A quick note before we begin, just like in SOI, I have done a tiny bit of guesswork on the numbers. I'm assuming every pack has one common/uncommon DFC (in place of a common), and 1/8 packs also have a Rare/Mythic DFC (in place of another common). I haven't seen this confirmed, but I'd be very surprised if it weren't correct. Also, there was a little bit of guesswork on how the print sheets worked to get the ratios right (particularly for DFC). However, it's worked out such that for Uncommons, Rares, and Mythics, the number you expect to open doesn't care if the card is DFC or not, and for Commons, you are getting about 1% more of the non-DFCs - not enough for the vast majority of people to notice over the course of the format, particularly given that card-to-card variation is going to swamp that for at least several hundred packs.
And just like in BFZ/OGW, you want to realize that the old cards are much rarer now: in fact, you're slightly more likely to open a particular card of one rarity higher from EMN than you are from one a single rarity lower in SOI. To give an example there, I'm slightly more likely to open Hamlet Captain than Byway Courier in any given draft, slightly more likely to open Bedlam Reveler than Rise From the Tides, and slightly more likely to open Tamiyo than her Journal. You don't want to know what your chances of opening Avacyn are now (but I'm going to tell you anyway: one out of 120 drafts; on the plus side someone at the table will open one only one out of 15 drafts now, so you get wrecked less often).
Without further ado, onto the analysis!
Madness is significantly down in this format. Madness cards themselves are down 26.5%, and discard outlets are down 15.5%. Note that this is also a significant shift in the ratio of Enablers to Payoffs - before, we had about 1.3 enablers for every madness card, now that ratio is up to 1.45. What that means is that the simple ability to have a discard outlet should be somewhat less valuable (both because there are fewer payoffs overall, and also because it is proportionally more likely to be redundant). But the actual effect of Madness itself might be better - if you really want to use those activations, you are going to want to be getting value, and there's just fewer options for that now. It's unclear how that will trade off against slightly fewer enablers existing.
Delirium is also less prevalent (maybe Emrakul doesn't have people crazy so mcuh now as completely gone, crazy Eldrazi?), down about 10% from last format. More importantly, though, it's going to be harder to turn on. We have a 9% drop in, for example, enchantments that go to the graveyard through play, but that's including Crop Sigil, which only goes if you already have delirium, Spontaneous Mutation, which likely only gets there if you use it as a trick to win a combat you would otherwise have been trading, and Choking Restraints, which you need to activate for potentially little value to get there. If we take those out, we'd be down over 60%. Of course, that's not entirely fair, because there were more unplayable Vessels than questionable Lunar Forces we have now, and sometimes you will get those things to happen, but the point stands that it's significantly harder. Enchantment was a fairly easy type before, if you wanted it; now your only really good bet is Boon of Emrakul (which should make that effect a bit better).
In terms of other types, there's about 5% more instants now, while sorceries remain within 1% of where they were before. These aren't huge issues, since those cards were already disproportionately being cut in order to keep creature counts high. Milling was very important though, since it tends to be your best bet at lands (and even more so now with fewer Forks in the Road and Warped Landscapes). That, too, is down about 5% from where it was previously.
So it's not a massive change, but it's a little less important than before, and a bit harder to get turned on.
Skulk is down 39%. It wasn't all that important before (a little more than the flavor text it was oft joked to be, but not tons), and I don't expect it to be now - there are a couple rares which try to use it to effect, but well, ok, that's rares for you.
Investigate is not present in Eldritch Moon, so naturally it's exactly 1/3 as present in this format as it was in the previous. The implications of this are that cards which care about investigating or having lots of clues (Graf Mole, Erdwal Illuminator) will get significantly worse. On the other hand, the actual act of investigating will probably get better, since you're less likely to get to the point where you have too many clues and not enough time to crack them all.
TribesAll the same tribes are back, though maybe not all at the same level of importance.
There are about 4% fewer humans than before, along with about 5.3% fewer cards that care about them. About 35% are now in white, and 31% in green, with the rest spread over the other colors, which is within a couple percent of where they were before. Of note, the green cards that care about the type are a bit more aggressive than the white ones.
The number of spirits is down about 6%; the number of cards for which the type matters is down from the already paltry 4.7 per draft by an additional 43%, all the way to 2.7 per draft. This tribe continues to not be an important factor in terms of typeline in limited, being more impactful here in terms of flying.
The number of zombies is steady from last format to this, but the number of payoffs or cards that care about them is up by over 14%... but still only at about 5 and 1/6 per draft. That's enough to do something with in certain cases, but not something you should expect to see on a draft-by-draft basis.
The count of vampires are up about 3%, but the number of payoffs is down 18%. Again, this wasn't a huge number before, but the only non-rare card (out of 2 total) that does something in this set is a cantripping lava spike that cares about the number of vampires (which is probably unplayable unless you get very all-in somehow).
