These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Crime & Punishment

 
  • Topic is locked indefinitely.
 

Bring me the head of a CFC titan pilot!

First post First post
Author
Previous Topic Next Topic
Voyager Arran
Brutor Tribe
Minmatar Republic
#561 - 2014-06-18 05:10:32 UTC
Leto Thule wrote:
Gevlon Goblin wrote:
Khanh'rhh wrote:
I have done some advanced maths to show how Gevlon is going broke.
Gevlon's Wallet - grrr spending = Gevlon's smaller wallet

I have to ruin your dreams. I never spent any part of my capital. I spent my weekly income every week. I have exactly the same amount as assets as I had before the project. I merely didn't accumulate any more, as any profit went (and keeps going until Goonies surrender or destroyed) into the GRR project.



These are the tactics of the lie.


No, I doubt Gevlon would ever care enough about anything to voluntarily subject himself to real financial loss.

I guess we'll just have to content ourselves with him furiously flushing his money down the toilet as fast as he can grind it up, month after month after month.
Catalytic morphisis
Ruthless Regiment
SL0W CHILDREN AT PLAY
#562 - 2014-06-18 10:16:45 UTC
Radric Davids wrote:
Gevlon Goblin wrote:
Khanh'rhh wrote:
Anyhow - I am going to help you out slightly, basic stats stuff.

To support the notion that a particular correlation is due to a causal relationship, one needs to propose a mechanism for the relationship.
So, by what mechanism do you think getting 200% reimbursement for our losses makes us quit the game? Is it because we get apathy from having too much money? Do our wallets get so large the client doesn't load?

Gonna need you to link A to B on this one and show how gaining money makes us leave.

You get 200% reimbursement for losing a JF to Marmite or a Golem to Mordus? Really? Because then I really have to start some clean alts and SA accounts?

By the way on Thursday I'll write something really nice about GSF. Honestly!


do you realize how horribly you misinterpreted the correlation coefficient in your blog? hilarious.

The correlation coefficient is a measure of the degree to which the movement of two variables are associated, not the degree to which the dependent variable is explained by the independent variable.

You are confusion correlation coefficient, which is [cov(x,y)]/ SxSy

and a linear regression coefficient, which is a slope coefficient on the independent variable (period 1 kills-losses) on the dependent variable (period 2 kills-losses). The slope coefficient would indeed explain how much a net loss in period 1 influences period 2 net losses. It is calculated using the Ordinary Least Squares method, which results in a linear equation of:

Y = a +b*x + e

Y being period 2 kills-losses, X being period 1 kills-losses, b being the regression coefficient you wanted (% of period 2 kills-losses explained by period 1 kills-losses)

Quote:
It makes sense, those who were above average active in the first half, are likely remain above average. 0.55 is not a bad correlation, but far from perfect either. There must be another factor.


Yes, you are missing many, many factors including real life events, wars, in-game events, economic effects, recruitment/alliance switching, deployments, doctrines, logi/scouting/dictor use, and many others.

Your model, and mistaken as it is, is heavily influenced by exogenous factors. Your correlation coefficients don't prove anything, in fact anyone with a brain could have told you that you would have found a somewhat significant result. active players are likely to have been active 6 months ago. However, you cannot calculate the effect of losing a 1bil freighter based on your 'calculations'

Your regression would be incredibly oversimplified, had you correctly performed it. 6 months of pvp activity is influenced by many more factors than a single 1 billion isk freighter loss in a prior 6 month period. Any explanatory power you might have found if you had done this properly, would be a result of the fact that you have excluded countless exogenous variables. I believe if you properly conducted a linear regression, your F statistic and adjusted R^2 would reflect that.


Quote:
Since the variance of the total activity is larger than of the net ISK (5612M and 4729M), if you gank an 1B ship of a GSF member, he will have 1B*5.61/4.73*0.42 = 0.5B less activity (kills+losses) in the next half year on average than he'd have without the gank.


You used the fact that variance is greater than the sample mean (which actually just means that the data is highly varied) to explain that you could make the following calculation:

ganked ship * variance / mean * correlation coefficient = effect on next period kill-loss

What equation is that exactly? I actually lol'd at that. You calculated a correlation coefficient, not a linear regression. Regardless, what the hell is that equation? I'm still giggling. You need the slope coefficient to predict the effect of a current period loss in the next period. You did not perform a regression. The correlation coefficient does not work like this. This equation is not a thing


The higher the correlation coefficient that excel calculated for you (dont blindly use excel formulas unless you understand them to avoid embarrassment like this), the closer the two are to a perfect linear relationship. Logically, that sounds reasonable for your chosen variables.

