Any way to make a randomly drawn FR/LG tournament "fair(er)"

Here's the deal: some guys I know have been doing random draws of the original 150 pokemon and battling with a team of 6. I think they've been allowing one "skip" if they pull something they don't want.

As you can imagine this has lead to some wildly imbalanced matches. One guy ended up with 5 OU-tier pokemon and another guy had all pre-evolutions.

They've invited me to join the next "tournament", but I want to see if I can improve their process a bit to make it more "fair". Here's what I'm thinking.


1. No EV training or breeding for IVs / hidden powers (actually, no breeding at all probably makes this the easiest)

2. Mewtwo banned (should I consider banning anything else? Snorlax maybe?)

3. Stratify the pokemon drawing somehow (this is where I need help)

4. Any other ideas you might have to make this a bit more fair


Clearly this is a little different than a normal "competitive" discussion, but I thought you guys might have some ideas. Specifically, how might I break the pokemon up in to different pools so that there's less variance between the strength of any given team?

Any other thoughts are appreciated!
 
Maybe consider assigning point values to each pokemon, like most draft leagues do? Make an algorithm to generate a team given, say, 20 points total. Mewtwo/Snorlax are worth 10 or so, so drawing a mewtwo means that the rest of the team must be pokemon with only 2 points, ie pidgeotto. The values are subject to change, especially if you want to derestrict the 6 pokemon rule so you could have both a mewtwo and a snorlax but nothing else. This comes with the drawback of losing some team combinations, meaning it's impossible to get either a stupid broken team or a crappy team, but that kinda just comes with the deal.

The other thing you could do is just keep re-generating teams until you get equally good ones based on like bst or tier or whatever you want. It's not clear whether you keep your team secret from the other player or not though, so that might make it a little tricky.

breeding and ivs are up to your own discretion, but I would argue against banning evs. EV training isn't luck-based like the other ones, and they can add some strategy- maybe you ev your charizard to survive a +1 alakazam thunder punch and activate salac berry and blaze, just for an extreme example.

edit: maybe also consider changing pokemon's levels to be different based on their power a la randbats?
 
Last edited:

mkizzy

formerly kenny
It'd be kinda cool if you did an auction. Give yourselves the same budget and allow yourself to bid on the pokemon as their drawn. Might take a few attempts to get the budgets balanced.

Or the opposite, like the above poster said give each mon a predetermined value and yourselves a budget.
 
Maybe consider assigning point values to each pokemon, like most draft leagues do? Make an algorithm to generate a team given, say, 20 points total. Mewtwo/Snorlax are worth 10 or so, so drawing a mewtwo means that the rest of the team must be pokemon with only 2 points, ie pidgeotto. The values are subject to change, especially if you want to derestrict the 6 pokemon rule so you could have both a mewtwo and a snorlax but nothing else. This comes with the drawback of losing some team combinations, meaning it's impossible to get either a stupid broken team or a crappy team, but that kinda just comes with the deal.

The other thing you could do is just keep re-generating teams until you get equally good ones based on like bst or tier or whatever you want. It's not clear whether you keep your team secret from the other player or not though, so that might make it a little tricky.

breeding and ivs are up to your own discretion, but I would argue against banning evs. EV training isn't luck-based like the other ones, and they can add some strategy- maybe you ev your charizard to survive a +1 alakazam thunder punch and activate salac berry and blaze, just for an extreme example.

edit: maybe also consider changing pokemon's levels to be different based on their power a la randbats?

Cool ideas!

I think they've been enjoying drawing the pokemon out of a hat (so to speak) together. I think a tier system might work better than points - i.e. try to divide the pokemon in to 3 - 6 tiers and then randomly draw from those tiers a set number of times. I don't really know how to handle the pre-evolution pokemon though... they'd always be bottom tier (for the most part)

As for the EV training, the only reason I was considering banning that is that some people don't know what that is or how to do it and others do. Maybe teaching everyone is the best way
 
edit: maybe also consider changing pokemon's levels to be different based on their power a la randbats?
This seems like the perfect way to go to balance out the pre-evolutions. Do you happen to know if there's a randbats level table somewhere for gen 3?


EDIT: I found this for a more current version of randbats. I wonder if I could knock 10 levels off this curve for everything and then just stick the pokemon in gen 3 tiers

LC: 88,
LC Uber: 86,
NFE: 84,
PU: 83,
PUBL: 82,
NU: 81,
NUBL: 80,
RU: 79,
RUBL: 78,
UU: 77,
UUBL: 76,
OU: 75,
Uber: 73,
AG: 71,
 
From the way you mention IVs/EV training, it sounds like you're not doing this on the simulator, yeah? And only the Pokemon available in the Kanto Pokedex. Like RS200, except FRLG150. I'd imagine the normal 386 tiers wouldn't be all that relevant here. I don't necessarily think every format is improved by applying Smogon tiers.

