Game Week 5 Review: MSCHF, Madonna, and Maps that Changed the World
Introducing a new visual comparison of the relative difficulty across game weeks!
Last week I was all like "boohoo we got bad feedback that's why this review post is coming 4 days late". This week a lot of you decided to compensate by putting in some very kind words in the feedback forms and on the Whatsapp/Facebook groups.
To show our gratitude, here is a review post that's 5 days late 🙈
Links:
There’s some other stuff we’re covering in this post apart from the quads, so here’s that sorely-missed Table of Contents.
Table of Contents
Quads
About that Feedback Form
Thank you very much for the kind words you've sent this week. A lot of screenshots have been taken and they've been passed around in the Setters and Readers groups, and you've honestly made our week.
We also want to call out that a lot of the feedback we received this week has been more objectively useful. In the past, people have tended to use that form mostly to complain about seat balances, how questions were either too easy or too hard, and how they feel about questions changing mid-week. This is fine of course, that's what a feedback form is for, so you can tell us how the quiz made you feel. But you should also take the opportunity to tell us about specific questions and quads you liked, and topics you think we aren't giving enough attention to. This is stuff we cannot see ourselves in the data (as opposed to seat balancing), and so is inordinately more useful to us. Go ahead and request a quad if you want, we'd love to act on it!
Being specific also really helps. We can't read your mind, you need to tell us what questions didn't work for you. Let me give you an example. This is a part of the feedback we received:
"Multiple questions that had close answers but were simply not accepted - I understand that generic answers are taken but you have to measure the difficulty against the specific answers for other questions."
This sounds important, but without the specifics, we have no way of figuring out what it's referring to. This can be useful feedback, but in its current form we're forced to ignore it, cos you didn't actually tell us which questions bothered you.
This week however we received quite a lot more of the good stuff. People have named quads and answers that they enjoyed being able to work out, and listed the quads that they thought were abject trash. That's more like it!
Game Week Comparison
Look what we've got! Based on a recommendation from Sandeep Hari, resident magician Arpita Shetty was able to put together a nice visual comparison of how game weeks have faired relative to each other. This isn't a measure of anyone's performance (this league is bad at that sort of thing), but more a measure of how the setters are doing in their quest to give you questions that are accessible but not easy, challenging but not frustrating.
The chart shows the distribution of game scores in Game Week 5. You can see that most games saw between 36 and 50 questions getting answered, out of a possible 60. There were a couple of participating low-scoring and high-scoring games too, but that's the minority. The total of all the columns is naturally 50, since that's how many games we have in a week.
On its own, this chart is quite useless. But look what happens if you see the same chart for all game weeks together.
This beautiful thing shows the same numbers as the previous graph, but in a way that lets you compare game weeks to each other.
Immediately you can see that Game Week 1 (blue) played significantly easier than any of the other weeks. The right-side-heavy peak means that the vast majority of games scored really well.
This might also be because the draws are completely random in Week 1, meaning experienced quizzers are part of a higher number of games.
Another thing that's apparent is how much harder Game Week 4 (yellow) was than other weeks. We all felt this, the content was quite a bit of a departure for the setters as well, and we can see the price you've all paid as a result. That peak at 31-35 means it was a genuinely tough week.
Among the non-outliers, Game Weeks 2 and 3 played quite similarly, and Week 5 played a little easier than both.
Tell us if there's something else you'd like to see here. This Arpita says she has time.
1. Madonna's 'Tears of a Clown'
Aparna Kallakuri told me this is the second time we've done a Madonna quad in B612. She remembered us doing one on the lyrics of Vogue. I didn't remember this (our bookkeeping for these things is terrible) but I asked around and found out that a quad like that did feature in a recent friendly, specifically Shree Patwardhan's Set 9. She's sure it wasn't that though, which means we're still lost. Do you remember a quad on Madonna's Vogue? What quiz was that?
The difficulty gradient was a smooth one, with Clown maybe being a little too straightforward even for an L1. The original version of the question didn't have an image, and we probably should've kept it that way.
The Toxic funda is one we've seen before, in the pre-season friendly. Even aside from that appearance this "Britney samples Lata Mangeshkar" chestnut is an old favourite, one that seems to make an appearance in open quizzes every few months. I recently heard some people calling such a question a 'Peter', which is just messy slang for 'repeat' and possibly my new favourite word.
2. Games by MSCHF
B612's current categorization structure is bad at handling topics related to internet culture, which is why this quad is just marked as "Art".
There was a nice difficulty gradient, albeit with a too-easy L4, one that we probably should've seen coming.
Blue Donkey saw some interesting takes. People who know very little about American politics probably had an easier time guessing it from the question, compared to those who know a lot. This is generally a sign of a poorly framed question since the authoritative language used by an amateur enthusiast will usually confuse an expert more than it confuses another amateur. In this case, it was the phrase "party that is opposed to corporations", which most people read as "not Republicans, must be Democrats", but threw off those with a little more context on the nuance in party stances.