What you tried to interpret that as, is the regression (or slope) coefficient. The slope coefficient is the extent to which the change in the expected value of period 2 kills-losses is explained by period 1 kills-losses. That is what you were going for, but they are very different in both calculation and interpretation.

Your calculation of the correlation coefficient explains precisely nothing. It has no weight over what you were trying to prove. Try again.

PS: de-whoring kills is ********. Basically only shows small gang and solo kills as a result. de-whored kills - losses will be VASTLY different than just kills-losses

TLRD: A correlation coefficient is not a regression coefficient. Try again with a linear regression and post the F test, T statistics, R^2, durbin watson statistic. You cannot predict the effect of a period 1 freighter gank on period 2 kills-losses using a correlation coefficient. Using excel formulas without understanding the statistic provides great laughs for everyone. Try googling regression vs correlation coefficient next time

**** son! Gevlon got owned! This made me chuckle, Blag math vs Real math :) For taking the time to write this out you are now my hero!

Actual Link free and scout free solo PvP'er
Khanh'rhh
Sparkle Motion.
#563 - 2014-06-18 11:11:37 UTC
Gevlon Goblin wrote:
Khanh'rhh wrote:
I have done some advanced maths to show how Gevlon is going broke.
Gevlon's Wallet - grrr spending = Gevlon's smaller wallet

I have to ruin your dreams. I never spent any part of my capital. I spent my weekly income every week. I have exactly the same amount as assets as I had before the project. I merely didn't accumulate any more, as any profit went (and keeps going until Goonies surrender or destroyed) into the GRR project.

There's 2 problems with this

1) Throwing away 100% of your income (when we know you poop-sock this game for hilarious amounts of time per day / week) is worse than just letting an existing pile of money whittle down.
You're basically sitting at your computer, for hours a day / week to make your pretend money, and then you're taking all that time, all that effort, and just throwing it at people who aren't making a difference. Who laugh at you even when you pay them.
You then spent countless other hours, writing blog posts to try to convince people (or perhaps just yourself) that you're having any impact at all, which literally no one, CFC affiliated or not, believes.

2) We know it's not true, since your weekly spend is going down. Either you're getting worse at the game, or you've realized you're hurtling towards 0 balance a lot faster than you thought you would* and you're trying to balance the books.

* - I will point out that your original "grrr goon POCOs" plan was to pay some cheap wardecs so that people could RF our POCOs without paying, for small gang (~5 man) PVP. Because you don't understand the mechanics involved, you didn't realize that to stop us taking money from them, you would need to take them from us. Basically, you never budgeted for months of paying 10bil a week, and you know it - you've just doubled-down several times instead of admitting you can't harm us in any way, and have shifted your goalposts from (major changes only):

1 -
Goal: "to Rf goon POCOs to stop money being made"
Result: Fail. Misunderstood game mechanics
2 -
Goal: "to take all goon POCOs in highsec"
Result: Fail. We sold them
Bonus result: Fail. RvB are all goons and still own them
3 -
Goal: "to collapse the CFC by highsec wars"
Result: Fail.
4 -
Goal: "to collapse the CFC by giving small amounts of money to NPC residents"
Result: Fail, no change in their activity
5 -
Goal: "to make CFC titan pilots lose their ships / be disinclined to use them
Result: Fail. None dead, still using them to RF POCOs and kill pods for lols.
6 -
Goal: "to say bombless bombers more than previously thought humanly possible"
Result: SUCCESS. Despite most of fountain, delve, and the south east being ground out by supers and dreads, Gevlon has never stopped saying bombless bomber.

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Arkady Romanov
Whole Squid
#564 - 2014-06-18 11:18:40 UTC
Goddamnit Khanh'rhh stop dunking him so hard. Leave something for the rest of us.

Whole Squid: Get Inked.
Catalytic morphisis
Ruthless Regiment
SL0W CHILDREN AT PLAY
#565 - 2014-06-18 11:35:35 UTC
Khanh'rhh wrote:
Gevlon Goblin wrote:
Khanh'rhh wrote:
I have done some advanced maths to show how Gevlon is going broke.
Gevlon's Wallet - grrr spending = Gevlon's smaller wallet

I have to ruin your dreams. I never spent any part of my capital. I spent my weekly income every week. I have exactly the same amount as assets as I had before the project. I merely didn't accumulate any more, as any profit went (and keeps going until Goonies surrender or destroyed) into the GRR project.

