Modern Horizons 3 Quiz Wrapup

With another quiz season ended, it’s time to review how judges did on that quiz. You can read the wrapups of previous update quizzes on our website. 

As always, in my role as Exams Manager I’ll start by thanking the people who put this quiz together. The quiz team did an incredible job putting together an excellent quiz that taught judges about some important rules and policy. 

Steven Zwanger is the Update Quiz Manager, and he led the team to figure out what was worth testing in a set with almost every mechanic. Questions were written by Seth Arar, Joe Klopchic, Nathan Long, Dan Milavitz, Tobias Vyseri, Steven Zwanger, and me, John Brian McCarthy (A special thank you to Seth, Joe and Dan for providing questions for the first time on this quiz!). Testing wouldn’t be possible without the JudgeApps development work by Paul Baranay and Dan Collins. And of course, our thanks to everyone who provides feedback on a question – this feedback helps us improve questions for other judges and write better questions in the future.

Nadu Not Miss These Stats

Each quiz was five questions long (four rules, one policy), and judges needed a 60% or better score to pass. The advanced question pool had 11 questions (enough for two attempts), while the standard pool had 16 questions (enough for three attempts). 

The standard quiz was attempted 95 times by 76 judges. Here’s a chart on how folks did:

95% of test-takers passed on their first attempt, and 99% passed within two tries. 

The advanced quiz was taken 15 times by 13 judges. Here’s the chart for the advanced quiz:

Ocelot of Questions

One of the purposes of the update quizzes is fine-tuning questions that will appear on other exams. The process of deciding which questions move from the practice exams to which other exams was a little trickier than normal on this quiz. As a general rule, I don’t like to see questions on a single mechanic on the L1 or L2 rules exam unless that mechanic comes up frequently, especially in constructed. Questions on Flying are appropriate for L1, questions on Cascade are pretty important for L2s, but the difference between Adapt and Outlast isn’t something that we need most judges to know offhand.

Unfortunately, the plethora of oddball cards in this set meant that a disproportionate number of questions on our exam were “Hey, remember this mechanic and how it works?” because we did need judges to know that for this set. So a number of questions that had pass rates that would normally destine them to the advancement exams are instead being sent to practice exams, to avoid making our tests too esoteric. 

Additionally, I want to be conscious of the balance between moving questions to the practice pools and moving them to the evaluation exams. We’re continually improving our exams, removing questions that data show are just too difficult or too easy, so we need a steady influx of replacement questions. But I also want to keep growing our practice quizzes, since I know a lot of judges value these – we’ve served 412 practice quizzes as of this morning. 

On this exam, here’s how promotion and relegation broke down:

Annihilating a few examples

A number of questions had interesting results, but among them, three stood out as both worth discussing in depth and unlikely to be promoted to an exam. 

No Strings on Me

Our first question is Q2863, a rules question on the Standard Update. This question’s pass rate is 47%, so I wanted to dig in deeper to understand why judges were having trouble with this question.

Andrea resolves Marionette Apprentice, announcing she is putting a +1/+1 counter on it when the enters-the-battlefield trigger resolves. In response to the fabricate trigger, Nate activates the last ability of Barbarian Ring to deal 2 damage to Marionette Apprentice. What best describes what happens next?

1. Fabricate is a replacement effect, so Marionette Apprentice enters with the +1/+1 counter. Since Marionette Apprentice is a 2/3, 2 damage from Barbarian Ring is not enough to destroy it.
2. Nate can respond to the fabricate trigger by dealing 2 damage to Marionette Apprentice, destroying it. When the fabricate trigger resolves, since Marionette Apprentice is no longer on the battlefield, Andrea must create a 1/1 Servo token.
3. Nate can respond to the fabricate trigger by dealing 2 damage to Marionette Apprentice, destroying it. When the fabricate trigger resolves, since Andrea previously announced she was putting a +1/+1 counter on Marionette Apprentice, she is held to that choice, but since the Apprentice is no longer on the battlefield, no counter is added.
4. Nate can respond to the fabricate trigger by dealing 2 damage to Marionette Apprentice, destroying it. When the fabricate trigger resolves, Andrea can choose whether to add a +1/+1 counter or create a 1/1 Servo token. If she chooses the counter, nothing happens. If she chooses the token, she creates a 1/1 Servo token.

Ready for the answer?

If we served this question in 2017, I suspect that it would have a pass rate close to 85%. That’s because, at the time, this was a question that was coming up all the time at events. Indeed, Fabricate was so common that it was even called out in the policy update.

It’s strange that this policy article was written seven years ago, because I seem to recall Kaladesh just coming out last year, maybe two years ago at most. In any case, it’s reasonable for people to have forgotten the ins and outs of the mechanic- irrespective of how actually long ago it was. Fabricate never really got a toehold in Modern or Pioneer, and sees play only occasionally in Commander (before making this statement, I did a quick scan of my decks on Moxfield, and found that the Angel of Invention in my Giada deck is the only card with Fabricate that I’ve likely played in the half-decade before Modern Horizons 3).

