Let me just begin by clarifying that I don’t disagree with this post in principle in the slightest. Having unique ways to play the game that challenge players to outsmart their opponents, especially as opposed to just winning DM fights, is an excellent thing.
That being said, I think there are a number of logistical problems with actually trying to add new gamemodes to our existing formats of comp tf2 that it is not sustainable in our current community with its size and focus. Note that, as a baseline to compare against, it is hard enough to add new maps for existing gamemodes to any of the formats - they face stiff resistance, it is near impossible to get serious playtesting done short of actually adding the map to an EXP Cup, one of which happens at best twice a year, and mapmakers may be apprehensive to create new maps for a community that just tends to stick to its preconceived notions about them anyway. Note that I believe that, volume of maps to be tested permitting, frequent EXP cups, if not to actually determine whether a map is suitable for inclusion in a regular season to give the mapmaker feedback on how to improve it so as to make it more fun or playable, are an ecellennt idea and if they were implemented, would make this idea a whole lot more feasible.
For gamemodes like KoTH, 5CP, and Payload, there are clear mapping standards for competitively viable maps - for example, for KoTH, there is the clear area-connector-area-connector-area-midpoint layout that most maps tend to follow, and for Payload there’s the unwritten rule that for each point there tends to be only one strong hold - such standards do not exist for other gamemodes and would require mapmakers already familiar with how comp tf2 plays (who are already few and far between) to improvise and theorycraft the map design such that it would not follow a standard that works in pubs but cannot work in comp.
Take for example the standard layout of Capture the Flag maps. Setting aside the outdated layouts of 2fort or Sawmill that make you literally go into a basement with two chokepoints to retrieve the intel, even on modern and well-designed maps like Landfall, the intel is right next to its team’s spawn, meaning that against a competent team, especially one with classes like Engineer, for example, retrieving the intel with just a dry push would be near impossible, and you would need Uber AND significant other advantages to retrieve the intel - other advantages like players up to get close enough to the intel to take it in a clean Uber push, and Uber to actually reach the intel (god forbid the other team has Uber too).
This is not to say that CTF is competitively unviable. It might very well be, but there might very well be an ideal CTF map design that maximizes both fun and competitive rigour for players in one of our existing competitive formats. The problem is that this ideal map design is so far unknown, meaning that mapmakers would have to play with simple trial and error to figure out what that map design is, which, given the constraints that already exist for competitive map designers and the mental block many people would have to seriously giving feedback for a CTF map and considering its merits (and knowing what feedback to give!) makes it a formidable task for any mapmaker that requires significant personal time and effort investment, and not necessarily one that will pay off.
Then there is the problem that some gamemodes have fundamental flaws that make it impossible to make a map that is competitively rigorous. For example, Mannpower has random power-up spawns, making it possible for one team to beat another through pure luck of the draw on what power-ups spawned on their side. Obviously, this is bad. Then, Player Destruction (the format you mentioned "What about a mode where it’s just a kill fest where people collect points and need to score them? ") has a gamebreaking flaw in that the player carrying the most points on each team has the self- and area-healing and ammo restore power at the rate of e Level 1 Dispenser, which makes it a whole lot harder for a team to kill any players who simply stick to their “team leader” as I believe the mechanic calls that player.
This is not to say that these barriers are insurmountable. For sure, my post is a bit disingenuous - you’re making an idealistic appeal, but I’m responding by citing realistic barriers, and not ones that cannot be overcome with enough willpower. My question is simply if that willpower actually exists. I believe that with enough effort, enough changes to the community to allow for, for example, more frequent EXP cups to give mapmakers feedback, and enough accommodation by existing players who encourage forward thinking mapmakers, anything you said in this post is possible. Am I for it? Also yes, but I’d definitely not be the one spearheading it just because of how much effort it is. What do you say about these challenges?