There's 2 problems with this

1) Throwing away 100% of your income (when we know you poop-sock this game for hilarious amounts of time per day / week) is worse than just letting an existing pile of money whittle down.
You're basically sitting at your computer, for hours a day / week to make your pretend money, and then you're taking all that time, all that effort, and just throwing it at people who aren't making a difference. Who laugh at you even when you pay them.
You then spent countless other hours, writing blog posts to try to convince people (or perhaps just yourself) that you're having any impact at all, which literally no one, CFC affiliated or not, believes.

2) We know it's not true, since your weekly spend is going down. Either you're getting worse at the game, or you've realized you're hurtling towards 0 balance a lot faster than you thought you would* and you're trying to balance the books.

* - I will point out that your original "grrr goon POCOs" plan was to pay some cheap wardecs so that people could RF our POCOs without paying, for small gang (~5 man) PVP. Because you don't understand the mechanics involved, you didn't realize that to stop us taking money from them, you would need to take them from us. Basically, you never budgeted for months of paying 10bil a week, and you know it - you've just doubled-down several times instead of admitting you can't harm us in any way, and have shifted your goalposts from (major changes only):

1 -
Goal: "to Rf goon POCOs to stop money being made"
Result: Fail. Misunderstood game mechanics
2 -
Goal: "to take all goon POCOs in highsec"
Result: Fail. We sold them
Bonus result: Fail. RvB are all goons and still own them
3 -
Goal: "to collapse the CFC by highsec wars"
Result: Fail.
4 -
Goal: "to collapse the CFC by giving small amounts of money to NPC residents"
Result: Fail, no change in their activity
5 -
Goal: "to make CFC titan pilots lose their ships / be disinclined to use them
Result: Fail. None dead, still using them to RF POCOs and kill pods for lols.
6 -
Goal: "to say bombless bombers more than previously thought humanly possible"
Result: SUCCESS. Despite most of fountain, delve, and the south east being ground out by supers and dreads, Gevlon has never stopped saying bombless bomber.

I approve this message

Actual Link free and scout free solo PvP'er
Pete Butcher
The Scope
Gallente Federation
#566 - 2014-06-18 12:15:44 UTC
Khanh'rhh wrote:
Gevlon Goblin wrote:
Khanh'rhh wrote:
I have done some advanced maths to show how Gevlon is going broke.
Gevlon's Wallet - grrr spending = Gevlon's smaller wallet

I have to ruin your dreams. I never spent any part of my capital. I spent my weekly income every week. I have exactly the same amount as assets as I had before the project. I merely didn't accumulate any more, as any profit went (and keeps going until Goonies surrender or destroyed) into the GRR project.

There's 2 problems with this

1) Throwing away 100% of your income (when we know you poop-sock this game for hilarious amounts of time per day / week) is worse than just letting an existing pile of money whittle down.
You're basically sitting at your computer, for hours a day / week to make your pretend money, and then you're taking all that time, all that effort, and just throwing it at people who aren't making a difference. Who laugh at you even when you pay them.
You then spent countless other hours, writing blog posts to try to convince people (or perhaps just yourself) that you're having any impact at all, which literally no one, CFC affiliated or not, believes.

2) We know it's not true, since your weekly spend is going down. Either you're getting worse at the game, or you've realized you're hurtling towards 0 balance a lot faster than you thought you would* and you're trying to balance the books.

* - I will point out that your original "grrr goon POCOs" plan was to pay some cheap wardecs so that people could RF our POCOs without paying, for small gang (~5 man) PVP. Because you don't understand the mechanics involved, you didn't realize that to stop us taking money from them, you would need to take them from us. Basically, you never budgeted for months of paying 10bil a week, and you know it - you've just doubled-down several times instead of admitting you can't harm us in any way, and have shifted your goalposts from (major changes only):

1 -
Goal: "to Rf goon POCOs to stop money being made"
Result: Fail. Misunderstood game mechanics
2 -
Goal: "to take all goon POCOs in highsec"
Result: Fail. We sold them
Bonus result: Fail. RvB are all goons and still own them
3 -
Goal: "to collapse the CFC by highsec wars"
Result: Fail.
4 -
Goal: "to collapse the CFC by giving small amounts of money to NPC residents"
Result: Fail, no change in their activity
5 -
Goal: "to make CFC titan pilots lose their ships / be disinclined to use them
Result: Fail. None dead, still using them to RF POCOs and kill pods for lols.
6 -
Goal: "to say bombless bombers more than previously thought humanly possible"
Result: SUCCESS. Despite most of fountain, delve, and the south east being ground out by supers and dreads, Gevlon has never stopped saying bombless bomber.


