The decks with an MUI less than or equal to 0.25 correspond to what we think of as “fair” decks, in that they are “curving-out” there mana requirements. The linear, aggro decks GW Bogles, 5C Humans, and Burn are here. Dave identified Bogles as Ambiguous, rather than Fair, but in regards to its mana requirements, it is low to the ground and does not distort its mana needs to accommodate a combo, so it is an “Even” deck, regardless of whether it is “fair” or not.
The decks with MUI greater than 0.38 are Uneven. These decks are capable of winning, in one explosive turn, from what might appear to be a losing position. The usual suspects are here, Living End, Affinity, Tron, Valakut, Storm, and Ad Nauseam. Counters Company can make infinite mana, search its deck for Walking Ballista, and kill you on the spot, so it is no surprise that it shows up here as well. Jeskai Control may seem out of place, but the combination of zero mana Ancestral Vision, and Emrakul, the Aeons Torn create a large MUI, and cheating Emrakul, the Aeons Torn with Nahiri, the Harbinger provides the “Oops, I win” combo that we expect from “unfair” decks. So, in my opinion, it belongs in this company.
The “Delayed” category includes the decks that utilize zero mana cards, and/or some large mana cards, that delay the mana requirements of the deck, and inflate the MUI. These decks will start out with a “Fair” game, but are attempting to leverage a game resource to end the game. RG Ponza is trying to ramp into big spells, Grixis Shadow is leveraging its life into a giant creature, UW Control plays a Sphinx’s Revelation and an Ulamog, the Ceasless Hunger, which inflate the MUI, but give it a long game. Eldrazi Tron is not Gx Tron. It plays fewer large creatures and more creatures at the four and five mana slots, so its MUI comes in at an intermediate value, and I expect that we agree that it loses some of the inevitability of Gx Tron as a result. But it is ramping into relatively powerful creatures that will take over the game if given enough time. GR Hollow One also appears here, mostly because in this deck, Hollow One is not a 5-mana spell, but more typically a 1-mana spell, and that keeps the MUI low. Now I am not asserting that this is a “fair” deck. It is “unfair” precisely because it can play Hollow One as a 1-mana spell on turns one and two. But it is not an “Uneven” deck and plays differently from “Uneven” decks. Much like Burn, it comes out fast, but if you neutralize or evade its fast start, you are often favored to win. The “Uneven” decks tend to be slower, but snap their trap suddenly after they have assembled their resources.
This use of MUI adds a fully quantitative value to Dave’s semi-quantitative “Fair”, “Ambiguous”, and “Unfair”. It is a value that is derived directly from the structure of the deck, and usefully separates many of the decks that we consider “Fair” and “Unfair” based upon fundamental constraints in deck-building.

This does make me want to ask if the goalposts of fair can be moved as a format grows and matures, or if what moves is strictly the goalposts for being considered overpowered. If the definition of fairness is indeed fluid, it might help explain why it’s so hard to define it in words.

Had a similar feeling of deja vu as well. Especially seeing the comparison based on the list of top 30 decks (tier one two and three). The odds an unbanning would create entirely new decks is slim, as are the odds of it meaningfully pushing something out of tier three.

Both this and the speed article would have been more compelling if they looked at top results of recent events in order to compile metagame shares (I assume mtgo cant be used for this anymore due to how data is released). Even if its just a four tournament sample size it means so much more to say the actual top tables went from 30 percent fast or unfair to 50 percent, rather than saying of the top x decks overall 7 were unfair and now 8 are unfair. Without weighting for actual presence/share its just not an indicator of anything. Esp in tier three where we’re adding ten decks that combined have less share than any tier one deck, but each counts the same value as that tier one deck.

I think the easier question to answer is what decks have suffered or been pushed out recently – ie is mardu pyromancer still a thing? Or conversely what decks are well positioned now.

I think we’re going to see some serious changes in the format with dampening sphere. A two mana rock that anyone can play that neuters storm, tron while doing splash damage to decks like grishoalbrand, snapcaster decks, ad nauseum, eldrazi, etc has to matter.

True, a 3/2 with haste for 4 isn’t playable, but that has nothing to do with its fairness. Remember, power doesn’t matter. It’s whether the mechanics break the rules of the game. Cascade lets you play a random spell off the top of your deck at no additional cost. Free mana is an unfair mechanic. Whether that makes Bloodbraid an unfair card is the question.

You do realize that Murphy’s Law dictates that this deck now exists and will 5-0 a league right?

I’m not saying Jund is unfair. I am asking the question of whether playing an additional card off the top of your deck for free is unfair. Free mana is generally an unfair mechanic, but does that necessarily equate to an unfair card? I don’t know, but it makes me wonder.

Just to be clear: I understood your joke there, but I could be just so mad to give it a try for fun! 😉

Interesting idea. But that means adding two colors (white for nacatl and black for BB). Isn’t traverse death’s shadow just better then?

TBH that Temur deck just needs to splash Wild Nacatl for early pressure and Bitterblossom in the side for midrange (and growing Goyf past Reality Smasher) and it would be fine. A Valakut or two might help beat grindy decks too.

I agree with Graeme. In a format with TKS, Gifts, and JTMS as common 4-drop options, no way is a 3/2 haste for 2RG a good rate. Cascade obviously offsets that, but the mechanic’s randomness helps balance it out.

This is why Temur All The Unbanned Cards jamming Jace, Bloodbraid and Ancestral Vision together doesn’t put up results. Such a deck doesn’t want to run normal <4CMC counterspells because they're 100% dead cascade hits (Jund's discard spells and Ponza's mana dorks are the respective decks' worst cascades and are still better than literal blanks) but without them its early interaction is limited to burn spells and the anaemic Izzet Charm. Successful BBE decks (whether Jund, Ponza or Gruul Eldrazi) work because the cards they run are all naturally good cascade targets, while Temur has to cripple itself to support the mechanic.

Personally I view BBE as extremely fair. It’s a creature and something else, a classic 2 for 1. Sure it’s a good rate, but that doesn’t make it unfair, just like Tarmogoyf isn’t unfair just because it’s big for its mana cost. BBE generates card advantage in a manner very similar to Pia and Kiran Nalaar, for instance.