Statistics 101 with BigPapa Aka BigData

Status
Not open for further replies.

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
Derailing: this post has been rated a "Derailing post" by forum members
Admin Edit: This is the continuation of the derail of this thread.


Not in that seduction no. Perhaps he has others that would apply.

Though, I don't see how that affects what I said previously, while speaking from my experience. Or what Chase said to you here:


Nope. I specifically included. . . drop-dead gorgeous. This is a girl whose beauty overwhelms subjective tastes for the most part.

It is a stretch for anyone to say every girl who's their type is also drop-dead gorgeous. It is also a stretch to claim all girls who are your type will be difficult to shag. You know what else is a stretch? Assuming you'll meet large numbers of girls, who are drop-dead gorgeous and your exact type on any given night. Thus a pragmatic step towards victory is equipping yourself to get past a tough hook. So that you can get past it, should you need to.



Cool, that leaves 1 who wasn't as easy as the other two. I suggest keeping a watchful eye on this growing tally with each passing year.

If you feel like you've seriously leveled up your game after a special year. Check the tally. Examine the new top 10. There will likely be more variance between your hottest lays who were easy, the one's who were not easy. . . and the ones where it felt like bedding them required moving mountains.

I am a data guy , and just could not help it.

Let's say that you live in a city that has 2 million people that are evenly split between men and women. This means that you have 1 million women living in the city. Let's say that the drop dead gorgeous women represent 1 % of the female population , this means that you have 10,000 drop dead gorgeous women living in that city .

Now let's start doing some statistic inference based on the population size of 10,000 drop dead gorgeous girls . In order for you to have a significant sample size ( basically girls that you sleep with ) and then to categorize them with a quality attribute based on what they were easily approachable or not ( easy , medium , hard) , the sample size needs to be between 370 ( taking into consideration that the confidence interval is 95 % and marginal error is 5 % ) and 622 ( taking into consideration that the confidence interval is 99 % and the marginal error is 5 % ) .

After you have the sample size in question then you can extrapolate the findings to the initial population , and say that drop dead gorgeous women indeed make it easier or not for you.

I am not saying this to be a dickhead , I am just saying that the opinions are quite divided , and that I noticed a pattern with the guys that are super advanced, and that is that they do not waste too much time with red girls . Maybe they do this because they do a lot of approaches ( and ofc you get more girls that are easily approachable) , maybe it is an ego thing , maybe they already reached the required sample size and noticed that drop dead gorgeous women do not necessary make things harder for you when you approach them . Have no idea , I am quite far away from banging even 370 drop dead gorgeous woman so I can not really have an opinion about this matter, I can only speculate.

Me personally I do not really care if she is easily approachable or not , I like the hunt more than the sex itself :)

What I can say is though, since there are lots of debates on this matter, and also see how super advanced guys see things, gaming girls that are in the red spectrum quantum ( btw great article ) does not necessary means that you have a better game than someone that goes for yellow & green girls in the quantum .
 
Last edited by a moderator:

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
I am a data guy , and just could not help it.

Let's say that you live in a city that has 2 million people that are evenly split between men and women. This means that you have 1 million women living in the city. Let's say that the drop dead gorgeous women represent 1 % of the female population , this means that you have 10,000 drop dead gorgeous women living in that city .

Now let's start doing some statistic inference based on the population size of 10,000 drop dead gorgeous girls . In order for you to have a significant sample size ( basically girls that you sleep with ) and then to categorize them with a quality attribute based on what they were easily approachable or not ( easy , medium , hard) , the sample size needs to be between 370 ( taking into consideration that the confidence interval is 95 % and marginal error is 5 % ) and 622 ( taking into consideration that the confidence interval is 99 % and the marginal error is 5 % ) .

After you have the sample size in question then you can extrapolate the findings to the initial population , and say that drop dead gorgeous women indeed make it easier or not for you.

