Table of Contents
Welcome back to this week’s Developer Update! This week’s edition comes after the release of the newest card
“This was my last week at Second Dinner. I loved working on Marvel Snap, and learned so much! Thank you to my peers and our players. Next week, I’ll join Fanatics Collectibles as Sr. Director of Game Design. I’m excited to meet more of the team and start building some games!“
With Glenn no longer answering questions, it’s unclear how many of your questions will be answered going forward. If another developer steps in to answer many of the questions Glenn would, these updates may continue was normal. If not as many questions get answered, these Q/A recaps will be put on hold. For now though, we still have the last set of questions Glenn has been able to answer.
If you don’t know if Selene Horseman of Famine is worth Snap Packs or tokens, make sure to check out our Bonus Challenge guide for a post-release review of the card and new decklists!
Answers and questions may be slightly rephrased for more clarity and ease of reading. This week’s topics will be divided into Card Specific Questions and Other Questions.
See all individual questions and answers updated in real time!
Card Specific Questions
Q: Will Kraglin’s conditional bundles variant become avalible again?
A: Griffin “I’ve definitely let the team know that there’s a desire for “pre-Kraglin” bundles to become available again to players. The team’s aware or this desire but no hard plans or dates for their return right now.”
Q: Could you change Galactus to banish cards at destroyed locations instead of destroying them? This would bypass the issue of cards like Colossus being destroyed despite his text.
A: Glenn “He could work this way, but he hasn’t and we don’t generally make changes because we could–we like to have a direct motivation that would make the change an improvement. All Ongoing abilities require the card to be at a location to function, so there’s an existing rules precedent for why Colossus can be destroyed this way–his location is gone, disabling the effect.”
Q: With so many cards that ramp energy, why can’t Arishem go back to his old state before his nerf?
A: Glenn “Even if Arishem was at some major deficit, it hasn’t even been 6 months since Arishem got second in the Golden Gauntlet World Championship. It’s OK for decks to ebb and flow in strength over time—we rely on that to keep the game fresh and dynamic. There isn’t some ideal state we’re moving every card or deck toward.”
Q: Can you at least make sure Arishem stays a viable card since he is so unique?
A: Glenn “Wish granted–the Arishem deck’s winrate is over 50%, and its cube rate is positive.”
Q: With Rocks being a supported card/archetype now, could we see more styles of rocks in the future?
A: Glenn “Maybe! We actually debated making Luna’s token card an Ice Rock.”
Q: With Thor now recieving his 3rd buff, why not buff Mjolnir like you did Stormbreaker?
A: Glenn “We generally prefer to buff cards directly. We buffed Stormbreaker in the past because buffs to Bill are worth more than 1 Power so they’re not as straightforward. Bill’s the unique case here, not Thor.”
Q: Has there ever been any consideration to change Mjolnir and Stormbreaker, so each weapon affects both Thor and Bill, instead of just their respective characters?
A: Glenn “Not really. That would raise the ceiling of each card considerably higher than a couple points of Power, which we’d likely wind up compensating for in some other way like making the floor worse. It tends to be more effective–or at least more measurable–to buff the floor.
Q: How did the team shift from Star-Lord not being a significant problem to nerfing him to a near-unusable state? It feels like the card was over-nerfed until a better balance decisions could be made.
A: Glenn “I’m not sure how you’ve arrived at the conclusion that Star-Lord had no significant issue. The card has received 2 nerfs about as rapidly as possible, and we aim to avoid double-nerfing as a practice.
When we are forced to take two shots, we want to be sure the second one will land. The purpose of any nerf is first and foremost to effect actual change within the metagame, not make the smallest change, so we swing bigger on #2.
We also learn more from a harsher change, because knowing where ‘too far’ is with certainty can better define the acceptable space. This practice has a variety of names, but Dan Felder likes to call it ‘triple tapping’ here: https://www.gamedeveloper.com/design/design-101-balancing-games. Dan’s got plenty of great readings if you’re interested in game balance.”
Q: With En Sabur Nur acting very similarly to Lockjaw, why can’t Lockjaw go back to his old version that can cycle multiple cards?
A: Glenn “Drawing multiple cards and getting lots of free Energy from one card is fun. It’s also strong in a warping way.”
Q: Why is Kang still banned from limited time modes despite his rework?