Well said sir.

http://evernus.com - the ultimate multiplatform EVE trade tool + nullsec Alliance Market tool
Radric Davids
GoonWaffe
Goonswarm Federation
#567 - 2014-06-18 12:54:52 UTC  |  Edited by: Radric Davids
Radric Davids wrote:
Gevlon Goblin wrote:
Khanh'rhh wrote:
Anyhow - I am going to help you out slightly, basic stats stuff.

To support the notion that a particular correlation is due to a causal relationship, one needs to propose a mechanism for the relationship.
So, by what mechanism do you think getting 200% reimbursement for our losses makes us quit the game? Is it because we get apathy from having too much money? Do our wallets get so large the client doesn't load?

Gonna need you to link A to B on this one and show how gaining money makes us leave.

You get 200% reimbursement for losing a JF to Marmite or a Golem to Mordus? Really? Because then I really have to start some clean alts and SA accounts?

By the way on Thursday I'll write something really nice about GSF. Honestly!


do you realize how horribly you misinterpreted the correlation coefficient in your blog? hilarious.

The correlation coefficient is a measure of the degree to which the movement of two variables are associated, not the degree to which the dependent variable is explained by the independent variable.

You are confusion correlation coefficient, which is [cov(x,y)]/ SxSy

and a linear regression coefficient, which is a slope coefficient on the independent variable (period 1 kills-losses) on the dependent variable (period 2 kills-losses). The slope coefficient would indeed explain how much a net loss in period 1 influences period 2 net losses. It is calculated using the Ordinary Least Squares method, which results in a linear equation of:

Y = a +b*x + e

Y being period 2 kills-losses, X being period 1 kills-losses, b being the regression coefficient you wanted (% of period 2 kills-losses explained by period 1 kills-losses)

Quote:
It makes sense, those who were above average active in the first half, are likely remain above average. 0.55 is not a bad correlation, but far from perfect either. There must be another factor.


Yes, you are missing many, many factors including real life events, wars, in-game events, economic effects, recruitment/alliance switching, deployments, doctrines, logi/scouting/dictor use, and many others.

Your model, and mistaken as it is, is heavily influenced by exogenous factors. Your correlation coefficients don't prove anything, in fact anyone with a brain could have told you that you would have found a somewhat significant result. active players are likely to have been active 6 months ago. However, you cannot calculate the effect of losing a 1bil freighter based on your 'calculations'

Your regression would be incredibly oversimplified, had you correctly performed it. 6 months of pvp activity is influenced by many more factors than a single 1 billion isk freighter loss in a prior 6 month period. Any explanatory power you might have found if you had done this properly, would be a result of the fact that you have excluded countless exogenous variables. I believe if you properly conducted a linear regression, your F statistic and adjusted R^2 would reflect that.


Quote:
Since the variance of the total activity is larger than of the net ISK (5612M and 4729M), if you gank an 1B ship of a GSF member, he will have 1B*5.61/4.73*0.42 = 0.5B less activity (kills+losses) in the next half year on average than he'd have without the gank.


You used the fact that variance is greater than the sample mean (which actually just means that the data is highly varied) to explain that you could make the following calculation:

ganked ship * variance / mean * correlation coefficient = effect on next period kill-loss

What equation is that exactly? I actually lol'd at that. You calculated a correlation coefficient, not a linear regression. Regardless, what the hell is that equation? I'm still giggling. You need the slope coefficient to predict the effect of a current period loss in the next period. You did not perform a regression. The correlation coefficient does not work like this. This equation is not a thing


The higher the correlation coefficient that excel calculated for you (dont blindly use excel formulas unless you understand them to avoid embarrassment like this), the closer the two are to a perfect linear relationship. Logically, that sounds reasonable for your chosen variables.

What you tried to interpret that as, is the regression (or slope) coefficient. The slope coefficient is the extent to which the change in the expected value of period 2 kills-losses is explained by period 1 kills-losses. That is what you were going for, but they are very different in both calculation and interpretation.

Your calculation of the correlation coefficient explains precisely nothing. It has no weight over what you were trying to prove. Try again.

