Just now, Stay On The Leader said:Haha, I have no sway that I'm aware of!
You write a blog that is frequented quite often from what I understand. Post in this forum quite often, and seem to have people that respect your opinion. I would call that sway.
Presenting conclusions based on incomplete data should ALWAYS have a disclaimer IMO, but even more so in this case.
4 minutes ago, Stay On The Leader said:I believe in it, of course.
My heart bleeds if that's true. Yes, I do maths/algorithms/machine learning for living.
Edited by Voitek1 minute ago, Lampyridae said:Wow. I knew I was handicapping myself with them but not this badly.
Good news, you arent, all this data is very innaccurate at best.
3 minutes ago, Stay On The Leader said:Haha, I have no sway that I'm aware of!
You've got loads of sway, look at the amount of people who want to argue with you 
3 minutes ago, Timathius said:Presenting conclusions based on incomplete data should ALWAYS have a disclaimer IMO, but even more so in this case.
This is the internet. Everything is disclaimered.
Fast forward to one year from now: "Guys, it's only been a year since the FAQ and Imperials disappeared, that's too small a sample size to jump to conclusions. Lets give it until 2020 and see where they are. Maybe Wave XIII will help them out. Now, who wants to play RebScumWing?"
1 minute ago, FlipmodeSH said:You've got loads of sway, look at the amount of people who want to argue with you
I thought I was just lucky that way!
3 minutes ago, Voitek said:My heart bleeds if that's true. Yes, I do maths/algorithms/machine learning for living.
So you think the results are significantly wrong, or are you throwing the baby out with the bathwater?
Just now, FlipmodeSH said:You've got loads of sway, look at the amount of people who want to argue with you
Some people have issues with accepting skewed numbers showed and presented as truth.
(I do for sure :P)
52 minutes ago, Admiral Deathrain said:Pretty crazy that A-Wings are even lower than Punishers and that B-Wings are lower than TIE Bombers. You would think at least someone would fly Snapswarms. They do look rather good, especially with OpSpec TIEs in the mix.
I think after seing A-wings in the top 8 of the last open people will start giving them a chance. The ship i'm most surprised seing so low is the ARC. It's a really good ship, especially when flown by Nora. The person i practice the most with has a skill level a bit lower than me. But it's now his favorite ship because hes able to compete with it and "sometime" beat me :-) Not a bad dial, 2 arcs with a title to enhance both and a pilot ability to add a dice on both offense and/or defense... I think its a sleeper hit. Hes playing her with that that nice BB8/marksmanship/kanan combo
3 minutes ago, Johen Dood said:Fast forward to one year from now: "Guys, it's only been a year since the FAQ and Imperials disappeared, that's too small a sample size to jump to conclusions. Lets give it until 2020 and see where they are. Maybe Wave XIII will help them out. Now, who wants to play RebScumWing?"
Yeah, any arguments, no matter how illogical they are...
4 minutes ago, Stay On The Leader said:This is the internet. Everything is disclaimered.
If only.
2 minutes ago, Voitek said:Good news, you arent, all this data is very innaccurate at best.
It is not, only the most recent entry. However things that aren't getting a lot of playtime can still be good. That is the thing, a meta analysis doesn't include everything that is good, only what is popular. Which usually is good, but not definitive. Look at Dengaroo - it took a long while before that list bubbeled up, yet it would have been just as good before that.
Snap-A-Wings could be one of those cases, if not as clear cut. I think they are good, but they don't break the bank.
3 minutes ago, Voitek said:Some people have issues with accepting skewed numbers showed and presented as truth.
(I do for sure :P)
I presented numbers as numbers. Drawing conclusions is at the reader's risk.
Just now, Stay On The Leader said:So you think the results are significantly wrong, or are you throwing the baby out with the bathwater?
No, the results are showing completely different thing than you are "selling" it as.
This particular set of data should be clearly labeled as: THE MOST POPULAR SHIPS IN TOP8 CUTS.
If you publish this data labelled as SHIP RANKING, I can only call BS on that.
Not including a disclaimer about that in bold, in the first sentence, above the table, is just pure populism. Am I telling the numbers themselves are useless/innaccurate? No, not at all, I'm sure they are actually showing ships popularity in TOP8 cuts. But that number all by itself does say very little about the state of the meta.
1 minute ago, Stay On The Leader said:I presented numbers as numbers. Drawing conclusions is at the reader's risk.
Upload a pic labelled as SHIP RANKING and expect people not to draw conclussion that is the ship strength/power ranking. That's just naive.
10 minutes ago, Voitek said:Good news, you arent, all this data is very innaccurate at best.
Inaccurate means there is a measurement or coding issue. Unless SoTL has a
The list presents exactly what it is meant to present. It's ranked data on what makes the cut, I don't need someone doing Rasch analysis to tell me that A-Wings suck. Agreed that ranking is misleading however.
Edited by LampyridaeJust now, Voitek said:No, the results are showing completely different thing than you are "selling" it as.
This particular set of data should be clearly labeled as: THE MOST POPULAR SHIPS IN TOP8 CUTS.
Except this data is CERTAINLY not that.
It's neither a measure of popularity nor a measure just of Top-8 cuts. It's a measure of success drawn from a data pool that is weighted overly-heavily towards Top-8 cuts. If anything a ship that is hugely popular is at a disadvantage in this ranking as they can't ALL win.
3 minutes ago, Admiral Deathrain said:It is not, only the most recent entry. However things that aren't getting a lot of playtime can still be good. That is the thing, a meta analysis doesn't include everything that is good, only what is popular. Which usually is good, but not definitive. Look at Dengaroo - it took a long while before that list bubbeled up, yet it would have been just as good before that.
Snap-A-Wings could be one of those cases, if not as clear cut. I think they are good, but they don't break the bank.
It really is. Most likely A-wing data is erroneous, to start with, not to mention limited amount of results being analysed and calculated compring to total number of lists being flown on every tournament.
Granted that the after nerf hasn't settled down just yet, but if you look at wave 9 to nerf was the most balanced time ever out of the list based on ships.
Double post...
Edited by StrikesbackSheesh.... Everyone has plenty of salt for anything you read on here.
It is simple enough to have a quick discussion of 'How much data is in that section?' and let people draw their own conclusions. This all seems rather personal.
Relax, it is Friday!
Voitek has a personal issue with me, which I think is colouring his approach to this unfortunately.
1 minute ago, Stay On The Leader said:Except this data is CERTAINLY not that.
It's neither a measure of popularity nor a measure just of Top-8 cuts. It's a measure of success drawn from a data pool that is weighted overly-heavily towards Top-8 cuts. If anything a ship that is hugely popular is at a disadvantage in this ranking as they can't ALL win.
As you are the one that created the chart, can you give us a bit more details what was actually being used to populate the data?
- all tournaments / large tournaments / Open series+Regio tournaments?
- all available lists / TOP8 / CUT whenever its TOP8, 16 or 32?
- if all lists uploaded were analysed, then if someone puts his list that ended up on position 237 on a 275 people tournament, is he going to be included as well?
1 minute ago, FlipmodeSH said:Sheesh.... Everyone has plenty of salt for anything you read on here.
It is simple enough to have a quick discussion of 'How much data is in that section?' and let people draw their own conclusions. This all seems rather personal.
Relax, it is Friday!
That is really all I am asking for is for him to add that to his first post. I don't think I have made it at all personal, and if I have I apologize.
Based on 25 events across the world, mostly store kits, with very limited data entry.
Done.