Wolves and Werewolves
More of these now care explicitly about Werewolves, whereas before it was almost always either or. Furthermore, the total number of these is down about 11% now, and the payoffs are down by 26%. This is one of the tribes that had some cohesiveness before, with cards like Howlpack Resurgence, Moonlight Hunt, Howlpack Wolf all seeing pretty serious play, but now there's far less reason, which in turn probably makes all of those cards worse. Also, they used to have mechanical cohesion with their flip triggers playing well together, but that's more or less out the window now - all but a mythic from the new set are based on pouring mana in (more on that later), which really function pretty independently.
New Mechanics and ThemesEmerge
The fist thing I want to note about emerge is that all of the Emerge costs total 1 less than the creature's normal CMC - so even if you're sacrificing a token, you are getting a discount. Next, strictly speaking, they're colorless, but they're expensive enough that I imagine that you aren't gong to be playing them in very many decks at all where you can't at least pay that color of mana to Emerge them out some of the time. (Note that sacrificing even an on-color creature can never reduce the colored mana portion of the emerge cost). Some of the cheaper ones might be good to keep in mind as sideboard options even when out-of-color in the durdliest of matchups, though. That they are, by themselves, very expensive, can potentially make them more susceptible to bounce (though once you have the mana to hardcast them, that flips around). It's also worth noting, though, that you are going in on one big creature, which makes you a bit more susceptible, at least potentially, to go-wide strategies, and potentially removal.
There aren't all that many Emerge cards in the set, to be perfectly honest, coming in at just under 9 opened per draft. That they are centered in one three color wedge (Sultai) will help you see more if you're in those colors, to be totally fair. And almost all of them seem at least reasonable to think about playing, so that helps, too. It's a bit hard to tell how much worse this mechanic will make enchantment-based removal, like Sleep Paralysis and Choking Restraints, but my best basis of comparison is the similar Exploit mechanic from Dragons of Tarkir - in both cases, you tend to be getting a spell-like effect worth roughly the cost of a card, then when you exploited and now just when you cast.
This comparison to exploit leads me to needing fodder to sacrifice, which more generally gets me to:
Sacrifice more broadly
We still have 6 and 3/4 other sac outlets besides emerge cards in the format (depending slightly on how you define a sac outlet), for a total that's 79% higher than we had in triple SOI. On top of that, we go from 11.8 cards per draft with death or sacrifice triggers all the way up to 18.9 (again, there's some slightly different ways you can define this, but the underlying point remains the same), getting us a massive 61% increase. Worth noting is that msot of these trigger on their own deaths, but a few care about other things dying.
On top of this, for more fodder, we are up to 40.3 Enter the Battlefield triggers per draft from 39.1 before. This is only about a 3% increase, but in my estimation, a bit higher percentage of the triggers now give you the value of a card (a la Enlightened Maniac), as opposed to being a nice tack-on bonus (see Gibbering Fiend).
This last point on ETB effects makes bounce spells on their creatures potentially a little worse, but bounce on your own creatures much better, and obviously blink and flicker effects quite a bit better. With the addition of EMN, the format comes up to 8.3 of these per draft from 7, an increase of 19%.
The way most werewolves transform is now a pretty simple mana activation. This has some things in common with Morph from KTK and especially, as several others have pointed out, Monstrosity from Theros block. There are more of these cards with low CMCs than there were with morph or especially monstrous, which is a bit different, but on the whole, I expect similar play patterns - they reward you and help you hold mana up at instant speed, as well as providing early plays which let you continue to use mana and have good plays later on in the game. Thraben Gargoyle is the one example of this kind of card in the last set - it was pretty good, and I expect basically all of these cards to be pretty good as well.
Beyond that, we also have Escalate, which is another mechanic that gives a lot of flexibility, and can also serve as a mana sink of sorts. These cards also seems to generally be good to me. And then we have just traditional mana sinks as well. When you combine all three of these effects, the number of mana sinks in the format jumps a blistering 125% (not to 25% more than it was, but 125% MORE, or 225% of the original SOI number). We do lose out on a lot of clues, which helped to be mana before, but even accounting for that, we remain at a 72% increase.
Because of all that, I expect EMN EMN SOI to be a format where you will usually be wanting to play with 18 land.
There are going to be about 3 more flash spells per draft now than there were before, a 55% increase over the previous format. This works well with the hold-mana-up-for-sinks plans, as well as other instants in general. So it makes a card like Silverstrike better, but on top of this, it's just something to be aware of in order for you to be able to play around it effectively.
I mentioned Mill before, but something important to add is that the number of cards with activated abilities in graveyards is sharply down, by about 53%. We're mostly left with a couple new zombies with similar return-to-play abilities as some we've seen before, plus all the old stuff.