PS: de-whoring kills is ********. Basically only shows small gang and solo kills as a result. de-whored kills - losses will be VASTLY different than just kills-losses

TLRD: A correlation coefficient is not a regression coefficient. Try again with a linear regression and post the F test, T statistics, R^2, durbin watson statistic. You cannot predict the effect of a period 1 freighter gank on period 2 kills-losses using a correlation coefficient. Using excel formulas without understanding the statistic provides great laughs for everyone. Try googling regression vs correlation coefficient next time



Gevlon, I would like either a response, or a re-write of your June 17th blog please. It is a misinformed travesty, and as long as it stands in it's current form it is testament to the fact that while you claim to be a math and business expert, you are in fact completely clueless. It is in your best interest to go back and run the regression you thought you were running, or remove your incorrect interpretation of the correlation coefficient.

I have sent him an eve mail asking for a re-write. Glossing over this just accentuates his denialism and arrogance
Agata Matahari
Royal Amarr Institute
Amarr Empire
#568 - 2014-06-18 14:39:12 UTC
Is Gevlon as hot as his statistics?
Big Lynx
#569 - 2014-06-18 14:59:23 UTC
Gevlon's quote of May 2nd in Foo's Eve Musings:

" I'm fighting Goons for no other reason than them being "evil". They never really hurt me and they have nothing I want. I simply want to put my mark on EVE history by this "crusade". What Noir did was evil. Hitting them is as good as hitting Goons. And let's face it, they are much easier target, with the prospect of an early and hilarious victory (titan KM)"

Now all is crystalclear for me. that pub is so amusing.
Gevlon Goblin
Science and Trade Institute
Caldari State
#570 - 2014-06-18 15:18:16 UTC
Catalytic morphisis wrote:
**** son! Gevlon got owned! This made me chuckle, Blag math vs Real math :) For taking the time to write this out you are now my hero!

You know what you "hero" didn't say?
He didn't say that my results and conclusion is wrong. You believe that they are wrong, based on what he said. But he didn't say it, since he didn't want to get caught lying. All he said is a bunch of math words you don't understand, with the meaning "his methods weren't prudent enough".

When you go to get the autogram of your "hero", why don't you ask him if my conclusion was indeed incorrect? Beware! He might blush, stand from one feet from another and say "actually, Gevlon said losing 1B will make a Goonie have half B less activity, while my absolutely perfect calculation says 0.412+-0.213 with 90% confidence, so he was technically wrong".

I doubt if many people on this forum even heard the word "regression" before. The question is "does being ganked (preferably in a bountied titan) makes a Goonie decrease his PvP activity or will he just laugh, reship and go on like nothing happened".

My conclusion: significant decrease.
His conclusion: Gevlon used wrong confidence interval and his data is noisy and didn't account for logies. And the exam season. And used "then" instead of "than" in the foreword.

My blog: greedygoblin.blogspot.com
Radric Davids
GoonWaffe
Goonswarm Federation
#571 - 2014-06-18 15:40:59 UTC  |  Edited by: Radric Davids
Gevlon Goblin wrote:
Catalytic morphisis wrote:
**** son! Gevlon got owned! This made me chuckle, Blag math vs Real math :) For taking the time to write this out you are now my hero!

You know what you "hero" didn't say?
He didn't say that my results and conclusion is wrong. You believe that they are wrong, based on what he said. But he didn't say it, since he didn't want to get caught lying. All he said is a bunch of math words you don't understand, with the meaning "his methods weren't prudent enough".

When you go to get the autogram of your "hero", why don't you ask him if my conclusion was indeed incorrect? Beware! He might blush, stand from one feet from another and say "actually, Gevlon said losing 1B will make a Goonie have half B less activity, while my absolutely perfect calculation says 0.412+-0.213 with 90% confidence, so he was technically wrong".

I doubt if many people on this forum even heard the word "regression" before. The question is "does being ganked (preferably in a bountied titan) makes a Goonie decrease his PvP activity or will he just laugh, reship and go on like nothing happened".

My conclusion: significant decrease.
His conclusion: Gevlon used wrong confidence interval and his data is noisy and didn't account for logies. And the exam season. And used "then" instead of "than" in the foreword.


What I said: gevlon calculated the wrong statistic using the wrong excel formula he doesnt understand.

What you just said: I dont understand it, and I dont understand what he said so he must not have proved me wrong.

