Resource Simple Questions/Simple Answers: Random Battles Edition

[Piplz]

But rands has always used levels to even the power gap between the strongest and weakest Pokémon.

Wouldn't that be a good opportunity to revisit that philosophy, or test it out at the very least?

I don't know how familiar you guys are with twitch plays pokemon, but eventually instead of them running cartridge games it became kind of a casino where you could bet on pokemon battles, with the mons selected randomly. All of them were level 50, but the stronger mons were given absolutely pathetic EVs and IVs with terrible moves, while the weaker mons were given better stats and stronger base power moves. Some of the balancing had to be absolutely extreme in order to make it work (Palkia had Bubble and HP dragon with a power of 30), but nonetheless, it more or less worked and most mons had a generally similar power level.

Now I'm not saying this should be applicable to standard randbats (i think level balancing with more realistic movesets feels more natural and should be preferred over having otherwise good mons run flame wheel or whatever), but for a spinoff/kinda experimental ladder that can't afford to do level balancing, wouldn't it be a cool route to explore?

 

[Celever]

Wouldn't that be a good opportunity to revisit that philosophy, or test it out at the very least?

I don't know how familiar you guys are with twitch plays pokemon, but eventually instead of them running cartridge games it became kind of a casino where you could bet on pokemon battles, with the mons selected randomly. All of them were level 50, but the stronger mons were given absolutely pathetic EVs and IVs with terrible moves, while the weaker mons were given better stats and stronger base power moves. Some of the balancing had to be absolutely extreme in order to make it work (Palkia had Bubble and HP dragon with a power of 30), but nonetheless, it more or less worked and most mons had a generally similar power level.

Now I'm not saying this should be applicable to standard randbats (i think level balancing with more realistic movesets feels more natural and should be preferred over having otherwise good mons run flame wheel or whatever), but for a spinoff/kinda experimental ladder that can't afford to do level balancing, wouldn't it be a cool route to explore?

The reason this would be absolutely horrendous in a skill-based format is because coverage is not the same across the board, and compromising legality would be a huge red line for us. Even in your example here, Palkia can't learn Bubble -- never has done, probably never will. Its weakest Water-Type move is Chilling Water in Gen 9, and before that it's actually Hidden Power Water post-nerf or Water Pulse pre-nerf (or Whirlpool if you prefer). We can't start hacking moves onto Pokémon to try and make it balanced, which is what TPP is doing, because at that point we'd be making a hackmons rands format. It also doesn't just effect STAB; for example, most stuff that learns Psychic and isn't Psychic-Type has Psychic as its weakest Psychic-Type attack. It would then become a question of "if this is too strong it just loses Psychic-Type coverage altogether", or we make a format where nothing has any coverage. Neither one of those are remotely playable.

TPP is its own thing (which I sometimes take part in FTR) and the fun of it is predicting lopsided matchups correctly. It's basically a second-person experience. Actual Pokémon battling is a first person experience, and lopsided matchups suck.

As for using IVs and EVs instead of levels to adjust, we strongly don't want to do that because part of the nature of rands is that Pokémon act as they authentically act. In teambuilt everyone minmaxes everything, but in rands base stats matter quite a bit more. If a Pokémon has decent special defence, the rands IV and EV system lets it sponge a bit, where if Pokémon started having maxed out SpAtk then that Pokemon wouldn't sponge at all, especially if it itself wants maxed out SpAtk so can't invest in SpDef (e.g. Grumpig). The IV and EV system lets the natural talents of Pokémon shine in a way that teambuilt doesn't really do.

 

[Piplz]

Legality is a completely fair point and I agree than it would probably become hackmons more than anything else at a pace that'd go way faster than I initially thought (as an example I actually didn't even realise Palkia couldn't learn bubble lmao, I never questioned it and assumed it was one of those really bad relearner move since they also get stuff like Swift or w/e). Later on they(TPP) started needing to hack quite a lot of things in, with EVs beyond the standard limits as well.

With that said I do think the coverage issue (for a spin-off format like this at least) is a little exaggerated - it's only the really big outliers on both ends that would need that kind of fine tuning imo (shoutout to Blipbug with its incredible 180 BST and very deep movepool, gl ever balancing that) but I think that realistically most LC mons wouldn't need many restrictions to be useable and generally at the right power level.