3. Discovery of King Tut's Tomb
That's right, 5 weeks into the season we finally have an unanswered question, which to me is an abject failure in question writing. I apologise. It seemed like a knowable term based on the little we read, but evidently, it was not.
4. Sairat and its Remakes
Look friends, look at this absolute sham of a difficulty gradient. This right here is a showcase of our bias as setters, cos we expected Sairat to play much, much harder than it really did. And everything else to play much easier.
I don't fully understand how Laila O Laila got missed, given that it seemed like a simple guess from the sample we played. Nor Channa Mereya, which would've been my first guess when faced with "Arijit Singh moon-themed song". But the data doesn't lie, so it's possible the questions were misleading or simply too hard to read for someone to make calm, composed guesses.
Let us know if you have any theories about why these played hard. Are they just not as popular as we were assuming?
🎯 Subrat M picked up a long-overdue musket in Season 4, scoring a perfect score in this quad!
5. Inspirations for Dora Milaje
A ‘world’ quad masquerading in a movie trenchcoat. The Dora Milaje featured in the last Black Panther movie, but knowing that doesn't help you get any of these, which is probably the way things should be.
Maasai and Samurai played well enough for their levels, as did Dora Milaje itself, but Dzilla gave us our second unanswered question of the league (and of the week).
As I've mentioned, unanswered questions are absolute mistakes, no question. This also means they're the questions that seem to most effectively anger our more aggressive participants. In the most entertaining bit of feedback, we received this week (a 350-word essay from a very disgruntled Seat 4, which is also the rating he gave the week), the player noted "There are simply too many iconic Japanese characters to pick from, and I assure you, Godzilla wouldn't feature in the top 10". Seriously, the whole thing is written like that.
6. 12 Maps that Changed the World
Mercator was everyone's "just missed" question of the week, and America played a little too easy for an L2.
Gall-Peters is actually a worthy L4 IMO, but it showcased a danger in adding oblique clues to questions that would otherwise be fairly difficult. The vast majority of people saw this question and passed in a hurry, not having enough context on map projections to even have a valid guess. But if you're an avid West Wing fan, that little mention of an episode featuring it would have you jumping for the name you do remember. You cannot resist, it fits too perfectly and it's as if the universe is telling you that all those hours of binging that show are finally about to pay off. Unfortunately, the West Wing episode calls it just "Peters Projection", which would've earned you a prompt in this quiz but not a point. This is incredibly annoying to you because now not only have you wasted a BA and dropped down in the passing order, but you've also been personally betrayed by Aaron Sorkin.
I heard from plenty of West Wing fans after their quiz this week, with Youtube clips and episode transcripts, no less. Fortunately, they were all extremely nice about it and nobody made too much of a fuss, but I'll go ahead and admit that ideally, we should've accepted "Peters projection" too, especially since we had added a hint like that. Mentioning someone's favourite TV show and then taking away a point is just cruel.
While we’re on the subject of maps, can you figure out what’s special about the project at the top of this section? It’s called an AuthaGraph projection, but don’t click on the link till you have a guess about why it’s cool.
7. Food Akbar Ate
Every once in a while a quad comes along that makes you wonder wtf the question setter was reading at the time. I love stuff like this because there's almost no way to actually prepare for them and they probably make a handful of people in the league very happy. "Hey, any quads to your liking this week?" "Yeah bro, Akbar's Dietary Preferences, fucking nailed it."
Played more like a non-ascending quad than it should've, with L1-L3 all having roughly even difficulty. OMAD played much harder and missed a prompt (or anti-prompt, as FLQL was calling it at one point) on "intermittent fasting", which is the more general description of the diet.
8. Hershey's Kisses
A fun quad, but broken down, in hindsight, by some questionable phrasing. While Pumpkin Spice and Candy Corn played out exactly how they were supposed to, the other two questions had some problems.
Kissables is a fascinating one. The question asked you to convert the "noun form to an adjective", which would have you scrambling for what the adjective form of ‘kiss’ could possibly be. The problem here is that while the 'kiss' in Hershey's Kiss is certainly a noun, its conversion to an adjective goes through an extra step. The players should've been reminded that 'kiss' would originally be a verb and only then converted to an adjective. Basically, you need to be describing an action and not a thing. This is a theory of course, but I think the question would've played it better if we'd just said 'verb' instead of 'noun'.
Remember how I said last week that Garden Of Earthly Worries was our most annoying question till then, but we were going to break that record soon? Well, this is it. Strawberry Ice Cream Cone, in spite of being answered more, has probably caused more consternation than any other question in the league so far. Working towards it slowly was the only strategy I saw people using that seems to work, i.e. attempt "strawberry" and then wait for a prompt, then attempt "strawberry ice cream" and wait for another prompt, and so on. Very definitely gaming the system, but hey if you thought of that, good for you! Points are well deserved.
