S10 Analysis: Why Swiss shouldn't be used next season
-
@rn fax. need to incorporate this next season @admins
-
Mothership I’d rather add in cascade and lakeside rather than borneo, it would keep it to 4 koth and 4 stopwatch
-
Was grinning the entire time I read this, excellent analysis! The graphs especially were very well thought-out/chosen and informative. You should consider going into Data Science.
-
Excellent write up BBL. I don’t really care to much about the math, but swiss is just super prone to screwing some teams over. In hindsight, I think my team (Vindicators) ended up getting punished for taking a round off Train Station during week 1. If we had lost 2-0, then we would have been placed against a bottom team; instead, we were placed against the current 2nd place advanced team. In a swiss system, I feel like the optimal thing to do would just to be to sandbag/lose early in the season and rack up points against the bad teams that die mid season (no disrespect to Broken Mice)
-
@Pain-Seer or add another steel-like for 3 koth 3 pl 2 steel like
-
@Pain-Seer said in S10 Analysis: Why Swiss shouldn't be used next season:
Mothership I’d rather add in cascade and lakeside rather than borneo, it would keep it to 4 koth and 4 stopwatch
oh you right, but if we add them all in, 9 week season, 10 team invite, 10 team Advanced with RR. !!!
-
Mothership I have no issue with this
-
This seems like something RGL staff should consider. Maybe put a question in the end of season polls.
I can’t tell if my sarcasm detector is broken - but I’m going to put this here for the sake of clarity
I know this isn’t what’s being recommended, but tangentially relevant quote from discord
A round robin for all divisions effectively means 14+ divisions
logistically, how do you handle who gets into which division
…
If HL has two matches a week, you’d have 14 a season and it would fix a lot of the issues when it comes to swiss
Source: °fa-info°(exa)Copy pasting from my discord #discussions post
I think to summarize my understanding: Swiss is rarely ideal but often necessary.
my biggest issue [with BBL’s Document] was that it felt less like an analysis and more like an argument with little weighing or discussion of pro/cons - a good case study, but the conclusions didn’t really have me convinced in the recommended actions.
From a 6s mentor POV , I feel obligated to mention that I’m frequently annoyed at the difference between AM 6s and IM 6s in their formats, and I really would rather move toward standardization of formats rather than a half-league Swiss or half-league RR, but I also can understand that a perfect solution might not exist. -
You can easily expand a season without increasing the map pool by making every week a pick/ban. Alternatively, if it has to be one map per week there’s always coalplant, borneo, ramjam, badwater, and gravel pit, which are not all completely ideal but they do all have fans and have all been proven to be playable and are not absolute garbage in hl (I’m personally not a fan of badwater but it’s definitely not on par with barnblitz or 5cp).
I think a longer season would be fun, and as stated above, it would better ensure that the best teams make playoffs - fewer fluke wins for worse teams that can have a large effect on placement currently.
-
no divisions
put everyone in the thunderdome
2 5 month seasons
win/loss
-
@watterson this but you can drop in and drop out at will and have to play at least 2/3 of the matches to qual for playoffs, which everyone plays in divs and is elimination
-
why dont teams just not die lmao