The Basics
My machine reviews evaluate cards with letter scores, ranging from S to F. All cards are graded in three categories -- leader potential, sub potential, and assist utility -- as well as an overall score. Rarity is taken into consideration as it affects accessibility (collab 7* are extremely hard to roll); the ultimate determinant of rating, however, is how the card in question compares to what is practically available or announced at the time the machine comes out.Many tier lists and rating schemes use inflated scales that go up to "SSS+", or some ridiculous nonsense. I believe that this is does not help give an accurate idea of how desirable monsters are relative to each other. Additionally, not using a broad range of letter grades and cramming everything into "SSS", "SS", "S", etc. ends up robbing a rating scheme of readability and ultimately meaning. Plenty of times I have looked at tier lists and been unclear as to what, if any difference, there was between the SSS and SS ranks.
Thus, I will always rate in the S to F scale. Because we rate in context of known information when the machine comes out, expanding the scale beyond S is not necessary: S will always be reserved for cards that are the best at what they do at the time that they are reviewed. Given the unpredictability of power creep, a contextual evaluation of power is almost always more helpful than any attempt at an objective one. Note that although I may speculate about game direction and possible future releases in the text of an analysis, I will never use mere speculation as the reason to increase or decrease a card's rating.
Finally, overall rating is not a simple average of the other three scores. This would imply that the card's leader potential, sub potential, and assist potential are always and simultaneously supposed to be taken into equal consideration. However, it is frequently the case that an exceptional card excels in just one of these areas, since no monster needs to perform all three of these functions simultaneously at any given time. The overall rating is meant to give an "at-a-glance" look at a monster's overall power, while the other ratings break down where that power is focused.
Rating Scheme
All examples are given in context of the time of writing, December 2017.
- S: the best of the best. Cards rated S are among the absolute best at what they do. In many cases, there is no viable replacement or alternative option for these cards. Alternately, the card may be part of a small group which can fill the same extremely powerful purpose. Often, these cards can be so strong that they open up possibilities that, previously, would have been unimaginable. A card that is simply good or even great without being among the best will not receive an S.
Examples of S ratings are Kiri and Tsubaki, for having entirely unrivaled dragon killing prowess that enables an enormous diversity of teambuilding. They are, in many cases, completely irreplaceable.
Certain cards such as Yasutora Sado, though not as glamorous as diamond rolls, might not see usage in the traditional sense, but are entirely unparalleled in their extremely useful niches. In his case, Sado is an irreplaceable component of many Fire farming builds. Cards like this also merit a rating of S. - A: excellent. These cards might not be the unparalleled best at what they do, but are powerful enough to serve as the next step down. They are still great cards overall that you would not have trouble fitting on the appropriate teams. Alternatively, a rating of A can indicate that even though something is the best at what it does, its thing isn't powerful enough to be game-changing to the same extent that an S-rating would imply.
An example of an A might be Tachibana. Her stats are nothing to write home about, but her combination of active and awakenings is excellent and powerful. She is overshadowed by Light Ideal and upcoming triple-7c cards and slightly held back by terrible RCV, but this doesn't make her less of a powerhouse. - B: solid and good, but not exceptional. B-rated cards are functional and useful, and you should not hesitate to put them on their team if they are what your team needs. That said, there are alternative options for the cards in this tier, either sidegrades or just straight up improvements. This tier also includes cards that are generally mediocre but have relevant and niche usages. Cards rated B are typically usable through the end of midgame content (Arena 1-2), and occasionally far beyond.
Two examples of B ratings are Momiji, who is quite usable for her combination of TE, FUA, and good base active but is held back by low stats, and is inferior to the truly good FUA options; and Diaochan, who is good only for a few select purposes, quite unremarkable otherwise, and often replaceable with better options if the player has access to them. - C: usable but mediocre. These are not cards you should be aiming to use, but can serve some functional purpose. They are not remarkable enough in either direction to require more of a description than that.
Note that even C-rated cards, due to the fact that old dungeons never change, are still usable through most of the game's content. They will, however, generally fail to bring enough to the table above Arena 1-2.
An example of a C is Kulia, who, though extremely pretty, is rather minimalistic on the gameplay side. Her adequate stats, 7c, and double SBR give her reasonable sub usage. At the same time she is simply unimpressive in the face of many better options. - D: not especially usable. You would probably not actively seek to use these cards on your teams, as they are easily overshadowed. In other words, you would not use these cards if you had a choice. Sadly, if you're considering a D-rated card, chances are you don't have a choice, in which case they are usable but disappointing. A D-rated card will begin to show its lack of power as early as the mid-game.
An example of a D is Cinderella, who is a painfully bad leader, relatively poor sub, and subpar assist at best. However, if your team really needed a bind clear, could use Light TPAs to any extent, and had no other options, you'd probably eventually resign yourself to putting her on the team until you rolled/farmed something nicer. - F: just the worst. Just as S is not given out frequently, it is also not often that a card has nearly no redeeming qualities. However, it does occasionally happen that monsters simply have minimal conceivable usage, and you would never expect to use them for any reasonable purpose. These special friends "merit" the lowest possible rating of F.
An example of an F is Yahiko, from the Kenshin collab, for being inferior to any possible alternative option in almost every way (until NA gets Surtr Cup, ten or so years from now). Things like Garbage Mage would also fall in this category.