9. Middle School Algorithms
Skip.
10. Pretender Athletes
Don't know about you, but I thought this was a very, very, VERY forgettable quad. Something about it just rubs me the wrong way. It doesn't know whether it wants to be a sports quad, a movies quad, a wordplay thing, or just a collection of interesting stories. The difficulty is all over the place, and it's not even specific enough to be a short list (there could've been dozens more on this list, some of which we'll never see now thanks to the inclusion of this quad).
Basically, this quads reads as though it was written in a hurry, minutes before the start of the first quiz of the week, which is not far off from what really happened.
11. Midnight's Children
This is more like it. I keep harping in these review posts about how 'Deep' quads (i.e. quads where 4 questions are extracted from a single subject) are better and more satisfying than 'List' quads (i.e. quads where 4 questions need with 4 distinct subjects that are part of a common list).
I stand by that here even when it's clear that the difficulty hasn't played the way we wanted it to (L3 and L4 were too easy), with the quad playing more non-ascending than ordered. It's still satisfying as an inclusion into a quiz and most of all, it makes me want to read the book, which is ultimately the goal that all quizzes should aspire to: make the participants want to explore new things.
Born at midnight on 15/8/47 had some people tripping up because the question asked for a common "trait", and so they went for a description of the children's powers rather than the actual attribute of sharing a birthdate. Clearer framing would've helped, although it ended up playing okay for an L1.
Indira Gandhi and Nanavati played easier than we expected them to.
The L2 is, in my opinion, a rare perfect question. Sharing it here.
Although written almost two decades before Rushdie met his to-be wife Padma Lakshmi, 'Padma' is coincidentally the name of Saleem's wife in *Midnight's Children*. Padma is the 'listener' while Saleem narrates the tale. This is a gender-subversive take on which story, that also helped Rushdie decide the number of midnight's children?
Answer: One Thousand and One Nights (Arabian Nights)A book described by its plot, a 'gender subversive' modification, and a clue about a number being important. So much going on. Someone give this question an award or something.
I asked Ananya if there was anything she wanted to add to this section, stuff that readers might be interested in about the making of this quad. Here's the best I could get:
“Apparently, a dog in Midnight's Children is named after JL Nehru. I didn't see this anywhere while researching, my Consti law prof said it in class.”
#TejujaCorrection: Ananya’s prof was misremembering. It’s actually a bulldog in Moor’s Last Sigh that’s named after Nehru.
It isn't much, but if it gets you a point in a quiz someday, please let us know.
12. The Works of Sarah Polley
🎯 Vinoo S and Notts both picked up muskets in this quad, scoring a perfect 4/4.
13. Keigo Higashino Adaptations
Once in a while a movie comes along that sends question-setters in all leagues scampering for new quad ideas. Monica O My Darling showed up on Netflix a few weeks ago and FLAMES and B612 have both made quads inspired by it.
I suppose this quad wasn't just that movie though, since Drishyam 2 also came out only recently.
The difficulty levels were good for a non-ascending quad, but the one that dropped was also one of my favourite questions from the quiz, simply because it's one of those "staring at you in the face" things.
Rapurasu no Majo (English: '[BLANK]'s Witch') is a 2018 Japanese film based on Keigo Higashino's novel of the same name. While the plot initially seems to involve geochemistry (deaths due to hydrogen sulfide poisoning), the title character's name alludes to a transformative French figure from the world of mathematics. Fill in the blank.
Answer: Laplace
Rapurasu is the Japanese pronunication of Laplace . So nicely done. Italicizing the 'transformative' might've been a mistake though, since we lost a bunch of people to Fourier. I hate italics. Have I mentioned that before? I hate italics.
🎯 Naveen Giles scored a musket in this quad, and I bet he could do it without the italics!
14. Supermodel Music Videos
I regret sometimes that we didn't ask more detailed questions in our league signup form. Stuff like age, gender, education, occupation, city, hometown, favourite colour, cat's diet, etc. Sure it would be creepy, but we'd get cool insights like "80% of the people who got this question right were people with engineering degrees who now work at banks".
Can't do that now, so I don't know if the 3 questions in this quad that played easiest were also answered by younger folks in general. There weren't a ton of clues in the Pillowtalk, Blurred Lines, and Freedom! questions, so anyone scoring points on those would have to have heard the songs.
15. Confluence Cities
Skip.
Special Mentions
Shyam Shreyas and Kumar Shobhit both picked up 7 different L4s in their games!
Seat Averages
X’s, or direct questions missed by all 4 seats.
Seat 1: 3.5
Seat 2: 5.52
Seat 3: 5.98
Seat 4: 5.36
Owns, or direct questions answered by each seat.
Seat 1: 6.2
Seat 2: 5.22
Seat 3: 5.0
Seat 4: 4.96