Page 2 of 2

Re: Classify backgrounds into difficulty

PostPosted: Monday, 27th August 2018, 13:10
by aperiodic
Interesting chart. Looks like somebody really likes NaWn.

Re: Classify backgrounds into difficulty

PostPosted: Monday, 27th August 2018, 15:49
by Arrhythmia
aperiodic wrote:Interesting chart. Looks like somebody really likes NaWn.


That green means that they're pure gold baby, ain't no combo better than *checks notes* NaWn?

Re: Classify backgrounds into difficulty

PostPosted: Monday, 27th August 2018, 17:47
by Berder
aperiodic wrote:Interesting chart. Looks like somebody really likes NaWn.

Dynast started streaking them and then I tried a copycat streak. NaWn is surprisingly very powerful and easy to streak using the formula Dynast invented: train Fighting and Stealth at first until you find a good weapon, spit poison a lot, worship Gozag, wear a shield, ignore Dodging and wear heavy armor when you find a decent one. Use Gozag for armor shops to help get good armor early. Prefer to use Broad Axes but a morningstar or demon weapon is also ok.

Re: Classify backgrounds into difficulty

PostPosted: Wednesday, 5th September 2018, 05:17
by EtadanikM
Can you do an analysis of the highest skills for end game characters? I think that might be a better look at the balance in the game, since ultimately, I'd say that, in order of importance for success:

choice of skills > species > god > "class"

Most people follow a very similar progression of skills based on strategy, and that seems to be where the game has minimal diversity. Then, species have a very obvious hierarchy, followed by gods, the latter being less important only because they're not chosen until significantly later. Finally, the "class" of the character has a large effect early on, but you can always abandon the "class" to go your own direction and most "classes" do not lock you into a play style.

Re: Classify backgrounds into difficulty

PostPosted: Thursday, 6th September 2018, 17:11
by gameguard
This is more of a popularity chart than anything. The data set is so restrictive that a handful of players can affect the outcome so much. If you take out Berder and Dynasts Wn streak, what would happen? DsWn is probably 80% my games. Was Wn actually the best class to streak with? Or was it used by players that can streak with any combo because its fun and "challenging." Is naga better than Mi??? This type of approach might work if crawl had x1000 players but not now. I know jack shit about statistics but even I can tell at a glance that the updated streak chart is extremely biased by what a handful if players were doing.