Final Check-in: Pre-London Cleanup

This is a very busy week. War of the Spark is now completely revealed, and so I have a few final cards to discuss. There have been a number of interesting cards for Modern, but we haven’t seen any obvious all-stars. Of course, that may be a lot to ask for, and more role-players are always welcome. The other thing is that Mythic Championship London is this weekend. That tournament’s results are critical for the future of Izzet Phoenix in Modern. However, an even more important factor will be the impact of the proposed change to the mulligan rule. While the proof is in the proverbial pudding, I’ve done some testing on my own, and have concerns about London.

Final Check-in

Before all that, let’s take stock one last time of the metagame before the Mythic Championship. Unfortunately, there is currently no Day 2 data for GP Yokohama. I’m not sure why there wasn’t one this week, though the lack of a Day 2 for GP Niagara also may indicate a policy change. Therefore, there’s no way to tell if the real concern about Izzet Phoenix’s Day 2 presence remains valid. There were five copies in the Top 32 and one more in the Top 8. I have noticed that non-US GPs have had fewer Phoenix lists Top 8 than the American ones, so this might not mean anything.

The Top 8 was won by Hardened Scales, which is fascinating, since it had to dodge three Dredge decks. Most Dredge decks run Ancient Grudge and Nature’s Claim, so I can’t imagine Grafdigger’s Cage or any artifact creature surviving long. Fortunately for Scales, it did dodge Dredge and thus skated to victory. Dredge hasn’t done that well in previous GPs, so I’m curious why it was so successful this weekend. The sideboard hate in the listed decks isn’t really lacking, so maybe it just ran well.

In any case, the overall metagame in Yokohama appears to be consistent with what we’ve seen previously. This in turn means I expect to see the same decks in the same frequency this weekend in London. I do predict that if the observed trends continue through London, Wizards will be obligated to intervene.

Card Discussion: Narset’s Reversal

Last week, I said that Dovin’s Veto would revolutionize control decks. First, it forces changes in the mirror, because landing and riding a planeswalker to victory is far harder. Second, it stands to change combo matchups, since Pact of Negation no longer wins counter wars. The control vs. combo dynamic will have to change again, because in the intervening week, Narset’s Reversal was spoiled. While its applications are very narrow, I expect Reversal to have a huge impact for combo decks.

Reversal copies an instant or sorcery, then returns the original copy to its owner’s hand. As Jordan mentioned, this gives you the option to turn opposing spells on their owners. I’m a bit skeptical of this usage. The original spell will be cast again, so it’s like kicking the can down the road with upside. Copying Path to Exile or Fatal Push with Reversal is like making the opponent pay an extra mana and sacrifice a creature to kill your creature. It’s not bad, but not exactly a game changer. I think the real value lies in redefining counter wars.

Just like Misdirection, this can be used to counter counterspells. Reversal remains on the stack while it’s resolving, and when the copy is made, Reversal is a momentarily-legal target. Upon resolution, Reversal leaves the stack, and the copy will fizzle for lack of a target. Since this doesn’t actually counter the counter, it should only be used to force a game-winning spell through. A more straightforward use is to simply to use it on the instant or sorcery being countered, à la Remand. While this does only work for instants and sorceries, it will also result in fewer judge calls.

A Place for Everything

While I could definitely see control using Reversal to answer opposing Vetoes, I think that combo will be its real home. A deck like Ad Nauseam needs to resolve the namesake card no matter what. Reversal provides a way out of a counter war and can potentially defend against Thoughtseize. Pact of Negation will be better most of the time, but in a world where Veto is present, Reversal stands to be prime sideboard material.

A more intriguing use for Reversal is in Storm. In addition to the previously mentioned utility of getting around counterspells, Storm could integrate Reversal into its combo. Copies aren’t cast, and therefore storm won’t trigger, meaning you can’t Reversal an opposing Grapeshot and win on the spot. However, targeting your own Grapeshot means it can be recast with additional Storm. This opens the possibility of forgoing Past in Flames entirely, an attractive option in a world full of graveyard hate. It does lose the synergy with Gifts Ungiven, but there may be other options. Watch Caleb Scherer closely; if this is a viable strategy, he’ll know.