For EVs and IVs I wholeheartedly agree with the fact that not using EVs lets base stats shine more and that can have its charm, but I don't see it at all for IVs. Isn't IV tuning extremely similar to level balancing at the end of the day? the only downside I see to it is that it adds yet another layer of knowledge

Lvl 82 84 SpA Mew Dark Pulse vs. Lvl 82 84 HP / 84 SpD Mew: 94-112 (31.5 - 37.5%) -- 0.1% chance to 3HKO after Leftovers recovery

Lvl 81 84 SpA Mew Dark Pulse vs. Lvl 82 29 SpD 84 HP / 84 SpD Mew: 94-112 (31.6 - 37.7%) -- 0.1% chance to 3HKO after Leftovers recovery

Lvl 82 29 SpA 84 SpA Mew Dark Pulse vs. Lvl 81 84 HP / 84 SpD Mew: 94-112 (31.8 - 37.9%) -- 0.3% chance to 3HKO after Leftovers recovery

Now of course in little cup trying to balance through IVs would be an exercise in futility as IVs might as well not exist at this level but still.

 

[senorlopez]

Why are trapping ability mons allowed to be leads? It just seems like potentially turn 1 game losing experience which could be easily remedied.

 

[A Cake Wearing A Hat]

Why are trapping ability mons allowed to be leads? It just seems like potentially turn 1 game losing experience which could be easily remedied.

Council voted to allow it because we do not purposefully nerf things that are too good in the vast majority of situations. Council votes cannot be overturned as a result of suggestions, feedback, pressuring, or external user input, so lead trappers will stay in the format indefinitely.

 

[SuperMemeBroz]

I really have no place to but this other than here but

Is there a reason why there isn’t a gens 3 or 4 battle factory considering there was a gens 6-8 battle factory format and there is literally a battle factory in the generation 3 & 4 games. I’ve noticed the interest of battle factory from a YouTube video I saw of a guy taking on round 16 of the emerald battle factory and it’s doing quite well. There are even sets you can simply import to make this a real format in the future. All I’m really asking is a BSS Battle Favtory format for generations 3 and 4

 

[A Cake Wearing A Hat]

I really have no place to but this other than here but

Is there a reason why there isn’t a gens 3 or 4 battle factory considering there was a gens 6-8 battle factory format and there is literally a battle factory in the generation 3 & 4 games. I’ve noticed the interest of battle factory from a YouTube video I saw of a guy taking on round 16 of the emerald battle factory and it’s doing quite well. There are even sets you can simply import to make this a real format in the future. All I’m really asking is a BSS Battle Favtory format for generations 3 and 4

Battle factory formats require a higher amount of manpower and quality control than regular Random Battles, as they typically require input from Smogon Contributions and Corrections Quality Control members in order to vet the sets for the format. This is hard enough for current gen, which is why it took so long for Gen 8 battle factory to be made. To ask this of old gen formats is even more of a task, as finding experts in old generation lower tiers that are also willing to participate in a randbats format and understand the difference between battle factory and their main format is a tall ask.

 

[SuperMemeBroz]

Battle factory formats require a higher amount of manpower and quality control than regular Random Battles, as they typically require input from Smogon Contributions and Corrections Quality Control members in order to vet the sets for the format. This is hard enough for current gen, which is why it took so long for Gen 8 battle factory to be made. To ask this of old gen formats is even more of a task, as finding experts in old generation lower tiers that are also willing to participate in a randbats format and understand the difference between battle factory and their main format is a tall ask.

Well because factory sets already do exist, wouldn’t it been easier than current gen due to less Pokémon and the affirmation of battle factory sets existing. I doubt most old gen randbats players would be willing to understand the differences of factory and randbats but that’s when we’d have to get other people that already know the fundamentals of factory to make it a less jarring task.

 

[kaboissonneault]

I myself have some interest in this, being in on this Emerald Battle Factory reemergence. If manpower is what's needed, I can always contribute. I am a game programmer by trade, so I picked up on the data and the procedures involved in the Factory pretty quickly, using pret's pokeemerald decompilation. I have not much expertise with web technologies (ex: Typescript) however, and no familiarity with the Showdown codebase. I am also not entirely familiar with all the obscure formats Showdown might have that could already implement some of the features we need.

