Thought experiment for you "do I have a winning model" types
Collapse
X
-
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#36Comment -
MonkeyF0ckerSBR Posting Legend
- 06-12-07
- 12144
#38In this case, the probability that his coin is fair is 0.028, regardless of how many millions of other coins are being flipped for you around the world and how many millions of other results you expect to receive.
Whether you think about results obtained for you by MrED in relative or absolute terms does not make any difference either. Probability of his coin to be fair is still the same - 0.028.
Comment -
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#40Going to bed now with smile on face
Job is done
We did itComment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#41Thinking helps.
When p=0.028 the coins have a fair chance to be identical and your knowledge about coins being identical
can very well be fair.
If you "know" that coins are identical and one of them comes with p=0.00001 you should examine your "knowledge" pretty fricking hard.Last edited by hutennis; 08-16-12, 11:26 PM.Comment -
MonkeyF0ckerSBR Posting Legend
- 06-12-07
- 12144
#43Thinking helps.
When p=0.028 the coins have a fair chance to be identical and your knowledge about coins being identical
can very well be fair.
If you "know" that coins are identical and one of them comes with p=0.00001 you should examine your "knowledge" pretty fricking hard.
You just want to argue.
Good bye.Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#44
These two questions
1. What is the probability in the head of Mr.Ed that he has an 'unfair' coin?
2. What is the probability in my head that Mr.Ed's coin is unfair?
This suggestion
I want you to think about the answers in terms of relative values not absolute values.
This statement
Uhh how about the two agents' different amounts of information?
The whole "thought experiment" is ridiculous.
Nothing but nonsense.
All attempts to defend it and, later, trying to get out of this emberasment are lame and ridiculous.
Good night.Last edited by hutennis; 08-16-12, 11:48 PM.Comment -
MonkeyF0ckerSBR Posting Legend
- 06-12-07
- 12144
#45It's called a hypothetical. You have failed the thought experiment as thinking is not something you seem to excel in.Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#46
If coming up with nonsensical hypothetical is the best that one can do, it is certainly the best not to come with any at all.Comment -
HUYSBR Sharp
- 04-29-09
- 253
#47Suppose I have 100 quarters in a jar and I wonder if they are unfair in the sense that they have an inherent tendency to land on heads or tails significantly more than 50% of the time. I know the coins are all identical. I hire 100 people to go home with one coin each, and ask them to flip their coin 200 times and record the results. I tell them to come back to me only if they think they have an unfair coin. Each of these 100 people does not know I have hired 99 others to do the same.
Suppose one of the 100 people I hired, let's call him Mr.Ed, comes back to me and tells me his coin landed 130 heads and 70 tails.
1. What is the probability in the head of Mr.Ed that he has an 'unfair' coin?
2. What is the probability in my head that Mr.Ed's coin is unfair?Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#49Now, this is border line idiotic.
Selection bias is the bias that occurs in a survey or experimental data when the selection of data points isn't sufficiently random to draw a general conclusion.
First. 100 coins in a jar presumed to be fair is as random as it gets.
Second. Report, 1 out of 100, given by MrED is as random as it gets.
Third. You didn't even ask to draw any conclusion about fairness of the remaining 99 coins based on MrEd's results. You ask about possible differences in assessment of probability that MrEd's coin is fair.
Not only this is an absurd notion in and off itself, but this question in general has nothing to do with selection bias at all.
You simply don't know what you are talking about.
Here is a perfect example of a selection bias.
If you are conducting a survey on a subject of validity of past looking models for predicting future results in SB, but the list of participants includes only HTT posters (very non random selection for obvious reason. 90% of participants WANT to believe that analyzing past data can predict future results).
The results of such a survey will be heavily, heavily biased and therefore completely useless as far as drawing general conclusions.
See the difference?Comment -
wrongturnSBR MVP
- 06-06-06
- 2228
#50Yeah, so if somebody wins a $100 million lottery, you must thought he cheated as there is only less than %0.000001 chance which in your mind should never happen.Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#51
Smart, intelligent, bright...
In fact this is just another very shortsighted, poorly thought off and ignorant comment.
In a country with a population of 300 millions there are enough lottery players so one winner is EXPECTED.
If only 100 people would have bought a ticket each and one of them would hit a jack pot, I can assure you that FBI would be all over him like a cheap soup.Comment -
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#52Now, this is border line idiotic.
Selection bias is the bias that occurs in a survey or experimental data when the selection of data points isn't sufficiently random to draw a general conclusion.
First. 100 coins in a jar presumed to be fair is as random as it gets.
Second. Report, 1 out of 100, given by MrED is as random as it gets.