Card Discussion: Finale of Devastation

The next card is Finale of Devastation. The first half of the card is a better Green Sun’s Zenith for an extra green. My testing indicated that Zenith’s power was much higher than expected, and now there’s a fixed version. I can’t imagine Wizards would allow the two cards to coexist, since the fear of too much tutoring is very real. I was skeptical of an unban before, now I’m writing it off entirely.

While Finale can still find Dryad Arbor to accelerate mana, it’s much slower, and jumping from two to three is less dramatic than one to two. It may be a thing anyway, but it won’t be as good. Therefore a lot of the appeal from Zenith is lost here. However, Finale compensates with some extra text. On the front end, Finale can find any creature, and it can also be used to reanimate creatures. The first part is very good, but I’m skeptical of the latter. Since you have to pump the creature’s mana cost into Finale to cast it, it’s not saving mana like Unburial Rites. Additionally, there’s very little difference between searching the library or graveyard for most creatures, so the reanimation clause is mostly a nice feature.

Obvious Home

Deck that seems like it would be the best home for Finale is Elves. The deck already plays a lot of tutors and generates unreal amounts of mana with Elvish Archdruid and Heritage Druid. I noted in my Zenith test that Elves really liked the extra tutoring and while the explosive starts are great, the real power of the deck comes from tutoring a lot. I tried out Finale in the proxied Zenith test deck and while it was noticeably worse than Zenith, it’s not enough to be disqualifying. I didn’t use Zenith to accelerate with Arbor much anyway, so in practical terms, the only difference was a slightly more prohibitive cost. This tells me that Finale will see Modern play.

Side Benefits

As compensation for the extra mana, Finale isn’t restricted to green creatures. Every creature is on the table, though take that with the whole salt shaker. Thanks to the extra green, Finale won’t be played outside heavy-green decks, and so will mostly find green creatures. I’m also having a hard time coming up with other colored creatures that I want to search for.

Searching for non-green creatures puts Finale in competition with Chord of Calling. Chord is an instant, but it also costs one more green. However, convoke means it doesn’t actually cost any more mana. Being an instant also opens the door for Restoration Angel tricks or Spellskite protection, which probably keeps Finale out of Kiki-Chord and similar decks.

Hatebears is another possibility, but finding small creatures is Collected Company‘s job. Tutoring for Thalia, Guardian of Thraben or Gaddock Teeg is very good, but Company provides so much value that it’s a tough fight. The versatility is great, but the competition is very stiff. I realize that the tutoring clause isn’t the only text on the card, but I’d argue it’s the only relevant text.

Cool but Impractical

I’ve skipped over Finale’s second clause so far because it is a mystery to me. I get what it’s going for, but I don’t see the point. I dump an incredible amount of mana into the spell, find Progenitus or something else huge and functionally unblockable, give it and whatever else is still untapped at least +10/+10, and then TIMMY SMASH! and I have a hugely satisfying win with a great story to tell my friends. If that was all Wizards wanted, I’m pleased to say they succeeded. But why should I bother?

There’s no strategically sound reason to go that big in Modern. The clause only triggers if X is 10 or more, so it takes a minimum of 12 mana to happen. If I’m dumping twelve mana into a spell to find a huge creature, it should win the game anyway. Besides, I wouldn’t cast Finale for 10GG or more in the first place. Tutoring for Ezuri, Renegade Leader or Craterhoof Behemoth accomplishes the same thing but cheaper and with trample.

For the other decks that could produce 12 mana in a normal game, it’s not a very necessary addition. Searching up Primeval Titan and then immediately swinging for 16 sounds great, and should win the game, but it’s not likely to happen. A Valakut deck with 12 mana available should have mostly won by that point anyway. Amulet struggles to get that much mana without first resolving a Titan, and already has ways to win immediately after casting the creature. Besides, Summoner’s Pact is a far more efficient and versatile, especially post-board. Tron could easily hit that much mana and find Ulamog, the Ceaseless Hunger, but Ancient Stirrings does that already.