A: Griffin “His inclusion has been vestigial from before his re-work and we considered removing it alongside Mysterio but we decided to holdback and do some testing first. If all goes well he’ll hopefully be off the ban list in future iterations.”
Other Questions
Q: With such a large number of variants getting released, when will we see variants added for the small pool of cards that don’t have any variants?
A: Griffin “The art team is aware of the need and desire for Variants for cards who may have none of very few. The team is making efforts to try to fill that gap but the lead time for Variant can be long as well as finding the right fit for an artist, their style, and a character.”
Q: How does someone become an artist for the game’s variants?
A: Griffin “Well, first it starts with putting in the hours and becoming and incredible artist in general.
In regards to Marvel Snap, we often commission artists for original work in addition to using existing pieces from Marvel’s extensive catalogue of comics.
Our art team is always on the look for new artists who they feel like can bring something new and unique. Feel free to DM me for more info, if you’d like.”
Q: Why don’t new game modes preserve the core gameplay of “snaping”? And why aren’t snaps designed for shorter gameplay?
A: Glenn “While we’re called Marvel Snap, the Snap itself is not what we consider our core fun, and the success of our game modes in some ways proves that. Our core fun, as defined prior to me even joining the team, is that Snap is all about the 3 locations–which 2 you’re trying to win vs. which 2 your opponent is trying to win. I think all our game modes preserve this, and some even enhance it.
You’re also conflating different things–games and sessions aren’t the same unit of measurement. But even so, most of our game modes do average shorter games and/or sessions than ranked. Some don’t, which is also fine–the purpose of a game mode is to diversify the available gameplay. Shorter for shorter’s sake is not a sensible goal; we’d just give everyone 21 Energy and a single turn if that was the case.
Conquest is a uniquely implemented chunk of the game. The team has considered updates to Conquest, but those are currently triaged behind higher-priority features, like Character Mastery and Draft. I’m optimistic we’ll see Conquest improved in the future though, and I agree that match length in that game mode is one of the core issues I’d hope to adjust.”
Q: Why are some bundles showing different prices for different regions?
A: Griffin “The team is aware of some pricing discrepancies across platforms in certain regions/currencies. They’re currently conducting a full review. We apologize for the confusion.”
Q: Is it hard to get good data on playrate when the card pool is constantly growing?
A: Glenn “No, not really. It’s fairly straightforward to scale play rate thresholds, and we don’t make so many cards that it’s especially burdensome to review the data.”
Q: Do you track how many hours or games each member of Second Dinner plays across different departments?
A: Glenn “We don’t track that value, as it’s not a useful number. It’s a ton of people, and every team has different reasons to use the app. Plus we have several different playtesting branches in development across all those teams, with internal accounts being constantly created, used, and discarded. Just today I played 3 different versions of the game myself, and I didn’t play any live games.”
Q: When players have game crashes, can you create a trace function to see why players games crashed to begin with?
A: Glenn “We have that. If you report a bug to us, it creates a backtrace that can be used to see the most recent activity on your account and identify what was happening prior to the error. This is why we stress to always report a bug in-client, even if the game or crash is over, because we can investigate activity earlier once we have the approximate timestamp.”
Q: Do you think it’s problematic to discuss card balance, when actual balance changes may be different?
A: Glenn “no—I don’t believe there’s any issue with honestly discussing our balance practices. I’ve always enjoyed doing it and adding detail where I can. But also… we pretty much only nerf cards that data confirms we should, so when you ask ‘why no nerf?’ a lack of sufficient data supporting that nerf is implicit in the premise.”
Q: What would you say is the most important statistic when deciding what to buff and nerf? And is there a minimum amount of games necessary to nerf a deck?
A: Glenn “I’ve covered this many times in the past. We look at the play rate, win rate, and cube rate of individual cards, as well as groups of cards that often see play together. There’s not a single statistic that’s always the important one, because context changes the information.
Nor do we use flat thresholds—for example, the play rate of Shou and Gambit HoD should not use the exact same thresholds, they’re different enough cards to evaluate differently.”
Thanks for all your questions!
That’s all for this week’s edition. While the future of Snap’s Q/A is unclear, I’m holding off on questions this time. If there continues to be enough questions being answered going forward, we will go back to the weekly updates. Until then though, thank you all for tuning in each week!
More Content