London Mulligan Coming to MTGO


I'm not on Twitter so I'll chime in here. In the abstract I like the London mulligan a lot. It feels much better especially when you get two bad openers in a row, and the experience of using it really won me over. Taking the same decks from a few weeks ago and applying the new mulligan rule to them just makes games more pleasant, on average.

Which is why I'm so glad that you were able to demonstrate what happens when people don't play the decks from a few weeks ago. The key point in the Sam Stoddard article about this mulligan rule was that it influenced deck design, and wow is that ever borne out by your Outcome list's evolution. I think that's about as clear an illustration as possible that the London rule is ill-suited for Vintage, if the goal is simply to reduce the frequency of mana-screw without substantially impacting deck design along the way.

@craw_advantage Thanks for taking the time to write up your thoughts. It's similar to what I've experienced - I didn't expect to like it, but I do. It adds more skill and consistency to typical games of Magic.

For Vintage I don't think the London mulligan is incompatible with the format, but the current restricted list is certainly not optimized for London. Vintage is by far the hardest format to fix though given the structural constraints. It's also the format least likely to affect DCI decisions.

@chubbyrain said in London Mulligan Coming to MTGO:

For Vintage I don't think the London mulligan is incompatible with the format, but the current restricted list is certainly not optimized for London.

Something that occurred to me after I posted is that I assigned a certain negative value to "influences deck design" because it was treated as a negative in the original Stoddard article, but that's not really fair--any mulligan rule would influence deck design, including the presence or absence of a mulligan in the first place, and we don't need to anchor our reaction to London to the idea that the deckbuilding trends encouraged by Vancouver are ideal. It occurs to me that there might be some other neat possibilities opened up as well--if mulligans can be an intentional part of your gameplan in combo, can they be part of your gameplan in aggro? Midrange? Could this be Sphinx of Foresight's time to shine at last? So I guess I'm coming around after all, provided that we can also get the appropriate DCI action to go along with it (which unfortunately isn't a given).

The biggest thing for me is that the whole myth/fear of Dredge/Combo taking over the format seems to be debuffed to me by these last couple weeks. People haven't really added so many anti-combo / anti-dredge cards to their main decks to even make that happen either.

While MTGO shifts much more frequently than paper, it doesn't change that quickly. That said, there is a trend to more Bazaar decks and fewer Blue control and Shops decks.

Week before the London:
Bazaars: 6
Gush: 7
Paradoxical: 7
Shops: 9

Last week:

What metrics are we using? If Bazaar becomes 1/3rd of the metagame, is that taking over? Do people really have to run main deck hate for it to be a healthy metagame (I was beat by MD Relic of Progenitus out of PO the one league I played with Dredge)? What about PO? I really think a 3 week trial is insufficient to gauge the actual metagame effects of London. This was probably a PR move to expose players to the rule change and give them a chance to experience the mulligan rather than have the MC be their only interaction with it.


Small samples are sufficient for telling you significance if there is a dramatic change. There isn't a dramatic change. Based on your numbers:
Number of outcome decks in top 32 is equal
Bazaars are slightly up
Shops are slightly down
Control/Gush is slightly down
Others are roughly the same

If Bazaar was that dominant now, I'd more expect to see it winning the tournament and getting more representation in the top 8. Only 2/10 in the top 32 made it to the top 8 (expected value would be 2.5/10).

Sure. I agree there's been no dramatic change. I think we can conclude that whenever something happens, at least a few people will have a tendency to overreact, whether honestly or out of hyperbole. There's certainly a large enough sample size of MTG history to conclude that. 🙂

This test period just feels really weird to me. Like, I don't know what to do from a brewing perspective as I don't want to invest more time for one more week of a rule that might not be implemented. I guess for a Modern MC the incentive is there but for Vintage, I had one idea, it worked out, and I kind of feel set for this 3 week block. And the new set comes out on Thursday, and then there are 6 days with London, and I'm unsure how much I want to build around London with the new set...

It really feels that we've just scraped the tip of the London iceberg, honestly.

last edited by Guest


I've also been playing a lot of Outcome Storm. It feels very nice with the new mulligan rule. A few 4-1's in the leagues.

I'll start my post be saying that my opinion on the London Mulligan is not firm. It's a complex topic and relevant data are limited. Three weeks really isn't enough to gather sufficient data to make statistical comparisons (at least with the limited data that our benevolent overlords permit us to see). So we're left with weaker forms of evidence (trends, intuition, etc.)

In any case, my initial reaction to the new rule is net negative.

There are certainly positive aspects of the new rule. As others have articulated here, the new rule may increase the breadth and complexity of decisions in the early game and in deck building. In theory, this should reward skill/experience and make the decision process itself more mentally stimulating and enjoyable.

However, several factors make me especially concerned about adopting the new rule.

First, I don't think we have (or will have) the data necessary to have an idea of how the new rule will affect the metagame and play experience. I recognize that taking risks can be an important part of improving the format/game, but I feel like the magnitude of this proposed change needs a commensurate amount of data to inform it, which I don't think we'll get.

Second, I worry that the new rule incentives strategies that are broken, fast, proactive, and all-in. This could make Vintage more like its caricature, the "turn-1 format." It may also make the luck of the die roll more important.

Again, hard to know for sure. The format doesn't appear to have undergone a cataclysmic shift in this direction. But I really wouldn't expect it to happen that fast, either. There's bound to be inertia in the decisions people make--people like what they play, have emotional attachments to certain cards/decks, and don't necessarily have the time/motivation to find out how to capitalize on the new changes.

