Table of Contents
- The Dataset
- Just For Fun
- Okay, Some Competitive Value
I’ve got some lighter fare for y’all this week. Our Marvel Snap Tracker collects information about your collection for the sake of syncing it to your account (here and/or at MarvelSnap.Pro). Doing so is really useful! It makes it easy to track your own collection progress and also lets you limit our deck database to only show decks where you own every component card.
It’s data I haven’t explored in Snapalytics before, so I thought I’d check it out. Most of what I found will fall squarely in the just-for-fun category. Don’t worry if you happen to hate fun — I’ve got you, too. I found one set of collection-based data that’s actually relevant to competitive play.
As always, I’m kicking us off with some info about the data set we’re looking at.
I wanted to make sure the data came from recent and active players, so I started with ranked matches from the past two weeks. Where I needed collection data, I included the collection of any player who recorded at least one such qualifying match.
Just For Fun
What avatars are used the most? For this, I actually looked at the avatars of the players’ opponents. I figured that would provide a much wider data set.
After removing duplicates, we observed almost 700,000 combinations of player and avatar!
|Rank||Character||% of Combinations|
We can see some trends! My observations are based on reasonable assumptions because our tracker only observes the character of an avatar, not the variant.
- Many people never change their avatar or else enjoy the faceless Default avatar
- The avatars that everybody starts with (Aero, Cyclops, M'Baku, Shang-Chi, Abomination) are also quite popular — not surprising considering their wide availability
- Rewards from recent seasons (Gamora, Goose, Cosmo, Wolverine, Hit Monkey, etc.) are popular
- Some bundles have been very popular (Groot, Deadpool, Rock, Mystique, Captain America)
We observed 179 unique avatar characters. Rock’s 13th place means it’s more popular than 166 characters! Second Dinner was onto something with their April Fool’s bundle this year.
|Rank||Character||% of Combinations|
|175||Jane Foster Mighty Thor||0.12%|
The bottom of a ranking can be really informative, too!
Kitty Pryde’s low rank is surely because her avatar came out in the middle of the past two weeks. I imagine it will rise dramatically, considering all existing players got her avatar for free.
Punisher is outrageously low compared to the rest at the bottom. It makes me think that the Punisher avatar from the Heroes for Hire season during Marvel Snap’s beta is the only Punisher avatar in the game.
It also seems like there are about 37 characters who have cards in the game but not any avatars. One of the missing ones is Stature — which is a real bummer to yours truly, as she’s the #1 avatar I’ve been hoping to pull since she was added to the game!
While it’s on topic, I’d like to share my 2¢ on avatars. Collecting them is boring (the rate is so slow) and disappointing (there are so many I’ll never use). I say remove them as a separate collectible entirely. Instead, players should be able to set any card in their collection as their avatar — including the ability to show off variants and splits! How cool would that be?! Please, Second Dinner? 🥺🙏
Card Popularity — Wide
How often a card has been upgraded to Infinity is a great measure for how well-liked cards are. There are two major reasons for this. First, you have to actually use a card to earn the boosters to upgrade it over and over again. Second, upgrading is 100% in the player’s control. Compare that to something like card acquisition, where even the rarest cards can be acquired randomly on the collection level track. So I’ve heard, anyway.
We’re looking at Infinities instead of splits because it’s so easy to forget to split.
In fact, there were 13,689 cards in the data set that had been upgraded to Infinity but not split yet. Make sure to occasionally sort your collection by Upgradable (descending) to catch these! Any card that can be split will be listed at the top.
Because most cards are only owned by a portion of the player base, we’ll divide the total number of Infinities by the number of players who own the card.
|Rank||Card||Infinities per Player|
The top 6 represent Series 1 and 2 cards that continue to be good even after completing Series 3. Sunspot’s dominant lead is certainly from the time he spent as the premier 1-drop — I’ve taken him to Infinity six times myself!
Mystique, Aero, Zabu, Doctor Doom, Death, and She-Hulk are the Series 3+ cards among the top 20. For the most part, they’ve always been good. Aero is much less good since her ability got nerfed, but before that she was a contender for best card in the game.
|Rank||Card||Infinities per Player|
|204||Ronan the Accuser||0.02|
Howard the Duck and Kitty Pryde are sitting at the bottom currently, but I’ve excluded them because they’re so new. I expect both will rise quickly, especially Kitty. By comparison, Iron Lad released a week earlier and has already shot up to rank 171! Admittedly, he was an unusually popular and good release.
My biggest takeaway from the bottom 10 is that Quake is so slept on. Y’all are missing out. 😤
Card Popularity — Deep
Time for a twist! Instead of dividing Infinities by the number of players who own the card, we’ll divide by the number of players who have upgraded it to Infinity at least once. This serves to pull up cards that are more polarizing. They are only loved by a portion of the player base, but that portion loves them a lot.
|Rank||Card||Infinities per Player|
See how “Infinities per Player” has grown? We’re requiring players to have upgraded the card to Infinity at least once, so the lowest possible score is 1.00.
