper.
@Rex2212 I think you could easily get 60/40 with only a 100 flips. 10,000 flips I think it would be hard to get more than 5400-4600.
If every toss of a coin has the chance of getting heads of 50%, do 10 consecutive tosses have 50% chance of getting heads every time?
Not really. The chance of getting 10 heads in a row is 1/1024, but the chance of Getting any sequence of heads or tails is 1/1024, because there are 1024 combinations of heads and tails after ten flips. If I just flipped 9 heads in a row, the chance for me flipping another heads is 1/2, and the chance for tails is 1/2. The chance for getting heads or tails will always be 1/2, no matter what.
Good old-fashioned research, by that I mean unbiased. A similar video today would be politically biased and throw a loaded question fallacy into their research on effects of probability.
this seems to work with a 50/50 coin toss. But how would this relate to a die that lands on 6 over and over again?
Assuming a fair die, the probability of landing a given number is 1/6. Take a die that has rolled 6 for five times in a row. If you were to think that because of these past rolls, the probability of landing another 6 is less than 1/6 (and thus probability of landing 1 through 5 is greater than 5/6), then you are exhibiting the gamblers fallacy. A real world scenario would be a gambling game. For instance, you may be betting on what number a die will roll. If after 5 consecutive rolls of #6, you better heavier that the die will land on 1 through 5, then that's gamblers fallacy
@@allen9111 But the fact that it can't land on 6 forever tells you intuitively that you should bet on a different number.
AJ SADLER?
where did they get the evidence that the loan officers statement is true
what is the name of this paper?
may you please share the link of this paper or the title and journal name of this paper?
Ni571 where are you
cool, thanks for explaining
I too don't understand how this relates to decision making concerning non-random things like weighing the pros and cons of whether to do this or that. Only fools would be influenced by their previous decisions in this strange way. Oh, yes, that's the majority of people, they behave in very strange ways. Yes, it makes sense now.
wow look at the genius over here. get over yourself bud
Stop liking your own comments
It's a PARADOX, because if the number of heads and tails is equal, then if it goes out of balance, it MUST eventually go back into balance, so the odds MUST change, but you just never know when this is going to take place. If you flip a coin 50 times, and heads come up most often, then I guarantee that MOST OF THE TIME, in the next 50 flips tails will come up most often. Because probability DEMANDS it!
So I think rather "the gambler's fallacy" is that if the first 50 coin flips are mostly heads then the next 50 coin flips MUST DEFINITELY BE mostly tails. But instead it's MORE LIKELY THAN NOT that this will be the case - which is of course a paradox, because the odds are STILL 50/50 at the same time they are not!
@@ALoonwolf You seem to completely misunderstand how probability works : when it's said that the odds of a coin toss are 50/50, it means that, as you throw the coin an infinite amount of times, both events (heads and tails) converge on a 0.5 probability.
Your reasoning is flawed anyways : you're saying that "If the first 50 coin flips result in mostly heads, then during the next 50 coin flips tails will come up most often." but this is exactly what the gambler's fallacy is. What I mean by that is that since each coin flip is independant from the previous ones, the result of previous coin flips *will not* affect the result of future coin flips. The odds will alway be 50/50.
Let's imagine that you flip a coin 100 times, and that, through a massive stroke of luck, the first 50 flips result in heads (which is exactly a 1/(2^50) or 1/1'125'899'906'842'624 chance, so very, very unlikely but not mathematically impossible). The next 50 flips COULD be 50 tails (through the same amazing luck), or they could be any one of the other 1'125'899'906'842'623 possibilities.
It is in no way a paradox : the probability is always 0.5 for both events.
Off to the casino with you
No sir. Head/Tail/Head/Tail etc for 50 turns almost never happens, just like all heads for 50 turns almost never happens. Any string of 50 results is just as equally likely -or unlikely- as any other string of 50 results. The odds of the next flip being heads are not affected at all by the previous 50 flips, or the previous or 50 million flips. The next flip is always 50/50!
@@HoloTheDrunk So are there two probabilities running concurrently? The probability of a 50/50 for a heads and the probability for 10 heads in a row.
So a judge determining a complex case or a loan officer determining a complex approval or denial is the same as flipping a coin....Not buying that piece of information. But I do think that flipping a coin is always 50/50 and that people have imperfect biases. Misinformation here, tread carefully and actually think about it.
That's not what they are saying. They are saying the loan officer skews their own decision based on the last decision. The officer is not the dice, the officer is the gambler. They figure, "Well I don't want too many approvals in a row so I'd better deny this one to balance things out".
How slick. An Asian professor of Finance at Chicago Booth, giving us a lecture on asylum and social justice, based on the premise of ‘Gamblers Fallacy’. Well done.
Most people dont want to share and its a sign of greed and greed is always punished. Not because its wrong because its laws of nature. Karma is the consequences of action and the only thingtoconsider is if you want the consequence. Simply as that.
Isn't this fallacy just people failing to understand independent events?