Reality: you wrote a blog post in an attempt to use big words and calculations to 'prove' that you are beating the cfc. Hilariously, you got in way over your head and spat out some numbers with no understanding of how to interpret them. You used the wrong formula, calculated the wrong stat which doesn't mean what you think it does, and instead of going back and doing an actual regression (hint, its another excel formula) and seeing the results you are denying everything. I read your post, laughed, and decided to kill some time at work to try to explain it in depth.

Regressions are not complicated. It is what you tried to do. A regression creates a formula (basically a coefficient) which predicts the effect of a kill-loss today on kills-losses tomorrow. You failed and used the wrong formula, and created some numbers which do not have anything to do with the influence of current period kills-losses on next period kills-losses.

Its all really quite simple. You tried to be smart, you failed, and your entire blog post is a failure. You spat out a statistic which does not say what you claim it says, and anyone familiar with basic statistics can see it at first glance. Denying it only makes it more hilarious. This is why I suggested you use the right formula, and see what it says. I'm curious
Catalytic morphisis
Ruthless Regiment
SL0W CHILDREN AT PLAY
#572 - 2014-06-18 15:43:32 UTC
Gevlon Goblin wrote:
Catalytic morphisis wrote:
**** son! Gevlon got owned! This made me chuckle, Blag math vs Real math :) For taking the time to write this out you are now my hero!
You know what you "hero" didn't say? He didn't say that my results and conclusion is wrong. You believe that they are wrong, based on what he said. But he didn't say it, since he didn't want to get caught lying. All he said is a bunch of math words you don't understand, with the meaning "his methods weren't prudent enough". When you go to get the autogram of your "hero", why don't you ask him if my conclusion was indeed incorrect? Beware! He might blush, stand from one feet from another and say "actually, Gevlon said losing 1B will make a Goonie have half B less activity, while my absolutely perfect calculation says 0.412+-0.213 with 90% confidence, so he was technically wrong". I doubt if many people on this forum even heard the word "regression" before. The question is "does being ganked (preferably in a bountied titan) makes a Goonie decrease his PvP activity or will he just laugh, reship and go on like nothing happened". My conclusion: significant decrease. His conclusion: Gevlon used wrong confidence interval and his data is noisy and didn't account for logies. And the exam season. And used "then" instead of "than" in the foreword.
thats rich gevlon, Took you a whole 3 hours to write out that response? and just because you don't understand it does not mean the rest of us don't, Just because you're small, synapse withdrawn brain cannot comprehend the fact that you're actually completely wrong there is no need to go simply trying to insult people, it gets you nowhere. I can safely tell you without doing a singular piece of mathematics or making up bullshit graphs and charts with no actual correlation to real life effects that your petty little "crusade" is nothing more than laughable to anybody who understands this games social/economic standpoint in terms of 0.0 alliances, You sir are simply wasting your own ISK and time trying to convince people of something which we all know is completely and irrefutably incorrect.

Actual Link free and scout free solo PvP'er
Catalytic morphisis
Ruthless Regiment
SL0W CHILDREN AT PLAY
#573 - 2014-06-18 15:45:50 UTC  |  Edited by: Catalytic morphisis
Edit: slow internet results in double posts

Actual Link free and scout free solo PvP'er
Lucas Kell
Solitude Trading
S.N.O.T.
#574 - 2014-06-18 15:49:30 UTC
Gevlon Goblin wrote:
You know what you "hero" didn't say?
He didn't say that my results and conclusion is wrong. You believe that they are wrong, based on what he said. But he didn't say it, since he didn't want to get caught lying. All he said is a bunch of math words you don't understand, with the meaning "his methods weren't prudent enough".
You aren't very bright, are you Gevlon. What he stated is that your method of analysis was highly flawed. He wouldn't say whether it is right or wrong as he doesn't have the data to make that judgement, unlike you, he doesn't leap to conclusions. But that doesn't mean you are correct.

Think of it like this. If you were to state that a cat in a box, which we can't see is white based on the fact that the box is brown, you might be correct or you might be incorrect, but your method of analysis is flawed. For you to make a reasonable conclusion you would need to correctly analyse the data available.

Gevlon Goblin wrote:
The question is "does being ganked (preferably in a bountied titan) makes a Goonie decrease his PvP activity or will he just laugh, reship and go on like nothing happened".
Well since participation numbers and members numbers of the CFC are increasing while the number of gank losses is increasing I'd say the conclusion is that no, the ganking doesn't decrease his activity. I'd suggest that perhaps there could be alternate explanations, like you comparing wartime statistics to peacetime statistics, CFC members becoming more efficient with more support roles or you know, a completely flawed analysis of the data.

