Process Guide Workshop for Create-A-Pokemon.

[X-Act]

Egg moves shouldn't be all sucky. Quoting myself, I said that "the most important moves in a Pokemon's movepool are usually TMs, and that the level-up list and several of the egg moves usually suck".

In other words, the level-up list usually contains only 2 to 3 worthwhile moves at most, but up to half, or even more, of the egg moves list could contain usable moves. The bane of the good moves are usually in the TM list, though, and this holds for practically every Pokemon that I can think of.

 

[X-Act]

So, shall I write a guide on how to create a movepool for a Pokemon?

 

[Gothic Togekiss]

So, shall I write a guide on how to create a movepool for a Pokemon?

I would yes as I want to know the ways of making a decent movepool too.

 

[eric the espeon]

This is a reply to DJDs post in X-Acts guide:

How will the TL chose what moves are classed as competitive? some things are usable but not truly competitive e.g. Dragon Rush Garchomp some would be competive but not on that pokemon e.g. EQ Blissey, Focus Blast Luke. and some are usable on gimmiky sets e.g. Mean Look on Gengar.

Should it be that the voters of CaP get to see the full movepools before the poll and say if something should be removed?

This is my idea of how it could work:
This is a WIP

1. What TM/HMs it gets.
Let the TL (or someone/several people appointed by the TL) write up a proposed list of what TM/HMs it should get, let people comment on it for a day or so, and have the TL (or appointed person/people) change the TM/HM learn list according to peoples suggestions (have a look in the currant movepool thread to see how I have done it).

The most disputed moves are polled by Bold text voting in the same way the Punches and W-o-W were on Revenankh, with each person having a Yes or No vote for each move.

2. What Lv. up moves it gets.
People discuss the possible competitive moves, the TL draws up a list of the most popular 10, Bold text vote (with each person being allowed 3 YES votes and 2 NO votes) to find the most popular 2-4, flavour moves will also be discussed, they will then be organized, by the TL or someone they select (into a Lv. up learn list).

3. Egg Moves
People discuss possible moves (competitive and non-competitive), the TL draws up a list (of about 20) of those with support and they are all voted on by bold text voting (so you say yes or no for each move).

 

[DougJustDoug]

I've left this hanging out here for a while. I've heard a few suggestions, and X-Act has already started the guide for the Complete Movepool phase. I plan to add this into the Process Guide sticky. I'm considering changing the structure of the guide a little bit, to allow details to be added to certain sections. This will be the detail for the movepool section.

The current movepool thread is a total mess. The Movepool section of the project needs a total process revamp. I'm glad Time Mage brought this up.

Based on the suggestions in the past few posts, I'd like to propose the following process for determining the movepool:

Movepool discussion and voting will be divided into three phases

Competitive Attack Moves
Competitive Defense and Support Moves
Complete Movepool​

The TL posts an initial Attack movepool in the OP of a discussion thread. People post regarding necessary additions and removals. No full moveset posts are allowed.

At the conclusion of the Attack discussion, the TL will open an Attack voting poll to vote on controversial changes to the Attack movepool.

At the same time the Attack voting thread is opened, the Defense and Support movepool discussion thread will be opened.

At the conclusion of the Defense and Support movepool discussion, the TL opens a Defense and Support voting poll to vote on controversial changes to the Defense and Support movepool.

Based on participation in the previous movepool threads, the TL will select several members to present a complete movepool submission. Since all the competitive moves will have been decided, the movepool submissions will merely add "flavor" moves to round out the movepool.

A guide should be written to describe the characteristics of an appropriate movepool. X-Act's guide is an excellent start. If anyone attempts to include a competitive move, not agreed in the previous polls, it will be the TL's responsibility to see that the submitter removes them from the full movepool submission.

The movepool submissions will be voted on by the community. No alterations, edits, or conditional votes will be accepted. By the time the final movepool is polled, all competitive aspects should have been decided. Therefore, the final movepool is really a vote of individual preference on what best fits the flavor of the pokemon.​

 

[DougJustDoug]

During this latest pokemon creation we have discussed several changes to the Process Calendar. I want to revise the process sticky to reflect all these changes.

Process Order of Events
(Multiple events on one "bullet", means the events kick off simultaneously)

[LIST]
[*]Concept Submissions
Topic Leader Nominations


[*]Concept Poll
Topic Leader Selection


[*]Main Type Poll


[*]Secondary Type Poll
Stat Spread Submissions
Art Submissions


[*]Style Bias (Offensive/Defensive/Mixed)


[*]Build Bias (Physical/Special/Mixed)


[*]Stat Rating Poll


[*]Stat Spread Poll


[*]Art Poll


[*]Ability Discussion
Sprite Submissions


[*]Ability Poll


[*]Attack Moves Discussion


[*]Attack Moves Poll
Defense and Support Moves Discussion


[*]Defense and Support Moves Poll


[*]Complete Movepool Poll
Name Submissions


[*]Name Poll
Pokedex Entry Submissions


[*]Sprite Poll
Pokedex Poll


[*]Server Implementation
Finalize Analysis
Misc (Pre-Evos, Height, Weight, etc.)