I feel like a Gen 3 Factory format would require a "get 6, pick 3" selection before the battle, which does not seem to be present in the Gen 8 Factory on Showdown. Even better if you can swap with your opponent after victory, keeping the team as long as you can continue your streak. Leaderboards would be by streak and swap count.

Simply having a randbats with the Factory sets would not be terribly unique as a format. The sets are not really the iconic part, though the Double Team spam does stand out. I am of course familiar with the sort of effort involved in this, and this is why I do not really expect more, unless someone knows something that can be reused to emulate these features.

 

[A Cake Wearing A Hat]

I myself have some interest in this, being in on this Emerald Battle Factory reemergence. If manpower is what's needed, I can always contribute. I am a game programmer by trade, so I picked up on the data and the procedures involved in the Factory pretty quickly, using pret's pokeemerald decompilation. I have not much expertise with web technologies (ex: Typescript) however, and no familiarity with the Showdown codebase. I am also not entirely familiar with all the obscure formats Showdown might have that could already implement some of the features we need.

I feel like a Gen 3 Factory format would require a "get 6, pick 3" selection before the battle, which does not seem to be present in the Gen 8 Factory on Showdown. Even better if you can swap with your opponent after victory, keeping the team as long as you can continue your streak. Leaderboards would be by streak and swap count.

Simply having a randbats with the Factory sets would not be terribly unique as a format. The sets are not really the iconic part, though the Double Team spam does stand out. I am of course familiar with the sort of effort involved in this, and this is why I do not really expect more, unless someone knows something that can be reused to emulate these features.

Well because factory sets already do exist, wouldn’t it been easier than current gen due to less Pokémon and the affirmation of battle factory sets existing. I doubt most old gen randbats players would be willing to understand the differences of factory and randbats but that’s when we’d have to get other people that already know the fundamentals of factory to make it a less jarring task.

Battle Factory, as it's called on the sim, is not "battle factory" as exists in the Battle Frontier. Battle factory, as a randomized showdown format, is "a team of competitively viable pokemon that are specifically EVd and built to fit an actual smogon tier". You fight in OU, UU, Ubers, etc. Completely nothing like what you're talking about.

 

[SuperMemeBroz]

Battle Factory, as it's called on the sim, is not "battle factory" as exists in the Battle Frontier. Battle factory, as a randomized showdown format, is "a team of competitively viable pokemon that are specifically EVd and built to fit an actual smogon tier". You fight in OU, UU, Ubers, etc. Completely nothing like what you're talking about.

Yeah I’m aware, what I want to have is the actual battle factory part. A randbats like 3v3 format except you have battle factory sets. I couldn’t care less about trying to do have 6 and pick 3 fundamentals, I only really care about the actual battling part

 

[mcooper527666]

Question, how come there isn’t a national dex random battles?

 

[A Cake Wearing A Hat]

Question, how come there isn’t a national dex random battles?

requires monumental work and balancing far surpassing that of any other random format.

 

[Celever]

Question, how come there isn’t a national dex random battles?

A National Dex random battles format doesn’t exist for several reasons:
1) Lack of Demand — NatDex rands is a semi-common suggestion we receive, about as commonly as LC rands (perhaps slightly more). We do not know how this would translate to player numbers, and we already make a lot of formats. It’s conceivable that NatDex rands wouldn’t hit the activity quota to stay, mainly because…

2) The Format Would Probably Suck — The worst part of rands which everyone hates is a bad matchup, especially when your team is actually quite good, but your opponent just has that one Pokémon with the perfect attributes to roll over your team. When you have every Pokémon and mechanic in the game, including discontinued ones, the chances of this happening actually increase significantly. Part of the reason for this is movesets cannot be optimised, because there are so many Pokémon each Pokémon might go up against. Pokémon with Fire- and Fighting-Type coverage whose STABs are resisted by Steel-Type would probably have to run both of them in NatDex, for example, whereas in something like Gen 9 there is usually an identifiable optimal one we can glean by looking at the specific Steel-Type Pokémon in the format.

