The secret of data crunching

By Yearfire, in X-Wing

21 minutes ago, baranidlo said:

Even without list juggler/fortress we would be better off, with more home-brewing and creativity and less copying.

What makes you think without the kind of data provided by list juggler/fortress people will copy less, and not copy just as much, but from a narrower pool of lists (what does really well at large events)?

10 minutes ago, LordBlades said:

What makes you think without the kind of data provided by list juggler/fortress people will copy less, and not copy just as much, but from a narrower pool of lists (what does really well at large events)?

Just anecdotal evidence. I see players in my area trying and testing all kinds of interesting lists since 2.0 came, and I think it is at least partly caused by lack of easy data source.

3 hours ago, GreenDragoon said:

Most people use excel and that's usually enough. I do my stuff in RStudio, using the statistics software R. It's very convenient to juggle around large data sets. Going through those 10'000 games in excel is probably not as convenient.

This. Excel is good for compiling and sorting data. It is a spreadsheet program, a powerful one. But doing graphs and statistics, this is ahassle in Excel, there are much better programs, which often can import xls. R, Statistica, SPSS, etc.

Then as several said here interpretation is tricky. You have to be aware of what question is asked, how it is asked, and what goes in as sample. Look at @GreenDragoon 's post about Poe and Fenn, that's a good and instructional example.

Then, while I personally find doing and reading statistics pretty fun, I am very doubtful about mathjuggler etc. These attempts at data analysis (grounded on very small and sometimes faulty samples) in themselves lead to self-fortifying "run-away-selection" of certain lists, which get more and more popular, due to being promoted by math analysis (regardless of proper done or not), pods and people (mis)interpreting these analyses, running even more of the popular stuff. Leading to a few pretty boring lists, the same c r a p everywhere.

I hope that FFG often enough shakes the meta by very slightly*) adjusting points.

*)not so much that ships are rotated out all the time, forcing to buy new ones constantly, and leading to sadness as ones beloved ships die off, by exaggerated point changes

14 minutes ago, baranidlo said:

Just anecdotal evidence. I see players in my area trying and testing all kinds of interesting lists since 2.0 came, and I think it is at least partly caused by lack of easy data source.

I agree, however I don't think the easy data source is, or ever was, ListJuggler.

IMO the easy data source is the big tournament results, and we didn't really have many of those so far.

52 minutes ago, baranidlo said:

Just anecdotal evidence. I see players in my area trying and testing all kinds of interesting lists since 2.0 came, and I think it is at least partly caused by lack of easy data source.

You don't need to force people to innovate and experiment by keeping them in the dark, it's part of our nature, it's fun, it feels good. So long as FFG does their part in keeping the game balanced and making sure that a subsection of the game isn't dominating due to being too points efficient, having more data available will just make it easier to innovate effectively, not harder. Without the data we get from an aggregate like this, people will have to work harder and longer chewing through the weeds to find new stuff that works, and some will get discouraged and just look at whatever did well at the few large tournaments whose results were streamed and pick up one of those lists. I would argue that once we move out of the current honeymoon phase that we're in now, having this kind of raw data will only serve to allow the game to maintain list diversity.

1 hour ago, baranidlo said:

Just anecdotal evidence. I see players in my area trying and testing all kinds of interesting lists since 2.0 came, and I think it is at least partly caused by lack of easy data source.

Could it also be that 2.0 is new, and players are trying out all the new and changed stuff?

1 hour ago, Tvboy said:

You don't need to force people to innovate and experiment by keeping them in the dark, it's part of our nature, it's fun, it feels good. So long as FFG does their part in keeping the game balanced and making sure that a subsection of the game isn't dominating due to being too points efficient, having more data available will just make it easier to innovate effectively, not harder. Without the data we get from an aggregate like this, people will have to work harder and longer chewing through the weeds to find new stuff that works, and some will get discouraged and just look at whatever did well at the few large tournaments whose results were streamed and pick up one of those lists. I would argue that once we move out of the current honeymoon phase that we're in now, having this kind of raw data will only serve to allow the game to maintain list diversity.

Ok, might be. However I remember very well how Listjuggler and Metawing reinforced the echo chamber mentality during the dark days of 1st edition and contributed to the super toxic environment in the forums. Sometimes less information can be good, if the noise level is high..