There’s a lot that’s interesting about this table, especially compared to the wide top 20. Cards on both lists are both widely and deeply loved (or, at least were for some time — RIP Aero).
The ones that rocketed up the list are the ones that are not upgraded by many players, but are upgraded many times by the players who do upgrade them. There aren’t too many surprises here. Agatha taking the top spot doesn’t surprise me at all. The people who play her do so a looot. Galactus and Thanos aren’t surprising either. Both enable unique archetypes, and their fans are dedicated. I mean, Galactus is the polarizing card!
Groot and Ghost are the biggest surprises to me. Ghost is almost negligible. Only 28 players in the data set have ever upgraded Ghost to Infinity. The rest of the top 50 are each represented by at least a thousand players (with an average around 5,600). Groot’s story is interesting. 86% of his Infinities were represented by the two adorable Groot variants from recent bundles. Behold the power of Baby Groot! This also must be responsible for his #7 spot on the avatars chart.
|Rank||Card||Infinities per Player|
|212||Ronan the Accuser||1.08|
Howard and Kitty are again excluded due to their recency.
Being on this list is most damning for cards with wide release. Namor is in Series 1. He is rarely upgraded to Infinity — roughly 6.3% of players who own him have upgraded him to Infinity at least once. Those who do upgrade him to Infinity, rarely do so again (with an average of 1.11 Infinities).
Card Popularity — Wide Pace
A card’s age and availability (e.g., Season Pass vs. 6,000 tokens) are both big factors affecting how many players own it and how many times it has been upgraded to Infinity. In other words, players who have been playing for months have had a lot more time to upgrade America Chavez than Iron Lad.
Of course, the new shiny will always enjoy some time in the spotlight. Well… almost always.
It’s hard to account for seasonal popularity. We can make an attempt to consider the age and availability of a card, though. It gets too complex to directly include how long each card spent in each series. Even more complicated if we consider cards that were available from the Season Pass or in a bundle. An easier way to approximate all that is to rank the cards.
We rank them by total Infinities and rank them by Infinities per Player. We then adjust those rank numbers and look at the gap between them. The size of that gap gives a hint at how much Infinities per Player is outpacing total Infinities.
There’s a transformation I glossed over above. Skip this paragraph if you don’t care about the math. Low ranks are much more volatile than high ranks (e.g., there are 2,800 Infinities between Sunspot and America Chavez at the top of the list, but only 11 between Shanna and Snowguard at the bottom of the list). To reduce the effect of this volatility, we take the square root of each rank before getting the difference.
The resulting score is abstract, but I’ll include it so you can get an idea of how much more popular one card is than another. It’s not perfect, but I think this popularity measurement is the closest at capturing how popular a card is among its owners.
|1||Jeff the Baby Land Shark||3.52|
|18||Negasonic Teenage Warhead||0.71|
Just by a gut check, this feels pretty right, right? I’ve upgraded Jeff to Infinity three times. 🦈🦈🦈
The big bads take spots #2–4. Nothing can stop Jeff from moving to rank #1. For real, though, the land shark’s lead in this category is huge. He takes rank #180 for Infinities and #98 for Infinities per User.
All of the season pass cards from the last seven seasons are on this list. For the most part, they’re good cards. That alone isn’t enough to make this list, though. My guess? It’s because everybody wants to play with each season pass card while it’s the new hotness, plus the pass includes a bunch of boosters for the card.
Stegron is the only card on the list represented by fewer than one thousand players (only 337).
86 of 215 cards have a per User rank that’s outpacing their Infinities rank (a positive indication of popularity). 10 have matching ranks. That leaves 119 with a negative “Pacing Score”.
Howard and Kitty didn’t need to be excluded from this one. Even with tiny numbers, they outscore about half of the cards.
Who wants to bet Hulk sees a nice boost in the coming weeks?
This passes a gut check for me, too. We see cards that almost every player owns but don’t see much play. When’s the last time you saw an Okoye, let alone one upgraded to Infinity?
I am surprised to see Discard (Swarm, Apocalypse, Blade) so low. There’s also Sword Master at #198, Morbius at #201, and Lady Sif at #202. Compare that to how well Discard does with win rate and cube rate and you’ve got proof that it continues to be an underrated archetype.
It’s worth noting that everything from #132–215 is from Series 2 or earlier. These cards take a hit because they are owned by newer players who haven’t had the time to upgrade them yet. And maybe that explains Discard’s low position best — it is not widely recognized as a good archetype among players still in the early game.
#131 (the lowest-ranking Series 3+ card) is Ronan, by the way. If a trombone emoji existed, it would go here. Wah-wah.
Card Popularity — Deep Pace
I wasn’t going to include this one, but I feel like the most ardent fans deserve to have their voices boosted.
This is a combination of the previous two approaches: we’re requiring that a player has upgraded a card to Infinity at least once, and then calculating the Pacing Score.