Gevlon Goblin wrote:
My conclusion: significant decrease.
His conclusion: Gevlon used wrong confidence interval and his data is noisy and didn't account for logies. And the exam season. And used "then" instead of "than" in the foreword.
He didn't make a conclusion about the CFC data, he just pointed out that your conclusion is about as accurate as rolling dice for the answer.

It's quite funny watching you desperately scrabble to hold on to any shred of dignity remaining after you essentially showed that you shook excel until a formula came out then guessed a result.

The Indecisive Noob - EVE fan blog.

Wholesale Trading - The new bulk trading mailing list.
Radric Davids
GoonWaffe
Goonswarm Federation
#575 - 2014-06-18 16:05:30 UTC  |  Edited by: Radric Davids
Gevlon, your conclusion has absolutely no weight, because it is not actually a conclusion. You applied the wrong process, got the wrong answer, and you interpreted it as a conclusion because you don't understand what you did wrong.

You took facts, but applied the wrong process, and now you are denying what exactly? That the process was not wrong? Explain to use exactly how you aren't wrong. So you don't understand how correlation coefficient differs from a linear regression coefficient, whatever. You can go back and fix it, and maybe the numbers will reflect what you think they will.

In it's current form, your 'conclusion' has absolutely no basis in statistics or facts.


You came to a statistical conclusion using the completely wrong statistic, and you are claiming that your conclusion is correct because you don't understand the underlying math that I explained to you a few posts back?

Use the wrong process, get the wrong answer. It's like logic 101, I cant believe i'm even saying this. I explained to you precisely what you did wrong, and exactly how you could fix it. I hope you do go back and do it properly by applying a linear regression, because I am honestly curious about the result. If you actually decide to do so, please post tests of statistical significance.

This is just too funny. Gevlon tried to get all mathy and smart, thinking nobody would understand. He got called out, and actually also shown how to fix the problem. Instead of fixing it and proving us all wrong, he denies it because we 'used big words' he doesn't understand.
Khanh'rhh
Sparkle Motion.
#576 - 2014-06-18 16:14:26 UTC  |  Edited by: Khanh'rhh
Gevlon Goblin wrote:
Catalytic morphisis wrote:
**** son! Gevlon got owned! This made me chuckle, Blag math vs Real math :) For taking the time to write this out you are now my hero!

You know what you "hero" didn't say?
He didn't say that my results and conclusion is wrong. You believe that they are wrong, based on what he said. But he didn't say it, since he didn't want to get caught lying. All he said is a bunch of math words you don't understand, with the meaning "his methods weren't prudent enough".

When you go to get the autogram of your "hero", why don't you ask him if my conclusion was indeed incorrect? Beware! He might blush, stand from one feet from another and say "actually, Gevlon said losing 1B will make a Goonie have half B less activity, while my absolutely perfect calculation says 0.412+-0.213 with 90% confidence, so he was technically wrong".

I doubt if many people on this forum even heard the word "regression" before. The question is "does being ganked (preferably in a bountied titan) makes a Goonie decrease his PvP activity or will he just laugh, reship and go on like nothing happened".

My conclusion: significant decrease.
His conclusion: Gevlon used wrong confidence interval and his data is noisy and didn't account for logies. And the exam season. And used "then" instead of "than" in the foreword.

So you lack the stats knowledge of even the common lay-person, who can successfully read his post and know that it irrefutably shows that, no, your conclusion is in no-way supported by your almost completely random mashing of numbers you call an analysis?

I've titled your autobiography for you:

'The Story of Gevlon Goblin: Shifting goalposts in a Dunning-Kruger field of dreams'

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Khanh'rhh
Sparkle Motion.
#577 - 2014-06-18 16:24:07 UTC  |  Edited by: Khanh'rhh
Radric Davids wrote:
What I said: gevlon calculated an irrelevant statistic using an excel formula he doesnt understand.

What you just said: I dont understand it, and I dont understand what he said so he must not have proved me wrong.

Reality: you wrote a blog post in an attempt to use big words and calculations to 'prove' that you are beating the cfc. Hilariously, you got in way over your head and spat out some numbers with no understanding of how to interpret them. You misinterpreted them, and instead of going back and doing an actual regression (hint, its another excel formula) and seeing the results you are denying everything. I read your post, laughed, and decided to kill some time at work to try to explain it in depth.