3) It Would Be Stupidly Hard To Make, Or Oversimplified — Our other formats have a limit of 10 roles. 1 of these is used on the generational gimmick (Tera, Z-Moves) and in NatDex that becomes 2. This would mean losing another role, further meaning some Pokémon would have bad item generation rules just to accommodate both gimmicks. On top of that, a format that needs to check for 3 rejections between those 2 gimmicks and Megas would result in some pretty complicated coding. This isn’t to mention the amount of work that would go into designing and modifying sets, which again we’re talking about every Pokémon in any game here, and many Pokémon would be left in “works on paper” mode in relative perpetuity because designers wouldn’t get the requisite playtesting with them, because appearance rates for each individual mon will be low.

By the way, NatDex rands has never actually been a “no”. It’s mainly been a “if we can find a way to make it work and somewhat fun and randomly have time to full send into that” and those conditions are unlikely to ever be met to be honest. We’re doing full old gen revamps still, DLC2 is on the horizon, we have another revamp on the way, potentially an additional current gen rands format that isn’t NatDex next on the agenda (no promises on that though) mixed in with running tournament circuits and all sorts of other stuff. Towards the end of Gen 9 we have factory formats to make too and they take priority too. Honestly I’d foresee this being more feasible in Gen 10 because we wont be doing full oldgen revamps during it lol.

 

[woodgoldtwo]

I remember reading somewhere that FFA randbats was going to get its own sets after DLC 2, is that true?

 

[A Cake Wearing A Hat]

I remember reading somewhere that FFA randbats was going to get its own sets after DLC 2, is that true?

FFA will be revamped whenever we are done revamping it, but due to the manpower required it's taking a while.

 

[Toxin Boost]

Azumarill's only gen 4 set is illegal due to the combo of belly drum and aqua jet, is this know about and if so why is it still there?

 

[Celever]

Azumarill's only gen 4 set is illegal due to the combo of belly drum and aqua jet, is this know about and if so why is it still there?

We don’t care about ensuring a Pokémon is completely legal, mainly because we think it would make the format less fun and competitive, but also because caring about legality means more than just move compatibility, and includes things like forcing shinies and specific natures and IVs that the rands code is really ill-equipped to enforce. It would just add a bunch of dumb memorisation and game knowledge that no one really wants to add.

 

[lemonademaker]

Is there a reason why basically every mon with guts in randbats gen 5 uses toxic orb over flame orb? Is it for balancing, or is this just an oversight.

 

[RSB]

Is there a reason why basically every mon with guts in randbats gen 5 uses toxic orb over flame orb? Is it for balancing, or is this just an oversight.

In gen 5 burn still does 12.5% damage per turn compared to toxic being 6.25% and increasing per turn. Over 2 turns which is the average about of time a guts pokemon would usually stay in for, the amount of damage taken from toxic doesn't amount to as much as you would take from burn. After 3 turns and onwards it evens out and toxic does do more but as I said before usually guts pokemon won't stay in for that long without either fainting, wanting to switch out, or simply doing enough work to win the game.

 

[lemonademaker]

In gen 5 burn still does 12.5% damage per turn compared to toxic being 6.25% and increasing per turn. Over 2 turns which is the average about of time a guts pokemon would usually stay in for, the amount of damage taken from toxic doesn't amount to as much as you would take from burn. After 3 turns and onwards it evens out and toxic does do more but as I said before usually guts pokemon won't stay in for that long without either fainting, wanting to switch out, or simply doing enough work to win the game.

Thanks. But then why do some guts mons have set up (like ursaring). Is it just for them to have a scenario where they can win the game like you said?

 

[RSB]

Thanks. But then why do some guts mons have set up (like ursaring). Is it just for them to have a scenario where they can win the game like you said?

Yes, something like ursaring is so bulky and powerful that it can usually sweep with quick feet and one swords dance up compared to most other guts users that are either too frail, slow, or simply lack the set up moves.

 

[Gamepro663]

Has anyone ever started a random battle and gotten a shiny Pokemon?

 

[Celever]

Has anyone ever started a random battle and gotten a shiny Pokemon?

yes, they appear at a 1/1024 rate. So in theory you should get a shiny every 1024th Pokémon you roll, but probabilities obviously aren't that clean.

 

[Gamepro663]

yes, they appear at a 1/1024 rate. So in theory you should get a shiny every 1024th Pokémon you roll, but probabilities obviously aren't that clean.

Kinda funny that the shiny I got was Giratina! XD