Leveling to a certain limit could add complications, especially as it seems some participants aren't familiar with all the mechanics. Though, maybe you're using emulator tools to speed that up?

I think it'd be worth it to consider using Base Stat Totals to stratify the Pokemon. Like, everyone picks a Poke from the 525-600 BST bin, everyone picks a Poke from the 495-515 BST bin, etc.. You could make six even bins of 25 Pokemon each, or engineer them so everyone gets more good Pokemon and fewer complete duds. I'd want to ban Mewtwo too, but maybe that's not in the spirit of the format.

Obviously BST isn't a great metric, but I think it's a bit more relevant than 386 tiering. It's based on in-game attributes, and so might seem a little less alien than "here, some nerds on the internet think Dodrio is bad into Skarm, so it goes in bin2". It looks like all you need is some way to bin the Pokemon, and tiers would be fine for that. But I thought this might be worth thinking about.
 
Last edited:
From the way you mention IVs/EV training, it sounds like you're not doing this on the simulator, yeah? And only the Pokemon available in the Kanto Pokedex. Like RS200, except FRLG150. I'd imagine the normal 386 tiers wouldn't be all that relevant here. I don't necessarily think every format is improved by applying Smogon tiers.

Leveling to a certain limit could add complications, especially as it seems some participants aren't familiar with all the mechanics. Though, maybe you're using emulator tools to speed that up?

I think it'd be worth it to consider using Base Stat Totals to stratify the Pokemon. Like, everyone picks a Poke from the 525-600 BST bin, everyone picks a Poke from the 495-515 BST bin, etc.. You could make six even bins of 25 Pokemon each, or engineer them so everyone gets more good Pokemon and fewer complete duds. I'd want to ban Mewtwo too, but maybe that's not in the spirit of the format.

Obviously BST isn't a great metric, but I think it's a bit more relevant than 386 tiering. It's based on in-game attributes, and so might seem a little less alien than "here, some nerds on the internet think Dodrio is bad into Skarm, so it goes in bin2". It looks like all you need is some way to bin the Pokemon, and tiers would be fine for that. But I thought this might be worth thinking about.
All good points.

I've gone ahead and put the pokemon in to the tiers just to see how it might look. This seems pretty far off, especially looking at base stat totals

https://docs.google.com/spreadsheets/d/1ORSuTddzFxpxjqmJWhYt7J_W51Rm9WUKQ9cigphiVos/edit?usp=sharing
 
Last edited:
Chansey would certainly be OU in this metagame. The only reason it's UUBL here is that Blissey exists in Gen 3, but if Blissey didn't exist, Chansey would be a metagame staple on its own. Honestly it might be on par with Snorlax or better, because if you don't have a strong physical attacker, you just can't kill Chansey.

Other Pokemon I imagine would jump up to OU are Dodrio and possibly Machamp. There are no longer Steel types everywhere to wall Dodrio, so it'd be very strong. It's slightly slower than Tauros, but actually hits harder. Machamp is the best fighting type and Cross Chop/Rock Slide is great coverage, but it's slow and Psychic-weak, so it won't be completely dominant.
 
Hi, I readed first couple of posts and the idea of making a draft-type system as someone suggested would probably best for you. When I am bored, sometimes I usually do a similar thing with my friends on PS. What we do is picking 8-9 random pokémons with the !randpoke command and then each of us will pick 6 of those and make a team with them. Instead of just being allowed a skip, you have a couple extra too and it makes building and the short prep fase funnier as you have to consider multiple mons. We do this with all gens though, and we just remove Ubers and we use only Fully Evolved mons, so battles are usually pretty equal and funny
 
Hi, I readed first couple of posts and the idea of making a draft-type system as someone suggested would probably best for you. When I am bored, sometimes I usually do a similar thing with my friends on PS. What we do is picking 8-9 random pokémons with the !randpoke command and then each of us will pick 6 of those and make a team with them. Instead of just being allowed a skip, you have a couple extra too and it makes building and the short prep fase funnier as you have to consider multiple mons. We do this with all gens though, and we just remove Ubers and we use only Fully Evolved mons, so battles are usually pretty equal and funny
If they're playing on cartridge/emulators on FRLG, they could also add post-game Johto mons, although I guess maybe keeping it to the original 150 was part of the point. Off the top of my head you get the legendary dogs, Tyranitar, Skarmory, as well as a few other good mons like Heracross/Misdreavus/Qwilfish and the Johto evos (Blissey, Kingdra, Scizor, Steelix, Crobat, P2 etc). A FRLG-only metagame might be a cool idea analogous to RSE-200.
 

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top