http://replay.pokemonshowdown.com/gen7randombattle-696959538
This was a completely unfair match where I just can't see how this could have possibly been fun for my opponent (and it wasn't particularly fun for me, given the knowledge of how rage-inducing the match would have been had I been on the opposite side - this fact simply lessened any enjoyment I could have gotten out of the match)
My opponent had 1 great mon (specs Exploud), one good mon (Excadrill), some situationally good water types (Mantine/Vaporeon) and two utterly useless sacks of you know (Kricketune without webs and Unown), whereas I had Aurora Veil Ninetales, Battle Bond Greninja, AV Kyurem (the perfect switch-in to specs Exploud under Veil - it took 21%!), SD Toxicroak, Xurkitree!! and my last was Pangoro which had some utility with Knock Off. I would go as far as to say I had 5 of the 6 best mons in this match.
We already don't balance randbats to be truly random -- it isn't a truly random experience. For example, for reasons of Pokemon strength, we do not allow most NFE Pokemon to be used. I've never received a Wartortle, for example, and that's fine. However, given the ability to record usage stats, I think it would be best if there were just one more additional countermeasure coded into random battles -- the concept of "Extremely Inbalanced Team Reshuffle".
I propose the following:
1) Each Pokemon receives a score based on their Win % on the high randbats ladder. The randbats ladder is large so you can make this a reasonable number that covers the middle game as well.
2) Many Pokemon perform differently based on their set - Geoxern should be scored separately from AV Xern, for example. Similarly, entire games** can and are decided based on who either has hazards or hazard-removal - for this reason, Mon A with Rocks and Mon A without Rocks should probably be scored separately.
3) Add each Pokemon's score on each team (adjusting for hazards) - if the scores are vastly different, reroll the teams.
How difficult would something like this be to implement? It's not asking for perfect balance, it is merely asking for a reroll in the case of extreme inbalance, such as in the replay shown above.
**For randbats on high ladder, I would estimate, 80% of matches come down to the following 4 factors (about equally important although we can't really control for #4)
1) who has hazards (and, in some cases, hazard removal - depending on whether or not the other person has hazards)
2) who has the fastest mons
3) are there any absolutely broken mons in this format that are on either team? (Sigilyph being a famous one that immediately ends the game just by appearing if you have nothing for it)
4) hax
(4.5 - Leftovers win games. Life Orb loses games. This is almost as unbelievably annoying and frustrating as hax.)
Thank you for reading my post.