[*]Playtesting
[/LIST]

Key changes

• There is a defined "project setup phase" at the very beginning. The process starts with concept discussion and poll. We've talked about this quite a bit, and I think everyone agrees it would be good to get this explicitly decided from the outset. Also, the Topic Leader is nominated and selected at the same time we pick a concept. This will allow a wide variety of concepts -- including things like "sets" of pokemon, each with their own TL. The TL (or TL's) will be given a defined "mission" after being selected. So, just like past projects, the Topic Leader's first task will be to kick off a typing thread.
  
  
• A Stat Spread Submission thread is started very early in the process. Based on past pokemon, it's obvious there are a lot of people that enjoy discussing detailed stat spreads long before the actual stat spread poll. I really don't want to police these early bird discussions in the normal threads. I'd rather create a dedicated thread for people that enjoy these discussions and let them get as detailed as they want. Does it matter that we don't know the style, build, or anything else about the pokemon? No. It didn't stop people from submitting stat spreads early this time. So, instead of fighting it, I'd rather embrace it. If people waste a bunch of time in this thread on submissions that turn out to be useless when the stat spread poll comes around -- so be it. This will be the place for people to "prove themselves" to the TL, for purposes of being selected to submit a spread for the actual stat spread poll.
  
  
• Same thing with art. I don't know why people like to make art submissions before we have even finalized the typing, but it happens a lot. Lets just kick off the art thread very early on, and let people post stuff there. If we don't make a thread for it, people will just post stuff in the main threads anyway. That takes the thread off-track and... you know the drill. This just eliminates another moderation headache.
  
  
• The old Base Stat Total poll has been replaced with a Stat Rating Poll, based on X-Act's algorithm.
  
  
• The movepool discussion and voting has been broken up into three phases, as mentioned in earlier process workshop posts.
  
  
• More events are conducted simultaneously, particularly at the end. I don't know about the rest of you, but this current pokemon has been very boring over the last week or so. Not much is going on, there's not much interesting to discuss, and the project is really dull. In the future, I hope to have multiple threads running through the end of the project to keep the momentum going all the way up to playtesting.

I plan to update the sticky fairly soon. Most of these changes have already been discussed here. But, there are some new twists that need some feedback. Please comment quickly if you think it needs revision.

I also want to include some better support for the TL's to keep the pace of the project at the proper level. I'm not really sure how to do this. Perhaps appoint Assistant TL's? Keeping all the topics moving, tallying votes, and composing OP's can be an overwhelming job at times. What can we do to alleviate the workload, so the project doesn't "stall" when the TL gets swamped?

 

[Dane]

• More events are conducted simultaneously, particularly at the end. I don't know about the rest of you, but this current pokemon has been very boring over the last week or so. Not much is going on, there's not much interesting to discuss, and the project is really dull. In the future, I hope to have multiple threads running through the end of the project to keep the momentum going all the way up to playtesting.

Weren't they always supposed to be going on at the same time? I'm not entirely sure why they haven't been this time, but I agree it's been slow this week.

In addition, the schedule needs to be stuck to. This Pokemon has had no spillovers so far, yet it'll likely go on 50% longer than it should have because of lag time between the end of one thread and the beginning of another.

I also want to include some better support for the TL's to keep the pace of the project at the proper level. I'm not really sure how to do this. Perhaps appoint Assistant TL's? Keeping all the topics moving, tallying votes, and composing OP's can be an overwhelming job at times. What can we do to alleviate the workload, so the project doesn't "stall" when the TL gets swamped?

I think part of the problem is in the bold votes. Due to the fact they can be changed any number of times, it's hard to keep up a steady tally. The easiest way to alleviate the burden is one that was suggested before this project, but wasn't used this time: a seperate discussion thread and poll thread.

 

[latinoheat]

what options been decided for the bias on offensive/balanced/defensive type things the 20% isn't cutting it. Also why have stuff like the name poll been allowed to have multiple choice it just slows things down, vote for one move on.

 

[DougJustDoug]

I have concerns about the bias polls. I thought we'd have a balanced poke this time. Instead we got a massively defensive poke. So defensive, that we had trouble with damn near every attack and support move in the book.

(NOTE FOR THE FUTURE: Be careful making defensive pokes. It's incredibly difficult to keep them from being broken.)

