Traces of Magic: CAP 10 Update

By Deck Knight.
« Previous Article Home Next Article »


The Create-A-Pokemon Project (CAP) is turning 10 CAPs old. Over the two-plus years we've been in operation, we have made some truly excellent and metagame-altering Pokemon that each have given new insight. It was a bumpy ride, but even after all the work of nine previous projects, we aren't running out of steam. The process has been refined and we can now move forward with more clarity and focus than ever. This article will backtrack a bit from the beginning of Colossoil, through revisions, and into our current CAP project, CAP 10.

A Gutsy Leader

CAP 9 was the first project where we implemented the policy of a "strong" topic leader. Discussed briefly in The Smog Issue 4, the new "strong" TL would be more assertive, command authority over slates, and provide more input into the project. During the Concept Assessment stage, we would determine questions the concept was supposed to answer about the metagame and how we would achieve those ends, loosely, in the polls to follow. Plus was selected as the Topic Leader of CAP 9. He was bold and outspoken and had the depth of metagame knowledge to keep the new process from going astray, and it showed in his selections. Colossoil is a terror to face for the secondary threats it was built to address, and it has become a monstrous addition to offensive teams.

One of the more interesting notes since the advent of the strong TL is that our selection of slates are much more exclusive. "No Secondary Typing", for example, had been included in every CAP previously, but in CAP 9 and CAP 10, it was not included as an option. CAP 9's secondary ability poll was also very exclusive in that it only had Guts and No Secondary Ability as choices. Guts won out by a small margin, and unfortunately the custom ability Rebound, which operates like an automatic Magic Coat on the switch-in, was largely overshadowed. Most of the criticism over TL selection slates has been very minor, limited to only a few contributors.

The strong TL model worked very successfully. Plus more than anything deserves credit for stepping into the role and performing it so well. CAP 9 was a (wall-)smashing success. This leadership model will be the basis for CAP projects in the future, so it is fundamentally important to emphasize.


The original three CAP Pokemon, Syclant, Revenankh, and Pyroak, were all created in the days before Platinum and required a much needed update. Additionally, all the previous CAPs also needed to be adjusted with HGSS moves. Umbreon Dan lead this project and used the community features to provide a makeshift voting site for revamped aspects of the original three CAP creations. Various respected community leaders each selected the other CAPs to get a new base of moves and they were moved through expediently. Truth be told, HGSS tutors were not overwhelming additions to any movepool.

Syclant had already gone through one nerfing process long ago and it was time to update again. The original Syclant was too powerful, but the new Syclant was underwhelming. It regained some of its lost Speed to jump up to base 120 and gained a lot more coverage moves. It also got back three of its original moves in Superpower, Focus Blast, and Spikes. Revenankh got a huge increase is Special Defense which some thought unwarranted, and went back to terrorizing CAP with the Bulk Up + ShedRest combination. Now that it possessed Will-O-Wisp, it was also capable of crippling physical attackers in an entirely different way. Pyroak’s entire outlook shifted towards offense and it gained powerful status attacks in Stun Spore and Sleep Powder, as well as Dragon Dance. Given the relative irrelevance of the HGSS tutors on the metagame, the other CAPs were not playtested again. Umbreon Dan’s article in Issue 7 covers the revision process and results in much greater detail.

MultiTYPE or MultiHYPE?

Enter CAP 10. Beej was selected from the nominees to lead CAP 10 against a massive headwind of novelty, given such an important “anniversary” CAP. Mutitype CAP had been the subject of many discussions in #cap and on the CAP battle server for quite some time, and it showed in many of the submitted concepts. Eventually, reachzero’s concept of “Utility Counter”, a Pokemon that could counter any specific threat but not every threat at once, was chosen. Concept assessment was the battleground for deciding between Multitype or going through more traditional means.

The CAP Policy Review Committee through our social group made a discussion topic about it. CAP usually runs in steps that go from typing through to stats, ability, and movepool. Multitype is problematic because it effectively decides typing and ability in one swoop, and also has a huge impact on movepool submissions and suggestions. Multitype was, in essence, a completed product. The initial question would be about how to implement it, but it shifted towards whether it was even fair to include it at all. If Multitype got its own massive poll jump, there would be almost no way to prevent the bandwagon it entails. Beej eventually decided to sever the idea at the head, and CAP proceeded normally.

From a discussion perspective, this was the most involved discussion of a single element since the Drought/semi-Drought playtesting during CAP 3 (Pyroak). That was the last time a single element would have so drastically altered the outcome of the project. After testing, we determined Drought, either permanent or temporary, was too powerful to allow and selected other abilities. Multitype presents the problem that it’s difficult to balance; either the Pokemon is too weak and cannot overcome the loss of item benefits from Choice items or Leftovers and the weaknesses that come with mono-typing, or the Pokemon is too strong and becomes seventeen separate offensive and/or defensive threats, as it does not need an item to give it effectiveness. For all the potential diversity Multitype offers, it is at the same time very limiting to the rest of the process.

What’s Your Angle(r)?

