omega nerd essay coming in. headings underlined if ur only interested in a bit.
First thing: comp player's n1 priority is winning--while we care about fun and liking the maps, nothing will trump trying to win when match time comes. So when people ban maps, they ban the maps that either 1) their team is bad at playing or 2) the other team is good at playing.
This is why the map distribution works out the way it does. It's not just that people like process, snake, and gully more; it's that they feel most comfortable on these maps because they are well made. These maps are consistent and flow well, so teams can focus on higher level play and not be held back by gimmicks. Teams know that they cannot be cheesed out of a win on these maps--the better team always wins here.
on granary and badlands:
Granary and badlands are gimmicks--both maps have hugely exploitable layouts (granary slows to a snails pace so only uber advantage is relevent/badlands last is hell to push into and out of). So what ends up happening is people pick these maps not because they know the map better, but because they can exploit these problems better than the other team. As in, these maps are only picked when one team feels so much worse than the other that they feel like they need to cheese out the win.
Notably, both granary and badlands are hugely dependent on ubers, while also making it really fucking hard to sack for medics. So if a team wants to safely hold back and wait for uber, there isn't much you can do to stop them. These maps have always been hard to sack for medic on, but in recent years (MEDIC-SCOUT SPEED WHY) it has just become unreasonable. Not only do these maps make the game slow and boring, but they encourage 1-dimensional play: stalemate, sack for medic, repeat until 1 of the medics fuck up.
These maps are good at producing upsets, but not for the right reasons. You can cheese these maps really hard by playing defensively, to the point where the decidedly worse team can sometimes come out on top. Similar to how any soldier can beat you on spire if they just sit on the fence.
on new maps:
New maps are in an interesting position: when they come out, everyone is bad at them. So you can either ban them and focus on the maps you know, or you can try to learn the map before everyone else and leverage that map knowledge for some wins against unprepared teams.
So why didn't anyone try to pull ahead on cardinal (well some teams have, like King's Krew, but in general)? Because the map isn't the same quality as the core 6s maps. That is to say, with so many branching chokes and wide areas, the map still feels cheese-ridden with how players can potentially wrap/flank/backcap. Nobody would ever feel comfortable that they would certainly win going into that map.
Propaganda has reason to go over differently. The map plays much cleaner, like the rest of the core 6s maps. That should be enough encourage teams to play it--to get ahead of other teams on a map where they can win consistently and not get gimmicked.
At the end of the day, every map in rotation should eventually be held to the same standard as process, snake, and gully. They should feel consistent and non-gimmicky, and teams would pick between them based on more nuanced concepts: which maps favour soldiers over scouts, holds over team-fights, etc... It's not that worse teams should never be able to upset better teams, but that they must get these upsets by carefully picking maps which benefit the few, slight advantages they have. New maps need to be added to reach this ideal state, so we shouldn't be shooting them down right off the bat.
on viaduct:
It's not a bad map, but I really don't consider it very good either. And that's not because people just don't know how to play it--it's the opposite. If you know how to play viaduct, it becomes pretty simple: team-fight for point, losing team respawns, losing team 4 man sacks to break even ubers/uber disadvantage, then losing team refights with uber ad so that they become the winning team, then loop until time hits 0.
The only compelling parts of viaduct are the team-fights; but frequent team-fights occur because it's KOTH, not because it's viaduct. I'd be much more interested in developing KOTH maps which did not give the winning team complete positional advantage, making it viable for the losing team to dry-push instead of sacking their whole team for med.
A KOTH map which encourages dry-pushing by evening out positional advantage, and therefore encourages more team-fights, would be far superior to viaduct imo. And having more than 1 KOTH in rotation would be great too. That said, KOTH should stay in the rotation as way to reward teams with coordinated team-fighting, and as a way to encourage teams to work on coordinating team-fights (which is arguably the most complex and underutilized aspect of tf2 imo). Viaduct just might not be the map to keep KOTH relevant tho.
nerd essay end me