I don't really know how to correct the bias polls. Right now, they are not doing the job.

 

[Dane]

what options been decided for the bias on offensive/balanced/defensive type things the 20% isn't cutting it.

Don't remember if anything was decided on, but using X-Act's calculator you can use the Physical/Special balance to find the balance. I'd say anything within [-10, 10], or "slight bias" wouldn't be counted as a bias one way or the other.

For example, 90/73 has a number of 10.6, or "Moderate Bias". If you raised it to 90/74, it lowers to 9.9, or "Slight Bias".

Also why have stuff like the name poll been allowed to have multiple choice it just slows things down, vote for one move on.

Because there's time to do it. It has no impact on anything other than the Pokedex, which can just use [Pokemon] until it's name is chosen. The problem this time is that it's the only poll going on for some reason.

 

[DougJustDoug]

I think we did too many bold votes on this pokemon. Bold votes are good for discussion, and they can alleviate multiple polls in some cases. That's good.

But, this project needs to be accessible to newcomers. I know some of you may not like it, but people need to be able to "drive-by" vote on this project. That's the way people get interested in the project.

If they vote, then they have incentive to track the results. If they track the results, they have incentive to track the discussion. If they track the discussion, they have incentive to post. If they post, they have incentive to get involved.

There are LOTS of lurkers on this project (and Smogon, in general). Many lurkers do not like to post. Smogon is an intimidating place, and posting can be a little scary. Click polls are a great low-risk way for people to "get their feet wet" on this project. I know we don't want a bunch of dumbass noob posts here. I'm fully aware of the whole "lurk moar" mentality. I agree with it. But, many aspects of this pokemon do not require expert analysis, forethought, and debate. Many times, all you need to do is, "Look at the options and pick your favorite". In that regard, the CAP project is almost unique it it's ability to appeal to a broad range users, from the newbs to the most seasoned competitive battlers.

Also, you can't track progress with bold votes. One of the great things about click polls, is that results are constantly visible. Yes, this can lead to bandwagon voting. But, it also encourages people to "check in" from time to time to see how the poll is progressing. With bold votes, you can't see the progress and the posts aren't exactly "thrilling" either. Almost all the bold vote posts, are just that -- bold votes. No enlightening discussion or anything -- just the options from the OP with some of them bolded. So people don't track the threads, and that curbs involvement.

Too many bold polls -- it's a total buzzkill on the project. It makes it harder to vote, it discourages people from checking in frequently, and it's less inviting to project newcomers.

The reason we started bold voting was to eliminate multiple polls on the same thing, and to encourage more discussion and less "drive-by" voting. I think those goals are commendable. But, on this project, we went too far with it.

Too many click polls are bad thing, but too many bold polls are a bad thing also. Next time, we need to make a concerted effort to strike a balance.

 

[Time Mage]

I basically agree with the schedule proposed. And yes, I STRONGLY agree that this project could have used some more speed. So the proposed changes are OK.

And BTW, the TO should adhere to the calendar, not just use it as a loose guidance. The calendar is the boss, the TO is the worker.

 

[X-Act]

I made the Physical/Special Balance and the Offense/Defense Balance just for this purpose. You can use that to know by how much a Pokemon is biased towards offense, defense or neither, and for the physical/special bias.

Just for a rough guideline, Gyarados (95/125/79/60/100/81) has an ODB of 9.4 (borderline balanced, slightly biased to offense), while Omastar (70/60/125/115/70/55) has an ODB of -9.8 (borderline balanced, slightly biased to defense). So yeah, an ODB between -10 and 10 would work. Of course, the nearer the ODB is to zero, the more balanced the Pokemon is, so in a perfect scenario, we would prefer the ODB to be as near to zero as possible to have a perfectly balanced Pokemon.

For physical/special balance, Dugtrio (35/80/50/50/70/120) has a PSB of 9.8 (borderline balanced, slightly biased to physical), while Moltres (90/100/90/125/85/90) has a PSB of -9.6 (borderline balanced, slightly biased to special). Again, a PSB between -10 and 10 would work, but the nearer it is to zero, the more balanced the Pokemon would be between physical and special.

 

[Time Mage]

By the way, I've just noticed... Now that we are going to use X-Act ratings for the new creations, it is very possible that, if someone used a very ridiculous stat spread, it would fit the proposed Base Stats rating but have a very high BST. Example:

5 HP/100 At/220 Def/50 SAt/220 SDef/90 Sp

It has "only" a rank of 307 ("Very good", lower than Mew & CO, and much, much lower than Garchomp's 813 rating) but has a ridiculous BST of 685.