The typing polls proceeded as normal. The biggest debate was between choosing a type with a lot of resistances and bad weaknesses vs. a Pokemon with few weaknesses and either immunities or key resistances. The first slate was between Electric, Water, Fighting, and Steel. Electric was eventually voted as the primary type because it has only one weakness and it resisted the most ubiquitous priority move in OU, Bullet Punch. Ground was a very serious weakness, but is arguably the easiest to address with both a move and an ability dedicated to its immunity. For secondary typing, we were looking to round out Electric by either mitigating the Ground weakness or continuing in the earlier vein of a pokemon with few weaknesses and key resistances. Water and Poison was the slate selected by Beej. Poison compounded the Ground weakness, but in exchange offered resistances to Bug, Grass, and Fighting and an important Toxic immunity, as well as the absorption of Toxic Spikes. Water provided another decent offensive STAB and addressed more of the special attacking threats with resistances to Fire, Water, and Ice. Water eventually won, and Electric/Water is CAP 10’s typing.

We implemented a new policy for stat ratings in CAP 10. In previous projects, we would select our stat ratings based on Offensive/Defensive Bias and Physical/Special Bias, as well as the Overall Rating. The old process made stats an “island”, especially considering how often slight biases in either category were chosen. In order to have a slight bias to defense with a bias to physical, all you had to make sure of is that your defensive stats totaled more BST than your offensive ones, and for either physical or special bias your biased stat had to total more than your unbiased one. Cyclohm is the best example of this, where slight bias to defense and slight bias to special resulted in a Pokemon with high Special Attack and high Defense. Because SpA + SpD was higher than Atk + Def, it was biased to special. Speed and HP act as multipliers, so they are more difficult to account for, but generally if HP + Def + SpD is greater than Atk + SpA + Spe, the Pokemon has a bias towards defense.

Interesting note on Cyclohm: At the time the ratings formula treated physical and special attacks the same, Cyclohm had a slight bias to special. When X-Act updated the formula to account for the generally greater effectiveness of special attacks, Cyclohm now has a slight physical bias.

In our new system, the Topic Leader would choose the Physical Sweepiness, Physical Tankiness, Special Sweepiness, and Special Tankiness ratings that could be used, and the job of the CAP community was to convince them of the proper ratings for fulfilling the concept. Because our Utility Counter needs to be well rounded, we chose Rank 5 Offenses and Rank 6 Defenses. There are some weaknesses in the new system. Since HP and Speed are multipliers, they greatly affect the makeup of the ratings. A low Speed or a high HP stat will allow you a great deal more raw offensive or defensive power while keeping the same rating score. One of the spread submitters caught wind of this and proposed a defensive spread of 254/44/45, which was perfectly legal. Ultimately the voters decide based on the calculations provided for a spread and its presentation, and I wrapped up my third stat spread win in a row with a submission of 151/84/73/83/74/105.

Even with Multitype gone, ability was still a major point of contention. The first slate of abilities was Filter/Solid Rock, Intimidate, Magic Guard, and Trace. The biggest battle yet was about to be waged between Intimidate and Trace. Taking a page from Poryon2, Trace was extremely useful in a variety of niche situations, countering several specific pokemon with great ease. Porygon2 itself took a while to fall from OU, and Trace was one of the things that kept it up so long. Intimidate was a different beast entirely. The concept plays an important role in Trace’s victory over Intimidate. The idea behind Utility Counter was to be able to counter any specific threat, but not every threat at once. The primary argument against Intimidate was that it allowed CAP 10 to counter too much at once, thus violating the concept. It would allow it to counter almost every form of Fighting-type offense and reduce the damage from Earthquake more than Filter/Solid Rock would. Trace cruised to a large margin of victory, but the argumentation that got it there was intense.

Secondary Ability was a similar slugfest between Magic Guard and Poison Heal. The general agreement was that the secondary ability should allow CAP 10 to counter different threats than the primary ability, and so abilities that reduced the effects of bad status jumped to the forefront. Magic Guard made CAP 10 immune to indirect damage while a CAP 10 with Poison Heal could attach Toxic Orb and be immune to all other permanent status and still regain HP in negative weather. Magic Guard was victorious by a small margin.

This Discussion Bug(s) Me:

The Counters Discussion wasn't nearly as heated as the rest of the process thus far would indicate. Most of it was deciding exactly which Pokemon should CAP 10 counter at any given point in time. It was generally agreed upon that the counter threats but break walls, and that a Utility Counter shouldn't also be a wall breaker if it could be helped. Swampert and the bulky Grass Pokemon thus jumped to the forefront of soft countering, at least. The thread then devolved into speculation on movepool and was wrapped up after that.

In the interim, Art and Name were selected. Everyone say hello to Krilowatt, submitted by newcomer Gamer128. Watt's in a name? Funny you should ask.

The Attacking Moves Discussion returned CAP 10 to its normal course of heated argumentation. At first it was suggested we ban all Bug moves because they defeat Celebi, which was categorized as a defensive threat of the wall variety (and thus not a prime target) too easily. This line of reasoning continued until posts arguing that Celebi need not be coddled, nor could it be if Krilowatt received Fire or Ice moves anyway, and the movement to oppose Ice Beam and support Bug moves began. I deserve some small credit for that one as I initially argued it and it went on from there. After all was said and done, Ice Beam, X-Scissor and Signal Beam made it while Bug Buzz was banned without a vote.

As of this writing, only the latter half of the movepool threads remains and minor details. Krilowatt will be up on the CAP server before you know it.


CAP 10 is shaping up to be one of the most contentious CAP projects to date. This is a fitting tribute to all the hard work that has been put in over the years and all the changes in policy and process that went along with it. We’ll get the final results after playtesting, but CAP 10 looks like it will impact the metagame in an untold number of ways. Stay tuned to #cap on IRC and Doug’s Create-A-Pokemon Server for all the excitement, and join us in the CAP forum if you haven’t already.

« Previous Article Home Next Article »