We have a PUA community so that we can combine individual sample sizes to much larger sample sizes, you seem to forget that.
Also your argument cannot be correct as it depends on city size. What if you live in a 20x smaller town, do you then require 20x less statistics? Makes no sense.
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
We have a PUA community so that we can combine individual sample sizes to much larger sample sizes, you seem to forget that.

I do not forget that , just that people are quite divided in how they see things . Some say that going for reds is time wasting , others do not . What I can say in this regards is that the Pareto Principle says that roughly 80% of the effects come from 20% of the causes ( this is a law as true as the ones about gravity) . Like I mentioned I am not really the guy to say that those 20 % causes have a higher density with reds , yellow or greens ( using the traffic light system lingo ).

Also your argument cannot be correct as it depends on city size. What if you live in a 20x smaller town, do you then require 20x less statistics? Makes no sense.

When comparing other big cities it is quite relevant , as they are very heterogenous places and can easily take a random sample size and compare it to other random sample sizes and see almost no differences.

When comparing a 20x smaller city you can extrapolate it to the target population in that particular town , as it is not as heterogeneous when compared to a big city . In the same time it will be enough to have a 20x smaller sample size for that particular town to be significant ( and can easily be extrapolated to other cities that have more or less the same characteristics with the town where you initially did the sample ) .

But more or less this is in line with what everyone agrees on , meaning that the game in a small town is different than the game in a big one.
 

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
I do not forget that , just that people are quite divided in how they see things . Some say that going for reds is time wasting , others do not . What I can say in this regards is that the Pareto Principle says that roughly 80% of the effects come from 20% of the causes ( this is a law as true as the ones about gravity) . Like I mentioned I am not really the guy to say that those 20 % causes have a higher density with reds , yellow or greens ( using the traffic light system lingo ).

Again your math is off. The Pareto principle 80/20 is NOT an universal law, it is a specific case of a general family of distributions. You can just as well have cases of 70/30, 90/10, 5/95 etc. Also you can't just postulate that something has a Pareto distribution just as you can't postulate a normal distribution without some assumptions.


When comparing other big cities it is quite relevant , as they are very heterogenous places and can easily take a random sample size and compare it to other random sample sizes and see almost no differences.

When comparing a 20x smaller city you can extrapolate it to the target population in that particular town , as it is not as heterogeneous when compared to a big city . In the same time it will be enough to have a 20x smaller sample size for that particular town to be significant ( and can easily be extrapolated to other cities that have more or less the same characteristics with the town where you initially did the sample ) .

But what is so special about 2 million that you use as a base for your argument?

We have guys here living in NYC with a population of 8 million. Do they require 4x the experience than the guy living in a 2M city?

You are just using an arbitrary input here to "prove" a general principle.
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
Again your math is off. The Pareto principle 80/20 is NOT an universal law, it is a specific case of a general family of distributions. You can just as well have cases of 70/30, 90/10, 5/95 etc. Also you can't just postulate that something has a Pareto distribution just as you can't postulate a normal distribution without some assumptions.

That is pure academics , the real life follows the pareto principle when it comes to effort vs reward :)

As an example , I saw a lot of data sets of companies and every data set was following the pareto principle . Same with interviews , same with girls , etc .

But what is so special about 2 million that you use as a base for your argument?

We have guys here living in NYC with a population of 8 million. Do they require 4x the experience than the guy living in a 2M city?

You are just using an arbitrary input here to "prove" a general principle.

It was just an example, in theory yes you would need 4x the sample size , but in reality I am pretty sure that in this particular case you will not see big changes .

Let me give you an example , beauty for women is more a state of spirit rather than arbitrary concept ( as it is for men ). This means that in cities where you have a higher density of beautiful women they will not feel as pretty as they would feel in a smaller city where the density is not as high. Basically it means that in Moscow even though she is a true 10 in the eyes of every guy in this world , she would feel most of the times as 8 or a 9 ( depending on where she is ) . Sadly also the reverse is true , where you have a girl that is a 7 from Moscow that goes abroad where the density of attractive women is not as high as it is in Moscow and she will feel more like 9 , even though in reality she is a 7 ( looking with the glasses of an arbitrary way of defining beauty from a man perspective ) .

