Oh! Oh! And another thing! You totally didn't account for the NU tier! You just lumped all the NUs together into the UU pile. I definitely feel Tangela and friends could use the boost!
http://rby2k10.proboards.com/thread/273/official-tier-thread
I got a Chansey with Thunder Wave, Counter, Soft-Boiled, and Reflect. Counter should not count as the sole attacking move.
if peeps still have suggestions for improvements, no reason not to~Are we still updating this?
I really like this idea and i think its how randbat levels should be chosen, but idk if raw W/L is the way to go about it. Legendaries/Broken mons ending up on awful teams or being given to less competant players could end up with completely broken levels of Ubers, or shit mons ending up with really good players or on broken teams could results in low level PU pokemon.Hello, folks. I've been playing this format for quite a while and the rough pokemon levels irked me; the tiers do not feel balanced. I've created a variation on the format in which pokemon levels are adjusted automatically based on their performance. Specifically for games in this new format, records are stored for each pokemon: how many times it's been on a winning team, and how many times it's been on a losing team. All pokemon start at level 50. Pokemon which win at least 51% of the time have their levels adjusted down, and pokemon which win 49% of the time or less have their levels adjusted up. Eventually, each pokemon arrives at a level which is perfectly balanced, so it does not contribute more or less to a team on average than any other; each pokemon's win rate settles at 50% given enough matches.
With Panpawn's help, I've implemented this format on the Gold server. It's called [Gen 1] Random Auto Level Adjusted. The Gold server does not have an especially active player base, and the match records there are basically just based on me playing the new format with my buddy. As a result, the pokemon levels are still pretty rough. However, having this format on the main server and getting some adoption on it would adjust the pokemon levels much more speedily. I would love to get this on the main server, if people think it's a good idea; so far, the only response I've gotten has been positive.
My pull request to the Gold server, where I introduced the new format, can be found here. Feel free to peruse the code. I'd be happy to answer any questions. https://github.com/panpawn/Gold-Server/pull/220
I really like this idea and i think its how randbat levels should be chosen, but idk if raw W/L is the way to go about it. Legendaries/Broken mons ending up on awful teams or being given to less competant players could end up with completely broken levels of Ubers, or shit mons ending up with really good players or on broken teams could results in low level PU pokemon.
Would there be a way to do it by something like KOs or damage dealt instead?
Yeah very good points, win loss is probably the best way to fairly measure support and defensive pokemonI think any issue of "bad mon with good player" or "good mon with bad player" would be drowned out over time as noise; on average, the bad mon is still bad and the good mon is still good. The way I've implemented this system is specifically designed to drown out noise; it doesn't matter if the win/loss records are not right initially. As long as people keep playing, the pokemons' true power levels will emerge in the statistics. I'm not a fan of the idea of using KOs or damage dealt since that undervalues pokemon that have strong support abilities (inflicting statuses, etc.). I think win/loss ratio is the purest measure one could use of a pokemon's utility.