With that all said, Marionette Apprentice looks like it’s a pretty good card. It’s showing up in  Yawgmoth, and it’ll conceivably appear on some Commander tables (I wish I could find room for it in Teysa). That doesn’t mean that Fabricate will rise to the level where players will expect judges to know it off the top of our heads, but it does mean that if you got the question wrong on the quiz, you hopefully read the explanation to either learn or refresh yourself on how it works.

No partial credit on partial fixes

Our next question is Q2368, a policy question on the Standard Update. This question’s pass rate is 42% – even lower than the previous question, and way too low for this question to appear on a policy exam as-is.

In a match against Noreen at an RCQ, Amos attacks with a 2/2 Bear token. Noreen controls a tapped Insect token and blocks the Bear with Priest of Titania. Before damage, she casts Revitalizing Repast on Priest of Titania so that the Bear dies and Priest of Titania survives. After combat, Amos plays a land and casts Wrath of the Skies with X=2, paying 2 energy, and Noreen puts Priest of Titania into the graveyard and puts the Insect token into her deckbox. In Amos’s end step, Noreen activates Polluted Delta, putting an Island onto the battlefield. On her turn, she untaps, draws, and resolves Deep Analysis, drawing two cards. Then Noreen realizes that Priest of Titania had indestructible and shouldn’t have been destroyed by Wrath of the Skies. The judge determines that both players have committed Game Play Error–Game Rule Violation, and issues a Warning to each of them. What is the best remedy?

1.  As a partial fix, Noreen returns Priest of Titania from the graveyard to the battlefield.
2. The game is backed up to the resolution of Wrath of the Skies. Wrath of the Skies is resolved correctly, destroying only the Insect token, and then the game continues from that point.
3. Issue Amos a Game Loss.
4. The game state is left unchanged.

I’ll start out by noting that what made this worth discussing is that A was the most popular answer here, with a couple people selecting B. I’m pleased that no one selected C. But clearly, a lot of L2s and L3s thought that the partial fix applied here.

Partial fixes are interesting. I like to think of them as “pre-approved backups” – while floor judges need to speak to the head judge or a team lead to get a rewind approved (or the choice to not rewind), any judge can generally perform a partial fix on his or her own, because there’s generally not much room for decision-making, so not much can go wrong. Did you forget to draw? Draw a card! Forget to order blockers? Do it now! I’ll note that this is significantly different from Flesh and Blood, where partial fixes are less prescriptive, but in Magic, we give you a small menu of “try this first, and if it doesn’t work, get help.”

The partial fix to move an object is the odd one out. First, because it’s the most complex. It reads:

If an object is in an incorrect zone either due to a required zone change being missed or due to being put into the wrong zone during a zone change, the exact object is still known to all players, and it can be moved with only minor disruption to the current state of the game, put the object in the correct zone.

That’s so many words, and all of them are important. Also important are the words that aren’t there, and that’s what tripped people up on this question. You can only use this partial fix to solve things that were supposed to move, not things that were supposed to stay where they were. If you’re confused here, you’re not alone! 

One great source to improve your policy is the Knowledge Pool archive – there are tons of great discussions of policy questions, and as long as you check to be sure the answer isn’t out of date, you can learn a lot just by reading the prompt, coming up with your own answer, then reading the discussion and seeing the solution.

Transform before Zof

Our final question is a policy question from the Advanced Quiz. Here’s Q2857, with a pass rate of 57%.  

During a deck check at a Pro Tour Qualifier, a judge observes that Alberto’s Zof Consumption is in its sleeve reversed, such that Zof Bloodbog is visible. Alberto’s sleeves are opaque. What infraction, if any, has occurred, and what penalty and remedy apply?

1. No Infraction. Remind him to reverse the card so the front face is showing.
2. No Infraction. No remedy is needed.
3. Tournament Error–Deck Problem. Issue Alberto a Warning, and remind him to reverse the card so the front face is showing.
4. Tournament Error–Deck Problem. Issue Alberto a Game Loss, and remind him to reverse the card so the front face is showing.

Ready for the answer?

This question received a lot of commentary, both before and after it was published. This question felt a little unfair to some test-takers, because it cites Toby’s blog, with answers not available in the policy documents! We decided to include it anyway – we want people to read Toby’s blog, and we’ve heard some reports of judges penalizing players for having the DFCs backwards or having to resolve issues that crop up from DFCs being flipped causing confusion.

I will note here, and I’ll even put it in bold: while Toby’s blog is fair game for update quizzes (and practice tests), questions whose source is solely the blog will not appear on advancement exams. You’re more likely to pass your exam if you read it regularly, but we’re sticking to the documents themselves for evaluation exams.

Riding off into the horizon

And that’s it for the Modern Horizons 3 Update Quiz. As always, I encourage you to leave feedback after taking a quiz – I read every single comment folks make on our questions, and I know the other exam team members do too. The Bloomburrow quiz is open and ready for L2+ judges to take it, so head on over to the exam hub and get started!