Intriguing Mulligan

Finally, it’s my turn to comment on the London Mulligan. This proposed change has been met with some cautious optimism, since in theory reducing the instance of non-games improves the experience and therefore the game. My worry has been that this will backfire. Arguably, Magic only works because of the risk of non-games from bad hands forces players to moderate their deck construction. I’ve heard that Arena players are exploiting the hand selection system to play aggressively low land counts. If the new mulligan reduces variance as it seems to be designed to, that opens a huge risk for exploitation. The string of articles mathematically proving that it benefits specific-cards decks like Tron increased my concern.

However, I didn’t want to say anything until I had some actual experience with the mulligan. My experience has been very neutral, and I can see what Wizards is going for. However, other players are finding that the London Mulligan is very useful for some very worrying decks. If their personal experiences accurately reflect the metagame reality, then this could be a very big mistake. They’re arguably already exploiting the new system until it breaks.

Tale from the Trenches

When the London Mulligan was announced, a Grishoalbrand player at my LGS, Black Gold, swore that if it were implemented he would grind leagues until he 5-0’d so many with turn 2 kills Wizards would have to admit their mistake. In preparation, he spent about a week trying to make a version that had a turn 1 kill work in paper, but it never came together. Then Wizards decided that one tournament wasn’t enough data and initiated a trial period on MTGO. I caught up with Ken last week to ask him if he’d followed through with his threat. The answer, surprisingly, was kinda.

Ken has been grinding competitive leagues and claims to be blitzing through them and racking up the 5-0’s. I can’t confirm it, but I also have no reason to doubt him. However, he hasn’t been using Grishoalbrand. Instead, he’s been playing Narset Cannon. His reasoning was partially that it’s really cool and weird and he enjoys it. The more important reason is that it abuses the London Mulligan better than Grishoalbrand thanks to its inclusion of Serum Powder. Specifically, he not only needs specific cards in hand to win but also certain ones in his library and going to London favors and rewards decks like that.

Narset Cannon wasn’t actually a deck beforehand, because it’s too brittle and inconsistent, but Ken says the London Mulligan changed that because now he can mulligan and sculpt until he finds a really busted hand and win on turn 2. And he has been winning on turn 2. He didn’t have statistics for me, but he claimed that his turn 2 win-rate is surprisingly high. He’d seen an uptick in turn 2’s while still on Grishoalbrand but not sufficient to actually stick with the deck. Admittedly, if he doesn’t win on turn 2 he frequently doesn’t win at all, but for him that’s a small price to pay. Testimonials aren’t data points, but it does support all the theory crafting’s conclusions which in turn suggests validity.

Some Positives and a Warning Sign

I have not been grinding leagues and don’t play that much MTGO in general. However, the experiences I have had and some testing I’ve done on the side raise troubling questions. As someone who primarily plays Aether Vial decks, the main change I’ve noticed is that I don’t have to mulligan as much. The actual gameplay didn’t dramatically improve for my decks, but over a two week stretch playing paper Magic with fair decks I had to mulligan past 6 in ~10% of games. During the same stretch on MTGO using the London Mulligan it happened ~5%. I also found the decision on which card to send easier than expected. I thought that there would be a lot of strategic decision making and potential to wreck myself, but in practice, it was obvious what the correct choice was every time.

However, in that same period, I also saw an uptick in glass-cannon decks. I’m used to hitting weirder-than-Grishoalbrand decks in about once every fifty matches. I’ve seen them roughly once every ten since the trial began. I don’t know if this is just players assuming that the mulligan benefits these decks and all the talk has become a self-fulfilling prophecy, it’s just me, or if this is a real thing. If the latter is the case, that is very worrying, and will probably lead to the abandonment of the new mulligan.

Judgement Week

I really hope that Ken’s experience with the London Mulligan is a fluke, and it doesn’t advantage Broken-Hand-Only decks as much as he claims. However, my experience can’t actually refute his claim either. I’ll be watching London closely next week. Hopefully, I’ll be able to report that my anxieties were only that and Wizards knows what they’re doing.

8 thoughts on “Final Check-in: Pre-London Cleanup

  1. I think a lot of the MTGO is people seeing how much they can get away with under the new rules. That being said, I love the new mulligan rule and would hate to see it go away if they can mitigate the potential issues with a couple bannings. I mean, if the requirement to have a better mulligan rule is that (just for example) Ancient Stirrings, Creeping Chill, and Goryo’s Vengeance end up getting banned, I would be OK with that. Now obviously, if they would have to ban a truck-load of cards, then that becomes much more problematic.