So, do we accept proposals like this, knowing that, no matter what the BST is, the performance won't be THAT good (will be about the same as one of the 600 BST faeries), or do we impose a BST restriction as well (not going higher than 600 BST, for example)? Well, the TL could just not choose that BST for the poll, or it could even go to the poll and let the people decide if they want something so ridiculous.

Anyway, I'd like to read your opinions on the matter.

 

[Dane]

I wouldn't accept that as a spread. It's obviously skewed to enable other stats to be ridiculously powerful. As for the specific spread you used as an example, it's like a Shuckle on speed, pcp, and steroids all at once (Shuckle is 20/10/230/10/230/5).

A BST should still be reasonable, even if we're using a different way to rank it.

 

[Helios]

Ridiculously powerful? If its weak to SR, Garchomp can 1HKO it with a SD dragon claw and its offensive stats are mediocre. Its a very inefficient stat spread; HP>defenses when considering efficiency. Just ask Blissey

 

[X-Act]

No, that is not Shuckle on steroids.

5 HP/220 Def/220 SpD is slightly less defensive than Mesprit's 80 HP/105 Def/105 SpD. 100 Atk/90 Spe is slightly better than Mesprit's 105 Atk/80 Spe. So consider such a Pokemon as being a Mesprit which can't attack from the Special side. Case in point: if it had also 100 base SpA, its rating would be 436, and Mesprit's rating is 424.

This example actually accentuates further my premise that BST is useless.

 

[Time Mage]

That's why I was asking. I posted that stat spread on purpose because I know it's not overpowered at all (Although HP Evs would improve the thing substantially), so it's not a matter of exploiting the ratings systems and making an overpowered pokémon (which proves how solid the formula is), but a matter of consistency with the game. In the same vein we try to make movesets similar to the ones in the games, having a simple BST restriction of "can't be higher than 600" would allow for most reasonable spreads without conflicting with the trend that all pokémon stats follow.

Still, I wouldn't be adamantly opposed to a "every BST is fine" approach, but since most reasonable spreads can be done with a 600 BST restriction, I think that should be imposed, for consistency with the games.

 

[X-Act]

Oh, in that case, that's okay.

I think that this system will produce better Pokemon in the long run. Syclant, Revenankh and Pyroak all were created with the BST in mind, and the first two turned out better than we thought they would be (and I'm sure that Pyroak will end up like that as well). Using rating instead of BST will be better I think.

 

[eric the espeon]

An idea to improve the stat ratings, why not account for evs? You could have an include best ev spread checkbox.

For example it would make Blissey and Shuckle (and other Pokemon with extreme stats) look much better, and if the best ev spread for high stat ratings was close to the most common spread in real Pokemon it would show that stat rating system is accurate.

I don't know how feasible this is, its just an idea.

 

[tennisace]

I have the same problem with bst, it should be within reason. if not, then you could get Shuckle in reverse while getting a low rating, a la Deoxys-A.

Edit: We also need a clearer definition of balanced. Do we want Kingdra (95 attack and defenses), Azelf (125 attacks and 70 defenses) or Uxie (70 attacks and 130 defenses).

 

[X-Act]

Kingdra is balanced, Azelf is offensive, Uxie is defensive. At least according to my definition.

Offense/Defense balance for Kingdra is 4.4, for Azelf is 42.5, and for Uxie is -19.2.

So Kingdra is the only balanced Pokemon out of the three.

 

[Time Mage]

You didn't really pick the best examples, tennis :/

Those three could be in a pokémon dictionary under the definitions of "balanced", "offensive" and "defensive".

 

[Captain]

An idea to improve the stat ratings, why not account for evs? You could have an include best ev spread checkbox.

For example it would make Blissey and Shuckle (and other Pokemon with extreme stats) look much better, and if the best ev spread for high stat ratings was close to the most common spread in real Pokemon it would show that stat rating system is accurate.

I don't know how feasible this is, its just an idea.

I don't think I understand exactly how'd you get the best ev spread without burning time. It sounds excellent but how much time will it take to come up with the best EV spread? People complain that the process is going too slow as is, making another thing to vote on will only slow it down more.

How about adding 50 to every stat (except HP), that's the equivalent of having all nonbase stat factors excluding nature. Add 7 to the normalized HP stat because that's equivalent to 252 HP evs. This obviously can't be achieved in DP, but that's not the point. If all the stats are maxed out you get to see the potential of each stat. If you compared all pokemon this way in comparison to the no EV way suddenly pokemon with extreme stats like Blissey and Shuckle look a whole lot better and more realistic, imo. Feel free to tear my post apart, it's probably really off base but I think maximum of each stat, even if it's completely impossible in reality, is better than the minimum (excluding nature because that's straight forward).

 

[tennisace]

What i meant was that they all have stats that are equal to eachother. Pyroak was supposed to be "balanced" but turned out "defensive".