Regressions are not complicated. It is what you tried to do. A regression creates a formula (basically a coefficient) which predicts the effect of a kill-loss today on kills-losses tomorrow. You failed and used the wrong formula, and created some numbers which do not mean anything significant.

Its all really quite simple. You tried to be smart, you failed, and your entire blog post is a failure. You spat out a statistic which does not say what you claim it says, and anyone familiar with basic statistics can see it at first glance. Denying it only makes it more hilarious. This is why I suggested you use the right formula, and see what it says. I'm curious


A regression of the data he's looking at, even properly done, still isn't going to give him data that supports his conclusion. His assertion that killboard stats = activity is an unproven theory. Even if you accept that as being true, it creates problems. He's claiming:
- activity can be shown by killboards (does someone who doesn't PVP therefore not exist?)*
- all other factors being constant, killboard activity is level over time (it isn't)
- that a "loss" = "a loss" in an organisation with up-to 200% SRP

It's about as useful as looking at shoe-size over time to chart the rise of fascism; even if a significant correlation existed, it still isn't the causal-link and you need to start looking at what factors are causing both (see my previous silly correlations).


* - based on his own killboard stats, one can conclude that ganking miners makes you stop playing the game, since he has no activity after that.

"Do not touch anything unnecessarily. Beware of pretty girls in dance halls and parks who may be spies, as well as bicycles, revolvers, uniforms, arms, dead horses, and men lying on roads -- they are not there accidentally." -Soviet infantry manual,

Radric Davids
GoonWaffe
Goonswarm Federation
#578 - 2014-06-18 16:47:35 UTC  |  Edited by: Radric Davids
Khanh'rhh wrote:
Radric Davids wrote:
What I said: gevlon calculated an irrelevant statistic using an excel formula he doesnt understand.

What you just said: I dont understand it, and I dont understand what he said so he must not have proved me wrong.

Reality: you wrote a blog post in an attempt to use big words and calculations to 'prove' that you are beating the cfc. Hilariously, you got in way over your head and spat out some numbers with no understanding of how to interpret them. You misinterpreted them, and instead of going back and doing an actual regression (hint, its another excel formula) and seeing the results you are denying everything. I read your post, laughed, and decided to kill some time at work to try to explain it in depth.

Regressions are not complicated. It is what you tried to do. A regression creates a formula (basically a coefficient) which predicts the effect of a kill-loss today on kills-losses tomorrow. You failed and used the wrong formula, and created some numbers which do not mean anything significant.

Its all really quite simple. You tried to be smart, you failed, and your entire blog post is a failure. You spat out a statistic which does not say what you claim it says, and anyone familiar with basic statistics can see it at first glance. Denying it only makes it more hilarious. This is why I suggested you use the right formula, and see what it says. I'm curious


A regression of the data he's looking at, even properly done, still isn't going to give him data that supports his conclusion. His assertion that killboard stats = activity is an unproven theory. Even if you accept that as being true, it creates problems. He's claiming:
- activity can be shown by killboards (does someone who doesn't PVP therefore not exist?)*
- all other factors being constant, killboard activity is level over time (it isn't)
- that a "loss" = "a loss" in an organisation with up-to 200% SRP

It's about as useful as looking at shoe-size over time to chart the rise of fascism; even if a significant correlation existed, it still isn't the causal-link and you need to start looking at what factors are causing both (see my previous silly correlations).


* - based on his own killboard stats, one can conclude that ganking miners makes you stop playing the game, since he has no activity after that.


Very true. I find it very hard to believe that pvp activity in a future period is heavily influenced by pvp activity/losses in the current period. I would think that other influences would have a much larger impact on future pvp activity, and measuring all of this by net isk killed is a huge stretch.

I just noticed the fail math, and decided to call goblin out on his bs. He clearly tried to get smart on us, and failed miserably.

Like every other conclusion/claim he has ever made, his idea that future net isk killed is a function of current isk killed is somewhere between a massive stretch and a completely manufactured casual relationship.
Ria Nieyli
Nieyli Enterprises
When Fleets Collide
#579 - 2014-06-18 17:03:59 UTC  |  Edited by: Ria Nieyli
Gevlon,

Are you actually autistic? I'm asking cause it seems that it was thrown around a couple of times as a possibility in this very thread. I'm just curious.
Charles Case
State War Academy
Caldari State
#580 - 2014-06-18 17:25:31 UTC
Gevlon have you ever toasted to yourself while talking to yourself in front of the mirror?