I do not really see why you are so aggressive in your communication though .
 

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
That is pure academics , the real life follows the pareto principle when it comes to effort vs reward :)

Pure academics which you started discussing.

As an example , I saw a lot of data sets of companies and every data set was following the pareto principle . Same with interviews , same with girls , etc .

Sorry but this is nonsense. You only need entry-level courses in university statistics to know better, or possibly even high school math.

It was just an example, in theory yes you would need 4x the sample size , but in reality I am pretty sure that in this particular case you will not see big changes .

Maybe I also should point out to you that statistical uncertainty generally scales as the inverse SQUARE ROOT of your number of observations?
 

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
People should also take note that BigPapa claims this

Nope , I have a really high esteem as I am naturally a quite handsome guy , have a really good career , good work life balance , just got into Harvard
to do a masters degree in Data Science
( it is the extension school so not quite the real Harvard , but hey I still will get a degree from Harvard haha )

Yet he does not know stuff an average first-year university student in math here in Norway knows.

One may just wonder what else he is exaggerating his abilities on...

 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
Pure academics which you started discussing.

I was just saying that looking only at your top 10 is not enough.

Sorry but this is nonsense. You only need entry-level courses in university statistics to know better, or possibly even high school math.

Maybe I also should point out to you that statistical uncertainty generally scales as the inverse SQUARE ROOT of your number of observations?

Depends on what context exactly you are referring , but as a rule the R squared is the coefficient of determination , and is calculated as explained variations / total variations . Basically you will get a percentage that will say how much the variable explains your data set .

It is also true that the more data you have , the more outliners you have which basically reduces the "importance" of the variable ( not by much though ) . In this case it is very important to look at the outliners as usually they have important things that they are pointing out ( like in the example of 10s that do not feel like looking like 10 I gave above ) .
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
People should also take note that BigPapa claims this



Yet he does not know stuff an average first-year university student in math here in Norway knows.

One may just wonder what else he is exaggerating his abilities on...


Well now you are crossing the line senor and arguing in bad faith :)
 

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
Depends on what context exactly you are referring , but as a rule the R squared is the coefficient of determination , and is calculated as explained variations / total variations . Basically you will get a percentage that will say how much the variable explains your data set .

It is also true that the more data you have , the more outliners you have which basically reduces the "importance" of the variable ( not by much though ) . In this case it is very important to look at the outliners as usually they have important things that they are pointing out ( like in the example of 10s that do not feel like looking like 10 I gave above ) .

Thanks for teaching me a new term, OUTLINERS. I have only heard of OUTLIERS. But then again I never went to Harvard.
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
Thanks for teaching me a new term, OUTLINERS. I have only heard of OUTLIERS. But then again I never went to Harvard.

that is called a typo haha , dunno what you try to prove , but it feels like arguing with my girlfriend . No homo and no trolling ;)
 

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
that is called a typo haha , dunno what you try to prove , but it feels like arguing with my girlfriend . No homo and no trolling ;)

Listen here, I do actually work in data science and I have actual STEM degrees. You just revealed yourself to be full of shit.
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
Listen here, I do actually work in data science and I have actual STEM degrees.
Same here amigo ( but if I remember correctly you mentioned multiple times that you are a programmer - maybe I misunderstood ) , just that I do not have the STEM degree yet ( just one in Economics - which btw is full of math ) , but will have one soon from Harvard :)

Again , you argue in bad faith , and picking on the slightest thing trying to prove a point, like typos . Even to the best of us autocorrect kicks in ;)
 
Last edited:

Carousel

Tribal Elder
Tribal Elder
Joined
Nov 11, 2019
Messages
576
And now you guys can go back to discuss the merits or pitfalls of traffic light game.
 

Rakkum

Cro-Magnon Man
Cro-Magnon Man
Joined
Apr 2, 2017
Messages
198
Well now you are crossing the line senor and arguing in bad faith :)

No offense but I think @Carousel has a point here. Maybe you are all the things you say but you kind of sound hollow, a bit off.