Of course, such an issue could be mitigated through restrictions. But we all know how long addressing format issues through restrictions can take. I'm worried about Vintage undergoing such a large upheaval and how long it will take things to settle.

Finally, I worry that the new rule could complicate our long history of knowledge and theories regarding deck building and in-play decisions. Honestly, I have not thought this one through as deeply, but I don't think I've seen that discussed elsewhere (but I haven't read much on the new rule, either, so I could absolutely be wrong here).

Again, my post is intentionally full of "coulds," "mays," and "appears." I really feel like I don't have a firm grasp on this. Looking forward to reading more discourse on the subject.

last edited by Zias

Well articulated post @Zias. Thank you!

My assumption was that Dredge would get worse with the London mulligan. Let's see if that turns out to be true.

last edited by Griselbrother

I feel pretty certain we haven't had enough time or games to decide whether the mulligan is better for all-in strategies.

Dredge, for example (the posterchild of mulligans), might not actually be better overall with the mulligan. If we can reach Bazaar with more cards in hand on average, that's a net bonus for the deck, but on the other hand, other decks may reach pieces of hate plus a backup (or their own fast plan) just as easy, giving us a net neutral.

What i do believe can happen is that, at least for the first period of time (some months), Dredge will uptick in popularity as people test it to confirm whether or not the mulligan is better. Which gives us both potentially biased opinions and an increase in Bazaars not necessarily due to increased power but due to perceived increased power.

So can I just run Oath shell with 4 creeping chill, then have Magister Sphinx as my only creature, powder/mull to get sphinx to bottom, then activate oath one time??

I wondered the same thing when creeping chill was spoiled in the single card thread, but getting Sphinx to the bottom was tough. Now..

Perhaps the probability of a single Sphinx in the opening 7 that you can bury through the mull is still too high.

From @evouga 's great work in the SMIP thread:

Realistically though, you wouldn't keep a 7 card hand since you don't put cards on the bottom, so if your only method of putting of putting a Sphinx on the bottom is the London mulligan you are 60% to find it? With Powders (I think)? Which is awkward since you can't powder away more than 2 Chills.

And this equation assumes you are only mulliganing to Sphinx, not a generally acceptable Magic hand that also has lands and spells.

The comments in this thread leave me pleasantly reminded of a meta pre-Brainstorm restriction. I enjoyed then the accessibility of the four Brainstorms to tailor your hands at any point in the game, which allowed you to enact your decks strategy as a function of time (early, mid, late game). The restriction of Brainstorm took away that adaptability decks had. All of the blue decks were relegated to redesign because they had to operate under a condition of more randomness. Cards like Recoup, Tinker-Bot, and even Yawgmoth's Will became liabilities in too many situations to remain the auto-inclusions they once were.

If the comments in this thread are true, then the London mulligan rule should push blue decks slightly back toward their once more interesting history. It should be theoretically more forgiving to play Oath targets or cards like Recoup, because players can put "dead" cards back into the deck, like what was done with Brainstorm. New vintage players tend to overlook the power of Brainstorm because it's not netting any extra cards, but by putting back two useless cards, you are netting two useful cards. (Some people refer to the phenomenon as net virtual advantage). The new mulligan rule essentially has the desired effect of a free turn 1 Brainstorm.

Therefore in the context of virtual advantage, outfitting decks with more "situational" combos, and mitigating mana screw the London mulligan rule should have a positive effect on Vintage, specifically on the limitations of blue/combo deck construction. (Also a natural predator of Dredge decks)

last edited by desolutionist

I believe the test on MTGO was just to gather overall player response and maybe any meta major shift. Since most people seem to enjoy the mulligan gameplay-wise and nothing huge happened in any format, I think there's no reason not to adopt London Mull - at least from the MTGO test perspective, let's wait for Mythic London this weekend.
I think that if the game experience is improved, there's no reason not to implement it. The worst case scenarios are changes on B&R lists or even pullback/change the Mull 1 year from now, which are fine by me.

The London mulligan will be permanent with the release of Core 2020 on July 12, 2019.


So now the question for shops; grafdggers cage or leylines now?

"We understand that players have had some concerns about Eternal formats, particularly Legacy and Vintage. We agreed, and gathering data on those formats is part of why we tested the mulligan on Magic Online. From what we saw, the metagames were able to adjust even better than expected. One characteristic of Legacy is that there are very efficient one-for-one answers to most threats, which has the effect of making raw card quantity important. That in turn means choosing to mulligan aggressively to a particular card or combo is more of a cost, and so we weren't seeing that strategy be very successful. Vintage is harder to gather data on due to smaller sample sizes, but we did not see any alarming imbalances jump out. For example, the win rate of dredge did not change markedly even with a greater chance of finding Bazaar of Baghdad in the opening hand. Again, this is perhaps counterbalanced by other decks also more reliably finding their sideboard cards."

I'm not normally that guy but I'm pretty sure that it is commonly understood that the MTGO meta is not the same as the paper meta because of the physical differences from things like short-cutting infinite combos and the number of physical shuffles players have to do in many lists.

I highly doubt they really controlled for that. Dredge on MTGO can mulligan 5-6 times within a minute where as physically that could take a solid 5 or so.

It looks like its going to be just what it is though, hopefully if this proves to be a mistake they are amenable to reverting the format back as opposed to just restricting cards that were otherwise fine.

I haven't been following the MTGO results, but a quick look at Saturday's Vintage Challenge shows only 2/32 dredge decks, neither in the top 8. What is the experience of people who have been testing the new mulligan on MTGO? Was the fear of Bazaar dominance overblown? What have been the other impacts?

  • 49
  • 10453