Though a common response on here is “but its ment to be crazy and have no balance”. No. All modes deserve balance, or they will severly lack replayability.
For example, why would I play soldier 76 in deathmatch when I could play roadhog and win with little effort?
My suggestion for balancing for that situation is to lower the HP of tanks so that tanks aren’t the only realistic way to win.
Other modes like CTF might need specific heroes to have values changed.
So yeah, make arcade more fun, let all roles be viable.
This is the kind of topic where you really need to have developer insights.
How many people play Arcade? What’s that distribution split like between modes? Do certain Arcade modes foster greater player attention from the normal game?
I would extremely doubt that Arcade has enough players to really quantify a lot of the kind of in-depth balancing the development team would want to do if that was something they were interested in.
Probably not. Give a mouse a cookie, yeah? ‘Ah, we’ve fixed Hog.’ “Yeah, so fix all the tanks.” ‘Ah, we’ve fixed all the tanks.’ Suddenly you’re at the point where they rework Symmetra just for Arcade modes.
It’s a lot of work and a lot of time that is clearly better spent catering to the rest of the game. [Insert a commonly cited and unsolved balance or design issue to prove my point here]
Back when they were introducing competitive open queue as a temporary game mode in the arcade, Jeff Kaplan posted some stats about the daily % of active hours in various game modes for a single day. Ignoring the Open Queue modes (i.e. not quickplay classic and open queue comp), arcade modes and workshop made up around 20-25% of the player time for that particular day, depending on platform and region:
This is the slippery slope fallacy. They’ve done this without any issue for Total Mayhem multiple times and nobody has come back begging for specific changes or larger refactors to heroes in other modes as a result.