  2. I dont like the rule because I like the fact that magic has that flop factor. There is a mathematical chance your deck doesnt do its thing. I enjoy having this be a factor in the game. I dont really want to see that eliminated.

    1. I feel that fewer instances of game wins and losses coming strictly down to opening hand variance is a good thing, as it rewards player skill more. Also, going down resources is still punishing. Sure, a Tron player is much more likely to have their 4-card hand of Tron and a payoff on a mull to 4 with the new rule, but a deck like Shadow can easily just take your payoff and go kill you while you’re trying to find another. There’s a big difference between making mulligans *less* punishing and totally removing punishment.

      1. Both points are true. The problem comes down to whether that lesser punishment is too low given the rate of reward for decks like Narset Cannon and Grishoalbrand that are already high variance.

        I suspect it will come down to the same problem with Preordain and Ponder in Modern. Everyone benefits from improved consistency. The problem is that high-risk, high-reward decks benefit more than the fair decks, and this has played out in every format and over time. I know that players point to Force of Will coupled with cantrips in Legacy keeping Belcher or Oops, All Spells down, but the odds of a given player having Force in their opening hand is only ~40%. It’s the fact that most Belcher hands can’t win in the first place that keeps them back. Thus I’m worried that boosting everyone’s consistency will incentivize player to play more inconsistent decks to take advantage of the improved mulligan.

        It may be that my fears are unfounded, but we won’t know until after London.

      2. You say it rewards player skill more and I say it depreciates the value/skill of deck building, which is why we are having a discussion about bad decks, and also degenerate ones, possibly being viable with this adjustment.

  3. I’ll try to be constructive with my feedback here but… The London Mulligan summary was quite dissapointing. It uses someone elses (partly unsubstaniated) experience and then summarises as if its fact. You start by saying that you didn’t want to comment until you had some experience, then continue to say that you only have a little data because you dont MTGO.

    From my perspective, having played a league a day (keeps the doctor away), I feel like the move is generally pretty positive. There’s an uptick in the non-interactive degenerate decks that people talk about but there’s some really important caviats:
    1. They’re more common in the 2-mans and friendlies than comp leagues, because people are just testing the new rule out. It’s not a “I know this is great”, its a “can this deck now be playable”.
    2. They’re not winning. You don’t see them at the 3-0 / 4-0 bracket, you see them at 1-1 and 2-2.
    3. When you face them, your postboard games require you to have specific hate cards (if you’re on a fair plan or a slower combo plan). You can mulligan to those more easily.

    I’ve found the new mulligan to be great, and so far don’t even feel like you have to ban anything. I am still behind when I mulligan to 5, but I can still play some meaningful magic. I agree that often “what goes back” is obvious, but that’s very deck specific. I’m playing a Birds of Paradise deck, and it can be tricky to know how much ramp and how many lands you should keep, since the opponent could be a bolt deck (as an example).

    I’ve found the new mulligan to be great, and I’ve seen plenty of interactive decks doing well at the same time.

    1. You misunderstand. I have been extensively testing outside of MTGO. It’s just that the testing and my limited experience online didn’t really tell me anything about whether the London Mulligan is a good idea. Since I play fair decks, I’m not noticing anything special. For me, it’s a big meh. My opponents seem to be playing wonkier decks, but that might not mean anything.

      However, players that are grinding tell me that it has an enormous effect on their matchups when they’re playing really unfair decks. This is why I shared Ken’s experience, however he’s not the only one telling me that. Thus I can’t shake my fears despite my fairly neutral experience.

      The bottom line is that we don’t have enough data to tell if there really is a potential problem or it’s all unfounded. Wizards has all the MTGO data but they don’t share it so we just have to wait to see what they think. Also, the MTGO experience doesn’t translate well to paper since nothing is automated. There may be some logistical/time related problem that’s never come up before which kills the rule. Again, everyone should withhold definitive judgement until after London.

Leave a Reply