Here's what I mean, a thread of yours:

I found a really interesting data set about the census in USA that can be generalized to the rest of the world , as the data for them more or less show the same trend .

I'm re-learning high-school math, have no math nor STEM credentials to show but what I can tell you is that you don't need to generalize this for the rest of the world. It's a universal fact that everyone with a secondary education knows already.

You see what I mean?
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
I'm re-learning high-school math, have no math nor STEM credentials to show but what I can tell you is that you don't need to generalize this for the rest of the world. It's a universal fact that everyone with a secondary education knows already.
You see what I mean?

I do not understand exactly your point . I said that I was playing with some census data from the USA ( school project meh ) and found that interesting fact that there are more boys born than girls ( contrary to the popular belief that boys are more "difficult" to have ).

I found that interesting and compared this USA census data to those of other countries and found the same trend .

If we look at countries like China ( where you are allowed to have only 1 child - recently I read that you are allowed to have 2 ) or some more extremist arab countries let's say , yes you will have a lot of abortions due to the "child gender" , because families prefer having boys to having girls ( usually it is because of a mix of culture + economics ) .
But if the statistics were not "rigged" because of this "cultural beliefs" let's say, also those countries would see the same trend .

My bad when I did not explain that in some countries this will not be the case , due to this " cultural" beliefs , but I am quite sure I did not add this because I do not really care what is happening there due to practical reasons ( like never thinking visiting those places in the first place ) .

So ... coming back to your post , what you wanted to say actually or to prove ? :)
 

Mike Silvertree

Chieftan
Staff member
tribal-elder
Joined
Nov 12, 2019
Messages
296
Location
A long time ago
Can any of you tell me; how many Angels can dance on the head of a pin?

This has taken a laughably dumb tangent.

The advice here is given free, and you are free to take it or not. We have no objective proof, just testimonials, which are, of course, anecdotal.

The actual point of the original post was to tell people to knock off the thread derailing and dick waving contests. That admonition is still in effect.
 

BigPapa

Space Monkey
space monkey
Joined
Jan 12, 2020
Messages
601
Except you dont need to analyze any data to find this:


Literally took the actual data scientist here 5 seconds to google.

oh you 2 guys , are really cute :)

there is a big difference between gender birth rate and gender distribution rate . I will give you 2 guys a hint : women outlive men by a couple of years on average and the population is getting older on average . You can do the math from here . If you can not , I will be more than happy to help you guys further . But again I also wrote this in my post that rakum was really nice to share it .

Rakum really cool that you are learning high school math . For sure it is a good investment ;)

Carousel from your side I am little bit surprised hearing this kind of things with your STEM background , PhD , working as a data scientist , and other impressing stuff .

To be frank I do not understand what is exactly your problem , but as I mentioned in my previous posts it is not an attitude that I am that fond of and totally unelderly ,and if you try to roast me then let”s agree before that you are not gonna use your forum rank in this roast war and that the other mods will not take your side .

Otherwise , let”s just toast a glass of wine and talk about interesting stuff that will benefit everyone on this forum , like guys who share the same mission would do :)
 

Teevster

Tribal Elder
Tribal Elder
Joined
Aug 23, 2013
Messages
1,714
Location
Scandinavia - For Now
there is a big difference between gender birth rate and gender distribution rate . I will give you 2 guys a hint : women outlive men by a couple of years on average and the population is getting older on average . You can do the math from here . If you can not , I will be more than happy to help you guys further . But again I also wrote this in my post that rakum was really nice to share it .

Aren't you just adding another variable to your model here?

Any model can be tweaked by subtracting and adding new variables to a model.

Unless we do so, i.e. limit about model we are at risk of ending up with kitchen sink regressions.

But again, I must admit I have stepped away from quantitative social sciences long ago. I could be wrong.

Best,
 
Status
Not open for further replies.
Top
>