Third. You didn't even ask to draw any conclusion about fairness of the remaining 99 coins based on MrEd's results. You ask about possible differences in assessment of probability that MrEd's coin is fair.
Not only this is an absurd notion in and off itself, but this question in general has nothing to do with selection bias at all.
You simply don't know what you are talking about.
Here is a perfect example of a selection bias.
If you are conducting a survey on a subject of validity of past looking models for predicting future results in SB, but the list of participants includes only HTT posters (very non random selection for obvious reason. 90% of participants WANT to believe that analyzing past data can predict future results).
The results of such a survey will be heavily, heavily biased and therefore completely useless as far as drawing general conclusions.
See the difference?Comment -
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#54In this case, the probability that his coin is fair is 0.028, regardless of how many millions of other coins are being flipped for you around the world and how many millions of other results you expect to receive.
Whether you think about results obtained for you by MrED in relative or absolute terms does not make any difference either. Probability of his coin to be fair is still the same - 0.028.
This has been pointed out to you twice now.Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#55Cut your loses man.
Always a good ideaComment -
wrongturnSBR MVP
- 06-06-06
- 2228
#56The funny thing is that you must feel so good about yourself after making this argument.
Smart, intelligent, bright...
In fact this is just another very shortsighted, poorly thought off and ignorant comment.
In a country with a population of 300 millions there are enough lottery players so one winner is EXPECTED.
If only 100 people would have bought a ticket each and one of them would hit a jack pot, I can assure you that FBI would be all over him like a cheap soup.
If someone winning a lottery is expected, then Mr. Ed's result can also be expected, and Mathy assumes the result is valid so he can present his question. To argue Mr. Ed must have done something wrong is like barking the wrong tree.Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#57What is more funnier, is that everybody else think that is exactly how you feel about your own argument, and come to the same conclusion. Haha.
If someone winning a lottery is expected, then Mr. Ed's result can also be expected, and Mathy assumes the result is valid so he can present his question. To argue Mr. Ed must have done something wrong is like barking the wrong tree.
No wander books make so much.
They get it too god damn easy too.Comment -
evo34SBR MVP
- 11-09-08
- 1032
#58One of the better Chinglish expressions I have ever seen. Almost makes it worth keeping this d-bag around. Almost.Comment -
wrongturnSBR MVP
- 06-06-06
- 2228
Comment -
InspiritedSBR MVP
- 06-26-10
- 1788
#60This is kind of like hut's argument that he repeats over and over again in just about every thread. You think you have a significant system/model that beats the books, but it is likely that you're really just one of the lucky ones.
It seems like the way mathy put it, this needs to be expressed in a bayesian form. Given the knowledge that mathy has, his posterior leads him to believe the coin are fair, but given the knowledge that Mr. Ed has, Mr. Ed believes the coin likely to be unfair.
???Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#61This is kind of like hut's argument that he repeats over and over again in just about every thread. You think you have a significant system/model that beats the books, but it is likely that you're really just one of the lucky ones.
It seems like the way mathy put it, this needs to be expressed in a bayesian form. Given the knowledge that mathy has, his posterior leads him to believe the coin are fair, but given the knowledge that Mr. Ed has, Mr. Ed believes the coin likely to be unfair.
???
You believe that your model can beat their model.
OK. What your beliefs are based on?
If you could somehow get your hands on their model, or reverse engineer it, break it down to its components then I understand.
You can say to yourself :"Aha, they are missing this!!!! I have everything they have and more. They have everything from A to Y, but I have that AND more. I have Z. And that Z gives me an edge. And that why I expect to beat them."
But you don't know that, don't you.
All you know are elements you consider.
What if they have all you have and then some?
After all, they are sophisticated enough, are not they?
They have everything they need to stay on a cutting edge, do they?
So. What your beliefs that you may have an edge are based on?
Your results over last 500 games?
Please, explain.Comment -
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#62Comment -
TMoney33SBR Sharp
- 08-29-10
- 388
#63My math skills are weak, but I think the intended observation is that Mr. Ed will believe there is a high probability of the coin being biased based on his limited number (200) of flips. On the other hand, MDC realizes that over 100 sets of 200 flips (20000) flips, a series of 130/200 is bound to occur and has little to no bearing by itself on the odds of the coin being true.Comment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#64It is really hard to understand.
I asked a question.
Very legitimate and logical question.
I expect that there should be no problem getting an intelligent answer to my question in a community full of successful handicappers.
Instead a picture of a person who would not be able to answer my question (must be a self-portrait) is posted and is awarded (must be by someone who would not be able to answer my question either but none the less gets frustrated simply by reading it).
Is anyone able to take a shot on answering?
I mean no disrespect. I just think it is a good question to ask and think about.
