There is no correlation between boot rate and percentage of real time games. Isn't that a lot more simple than writing up a script, plotting the data and uploading the graph to an export device?

export device

What exactly is an "export device"?

So I could just claim that no correlation exists... or actually prove it and show that there's no good reason to separate RT and MD boot rates since RT% has no discernible effect on boot rate.

(Also, the code for this was ~15 lines. I already had the data for other reasons.)

Edited 8/29/2015 22:42:59

@Genghis What can you ask me about The Circle?

You upload a file, it becomes a url for Internet.

Ifk ox. I get you mixed up with Only the pie sometimes.

Oh. People are more likely to understand you (at least in the English-speaking world) if you call it an "image host" or (if you feel like being fancy) an "image-hosting service." "Device" is usually used to refer to a physical piece of mechanical or electronic equipment- for example, my Parrot AR drone is a "surveillance device" (not really). On the other hand, offering software and space for someone else to use is a type of "service"- an action that you offer to do for someone else, often in exchange for something.

Congratulations on your newfound tech literacy. You can now use a computer, gramps! :)

Edited 8/29/2015 22:51:18

"Okay now click the start button in the bottom corner."
*click*
"Check it out dad! I'm computing!"

Knyte, i don't need your patronage you scrupulent box-knave. Perhaps return to my quarter when you're literate in actual world language, savvy? Also, i find value in a good book before battle. I do not prefer tacitus however; his pages are quite rough on the nether regions.

2/10 trying too hard

@knyte Perhaps an inverse Bell graph?

^ Not a statistician, but aren't Bell curves for 1D distributions?

I don't think that's something I can use to perform a regression, but if there's a way I'll do it.

EDIT: TIL that Gaussian regressions are a thing. Doing one right now :D

Edited 8/29/2015 23:15:24

How about removing some data as the random 90 percent boot rate. It seems to be not relevant.

Bias.

knyte after seeing all your programming and knowledge of data handling, and the fact that you aren't in a clan, why aren't you in the Statisticians clan? (https://www.warlight.net/Clans/?ID=100)

Hostile, how can that be bias if the 100 percent boot rate is on both sides, both 100% rt and 100% md? Bias would be to take it away.

^ Because I'm not an actual statistician and haven't ever been employed in a data science capacity in my life. I have a bare-bones knowledge of data that is nowhere near the standards I expect that clan to have.

This was literally my first time conducting a polynomial regression, ever. I was expecting someone with statistically knowledge (Corvus5, Math Wolf, etc.) to pop in and tell me how to do things better. I'm someone who occasionally PMs Math Wolf with questions about stats and then pretends to understand his answers, not a real statistician. :)

Edited 8/29/2015 23:37:39

Is simple batch files with other programs good for statistic program ? Batch is fast with sorting because of regex and it is the only way to make pipes as far as I know.

Pipe is output of program to input of other program.


How to fortran or buy a computer for scientific high level languages. Fortran is best. Every computer that is not mad with fortran in mind is crap.



In mechanotronics, we learn that everything is crap because taking shortcuts. Even if the application of your electronics and not meant to be precise, it is still crap if using shortcuts.

Must always make high quality and be careful. You can only charge for hours and different procedures/tools your application requires. Don't charge for making good electronics.

Edited 8/29/2015 23:47:23

Eh I just used scipy/numpy/matplotlib. I'm not a real statistician by any stretch of the imagination.

Why not r ?

Matrices are cool. So are lattices.

Do you know what a mathematical lattice is ?



Edit : I meant " lacet mathematique"

Path from point to point that returns to origin.



Also, did anyone tried to read chesterton. Google gives me crap even if Chesterton was said to be the most rude and prestigious critic of its time. He made ton of critics. I wondered if it is worth reading.

Edited 8/29/2015 23:56:10

I'm obviously biased, but I'm curious as to why people put 20 above us. I honestly could understand someone saying GG and WG was better than Apex, because GG and WG have added people to their rosters and they're bigger clans than Apex are. They always compete at a very high level everywhere they are. However, 20 is half retired. 20 didn't even put together a roster to play in A league for Clan League. I'd love to see them do it, so Apex could absolutely destroy them and send them home crying. Their members are rusty aside from Piggy and HHH, and they lack experience in some of the templates we play in Clan League. They also lack experience in the new templates for 1v1 2v2 and 3v3 going around nowadays. Everyone else is up-to-date in the Elite clan category except for 20.

Is it the fact that they're legendary status? Timi, Frank, grona, brisk, and Widzisz impress me on a daily basis with their playing ability, and they're all still playing. Dunga, Luxis, and Huruey were huge in the elite scene back when 20 players were also playing.

Unpopular opinion? Maybe. I just could never rate 20 above any of the other 'elite' clans right now. The way I see it, Apex, WG, and GG are pretty much dead even right now and the next clan league will decide whether or not the roster changes and WG/GG put them ahead of Apex, or if Apex really is still ahead of everyone. So my list would be:

1. Apex,WG,GG
4. 20
5. Lynx,Blitz
7. French
8. AHoL
9. XS
10. Warlighters

No specific reason. Python is just my general-purpose quick-to-write-quick-to-debug language. Plus, while I'm familiar with the statistical side of R, I haven't quite learned how to make HTTP requests in that language while I have learned how to handle data in Python. If I just want to write a quick script and leave it running while I make dinner, it's Python.

Yes, I agree that lattices are cool. As long as it's not set topology, I can find any branch of math fun and interesting.

@Lolo: it's more of a contrarian-I-can't-really-think-of-someone-better-than-these-guys-so-why-not-go-for-nostalgia thing. Y'know, like saying the 60's Celtics were a better team than the 90's Bulls or ranking the Lakers near the top in a list of franchises. They were great at their peak, but you're right that we're all being jerks by deliberately using them as an excuse to not give the #1 rank to Apex/GG/WG.

Edited 8/30/2015 00:00:42

I also kind of find it funny since 20 asked if Apex would merge with them. I find that a lot of GG/WG members would fit the 20 clan mentality of being comprised of the 20 most impactful players of Warlight.

Edited 8/30/2015 00:03:56

Unrelated, but I've always found it funny how Genghis/Kapy leap at me with insults whenever my behavior hints at any semblance of intelligence. Last time I remember Kapy calling me illiterate for writing more than four paragraphs; this time it's Genghis+Kapy coming after me for knowing how to use PyPlot.

Odd that they're the ones telling other people to improve their behavior.