FixingI mentioned I think it's going to largely be an 18 land format already, but it's worth noting that there's also very little fixing in this set. You have Terrarion, a green 2 mana 0/3 common that needs to die to Rampant Growth you, and a very strange 3 mana rock. Then there's the smattering of duals and fixing you had from last pack (Fork in the Road, duals, Wild-Field Scarecrow), but that is much diminished now. Even if you assume you can always kill Primal Druid to get the effect, fixing is down 35%, to less than 10 pieces opened per draft now. In fact, you're more likely than not to see more fixing in the last pack than in the first two combined. Worse still, a lot of that is tied up in uncommon duals from SOI, which will be very unreliable to be your colors. Almost all the rest is green. So 3 colors is a very strong no-go, and you really don't want to splash unless you have a massive incentive - I'm talking Tamiyo here.
As a side note, there are way fewer lands in this set as well, such that the number is down nearly 60%, and almost all of those are in pack 3. That does mean that you'll see more spells, which in turn means that you'll likely be taking from a slightly higher quality cut of cards, in terms of what you actually play - though this effect will be minor.
RemovalAs a note here, I am not including combat tricks or bounce spells as removal; there are some other cards which I had to choose one way or the other, as they aren't hard removal per se, but but often incapacitate a creature; I tend to include these as removal, but it's a bit of a judgment call - certainly all the numbers here are within a margin that, depending on what's good or bad in the format, it could 'go either way'.
More about specific kinds of removal when I get to creature size in a moment. For now, I'm going to say that there's about 7% more removal overall now compared to before. At instant speed, though, it's about 6.7% less; when you look at interactive spells at instant speed (including bounce and flicker), we're about the same, and when you add in the flash creatures, the current format is again in that same 7% range lower than we were before.
Creature SizeOverall, creatures are slightly more expensive than they were in triple SOI - an increase of about 1 CMC, which is about 3.4%. However, this doesn't take into account things like Emerge, and in any case this is well within the margin of "but how good are the cards" changing the conclusions there.
The average toughness of a creature now is 2.70, compared to 2.74 previously, and the average power is down to 2.39 from 2.57. Again, there are playability issues here, but there is something significant in terms of the shift to a lower Power:Toughness ratio. That means the format will likely be a little slower, with more groundstalls. More groundstalls make evasion better, but the rate of flying is also down by about 13%, so those cards probably go up in value ever so slightly.
I would be remiss to not note that these figures don't take into account the 2/2 zombies and 1/1 flying spirits which still litter the plane, nor do they account for the now fairly common 3/2 Eldrazi Horrors.
When you look at actual distributions, 2 is the most common power for a creature, followed closely by 3, and a bit further behind by 1. Only 9 or so creatures per draft have power greater than 3, at least without looking at the back half of DFC. On toughness, 3 edges out 2, with 1 still being ahead of 4.
More concretely, when you're looking at creature combat, defensively, a 2/4 is where you start to get a very stable body. It stops the tokens, eats most of the 2s and 3s, and bounces off of most things until a few 4 drops and a decent number of 5s can rumble through. Moving up to 3/4 allows you to eat almost all 2s and 3s, and most of the 4s, and 4/4s are bigger than just about everything below 5 mana, and not smaller than much even then. The 5th point of toughness lets you live to block to the vast, vast majority of attackers.
Offensively, the 2nd point of power is as important as ever, to provide a threat and be at least able to trade with basically any playable creature. The 3rd point gets you to where you at least trade with cards up to 4 and 5 drops (and even with some of those). The 4th point of power means very little is going to straight-up eat your creature, and very little will bounce with it as well. There are a few X/5s in the format, but for the most part, 4 power will be able to attack more or less freely. The third toughness is important, as usual, to be able to get through 2/Xs, but the 4th is pretty important, here, too, with a healthy number of 3/2s, 3/3s, and even some 3/4s, plus all the Eldrazi Horror 3/2 tokens.
In short, the creature sizes vary pretty typically for a modern limited format. It's not the world were Bears rule the roost and Pikers are fine to hold off the bears, as Origins was, nor is it the one with lots of giant monsters all around, as BFZ was - though it's a lot closer to the latter than the former.
In terms of how the creature sizing interacts with toughness-based removal:
1 or less toughness = 19.5% of creatures
2 or less toughness = 48.6% of creatures
3 or less toughness = 79.6% of creatures
4 or less toughness = 92.7% of creatures
5 or less toughness = 97.6% of creatures
6 or less toughness = 99.0% of creatures
7 or less toughness = 99.96% of creatures
13 or less toughness = 100% of creatures (base stats)
Again, these numbers don't take into account tokens, which typically have 2 or sometimes 1 toughness.