Some body must have reverse engineered Midas around here or at least knows how to beat it without knowing what the hell they are trying to beat in a first place.Comment -
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#65You are incomprehensibleComment -
hutennisSBR Wise Guy
- 07-11-10
- 847
#66
In order to outperform your competition first you have to know exactly how your competition comes up with product you want to compete with. Only then you can apply something extra that gives you an edge.
If you want to outperform Ferrari's engine first you have to know how exactly they do what they do, be able to do exactly the same thing
and then figure out what Ferrari missed that you can add and produce something better.
For your model to have an edge over "Vegas" model you have to do everything that "Vegas" does and then some.
Otherwise there can be no edge.
If you don't know what exactly goes to their simulation engines what are the basis for your belief that your model can have an edge
over their model? How can you compere if you have no idea what you compere it to?
Those are my questions.
What is so incomprehensible about them?Comment -
AnalysisSBR Rookie
- 08-18-12
- 3
#67
In order to outperform Ferrari's enginge, all you have to know is how well it performs and (!) how to build an engine that performs better.
If your model, over time, performs better than the bookies - yours is better. What's in yours, or what's in theirs doesn't mather.Comment -
mathdotcomSBR Posting Legend
- 03-24-08
- 11689
#68In order to outperform your competition first you have to know exactly how your competition comes up with product you want to compete with. Only then you can apply something extra that gives you an edge.
If you want to outperform Ferrari's engine first you have to know how exactly they do what they do, be able to do exactly the same thing
and then figure out what Ferrari missed that you can add and produce something better.
For your model to have an edge over "Vegas" model you have to do everything that "Vegas" does and then some.
Otherwise there can be no edge.
If you don't know what exactly goes to their simulation engines what are the basis for your belief that your model can have an edge
over their model? How can you compere if you have no idea what you compere it to?
Those are my questions.
What is so incomprehensible about them?
Here is the short counterexample to your claim (in my best attempt at hu-english): Vegas model is opener lines, like example RedSox/Yankees -105/-115... but almost always the line is move and so closer lines different from open line and who yu tink move dis line? not vegas.... must be someone... with better FerrariComment -
AnalysisSBR Rookie
- 08-18-12
- 3
#69This is the same old topic you keep bringing up in every single thread.
Here is the short counterexample to your claim (in my best attempt at hu-english): Vegas model is opener lines, like example RedSox/Yankees -105/-115... but almost always the line is move and so closer lines different from open line and who yu tink move dis line? not vegas.... must be someone... with better Ferrari
In order to outperform your competition first you have to know exactly how your competition comes up with product you want to compete with. Only then you can apply something extra that gives you an edge.
If you want to outperform Ferrari's engine first you have to know how exactly they do what they do, be able to do exactly the same thing
and then figure out what Ferrari missed that you can add and produce something better.
For your model to have an edge over "Vegas" model you have to do everything that "Vegas" does and then some.
Otherwise there can be no edge.
If you don't know what exactly goes to their simulation engines what are the basis for your belief that your model can have an edge
over their model? How can you compere if you have no idea what you compere it to?
Those are my questions.
What is so incomprehensible about them?
Why do you need to have the slightest idea about how the Ferrari engine is built to be able to build one that outperforms it?
If your model is more accurate than the bookies (over time of course), then yours are better. There's absolutely no need to know anything about what's in the bookies model (which btw isn't only maths)Last edited by Analysis; 08-18-12, 11:16 AM.Comment -
wrongturnSBR MVP
- 06-06-06
- 2228
#70In order to outperform your competition first you have to know exactly how your competition comes up with product you want to compete with. Only then you can apply something extra that gives you an edge.
If you want to outperform Ferrari's engine first you have to know how exactly they do what they do, be able to do exactly the same thing
and then figure out what Ferrari missed that you can add and produce something better.
For your model to have an edge over "Vegas" model you have to do everything that "Vegas" does and then some.
Otherwise there can be no edge.
If you don't know what exactly goes to their simulation engines what are the basis for your belief that your model can have an edge
over their model? How can you compere if you have no idea what you compere it to?
Those are my questions.
What is so incomprehensible about them?Comment
SBR Contests
Collapse
Top-Rated US Sportsbooks
Collapse
#1 BetMGM
4.8/5 BetMGM Bonus Code
#2 FanDuel
4.8/5 FanDuel Promo Code
#3 Caesars
4.8/5 Caesars Promo Code
#4 DraftKings
4.7/5 DraftKings Promo Code
#5 Fanatics
#6 bet365
4.7/5 bet365 Bonus Code
#7 Hard Rock
4.1/5 Hard Rock Bet Promo Code
#8 BetRivers
4.1/5 BetRivers Bonus Code