This category captures card popularity among its upgraders. It’s best at showing which cards have the most intense fans.
|9||Jeff the Baby Land Shark||4.47|
|11||Negasonic Teenage Warhead||4.25|
Ah, here are my people. The Quake and Valk fans. There are dozens of us and I love each and every one of you.
Valkyrie is the card I’ve upgraded to Infinity the most — eight times!
Snowguard makes the cut with a whopping seven players who have taken her to Infinity at least once. Seven players who I wouldn’t want to meet in the dark.
And there y’all go breaking my heart again — I love Jessica! 💔
Hit Monkey being down here is fascinating. Diving from #10 on the last popularity metric to #212 on this one. It means tons of players who own him have taken him to Infinity once, but hardly any of them took him to Infinity again. Oh, hey, that includes me! Exactly once.
Variants are a huge part of what makes one player’s collection different from another!
First I looked at the most commonly owned variants. 33 of the 37 most common variants come from giveaways, season passes, and bundles.
I looked at the top 100 most common variants that didn’t come from giveaways, season passes, or bundles. Here’s the breakdown:
The remaining 34 are just labeled “Variant”.
|1||Shocker||Steampunk (Variant Rush)|
|2||Yellowjacket||Steampunk (Variant Rush)|
|3||Bishop||Steampunk (Variant Rush)|
|4||Falcon||Steampunk (Variant Rush)|
|6||Sentinel||Variant (Alan Davis)|
|7||Nightcrawler||Variant (Travis Charest)|
|8||Ebony Maw||Winter Vacation (Variant Rush)|
|9||Rogue||Winter Vacation (Variant Rush)|
|10||Jessica Jones||Variant (Alex Maleev)|
|11||Patriot||Winter Vacation (Variant Rush)|
|12||Abomination||Winter Vacation (Variant Rush)|
|13||Rockslide||Winter Vacation (Variant Rush)|
|15||Iron Man||Variant (Salvador Larroca)|
Here’s a weird discovery: Cerebro was a lot rarer than the other cards in the Steampunk Variant Rush — about 3.2 times rarer! Contrast that with the Winterverse Variant Rush, where all five included cards appear to have been equally rare.
Variant Popularity — Wide
Again, we’re looking at Infinities because it’s something that’s entirely in player control. The data set has 800 unique variants, and many are owned by a small number of players. We’re gonna keep it simple and focus on Infinities per Player. First, looking at all players who own a variant.
|2||Kim Jacinto||Shang-Chi||Super Rare||2.38|
|3||Chibi||Agatha Harkness||Super Rare||2.25|
|4||Savage Land||Devil Dinosaur||Bundle||2.14|
|5||Luca Claretti||Mystique||Super Rare||2.11|
|9||Variant (Mico Suayan)||Wave||Season Pass||1.97|
|10||Variant (Jee-Hyung Lee)||Iceman||Super Rare||1.89|
|11||Variant (Guillaume Singelin)||Devil Dinosaur||Super Rare||1.83|
|14||Artgerm||Doctor Doom||Super Rare||1.75|
|15||Neutral Zone||Blue Marvel||Super Rare||1.75|
|16||Ryan Brown||Lizard||Super Rare||1.74|
|17||Variant (Marco Checchetto)||Magneto||Super Rare||1.73|
|18||Alex Ross||Doctor Doom||Super Rare||1.73|
|19||Variant (Stephanie Hans)||Angela||Super Rare||1.71|
I think I’m most surprised that there aren’t any Ultimate variants that make the cut. Considering we’re looking at Infinities per Player, a card with few owners could totally appear on this list.
The rarest card on here (for now) is the Wave — she’s from a season pass during the Marvel Snap beta. I say “for now”, because it appears that variant will be available in a bundle at the end of June.
Seeing four cards from bundles on the list actually isn’t so surprising. Anybody who owns the card bought it on purpose. You’d hope that means they like it!
Variant Popularity — Deep
We’re still looking at Infinities per Player, but this time, we’re requiring that a player has upgraded the variant to Infinity at least once.
|2||Chibi||Agatha Harkness||Super Rare||3.49|
|5||Kim Jacinto||Shang-Chi||Super Rare||2.85|
|8||Luca Claretti||Mystique||Super Rare||2.62|
|10||Savage Land||Devil Dinosaur||Bundle||2.50|
|16||Luca Claretti||Agent 13||Super Rare||2.26|
|17||Variant (Guillaume Singelin)||Devil Dinosaur||Super Rare||2.21|
|18||Variant (Dan Mora)||Sunspot||Rare||2.20|
|19||Variant (Mico Suayan)||Wave||Season Pass||2.20|
|20||Variant (Arthur Adams)||Shang-Chi||Rare||2.19|
There are some real surprises here! The biggest of which are represented by just a handful of players. Ghost, Helicarrier, and Mantis have been upgraded to Infinity by 3, 6, and 8 players, respectively.
Shang-Chi impressively holds three spots in the top 20 — as many as Agatha!
We did get an Ultimate appearance! I’m not surprised to see it’s White Queen that tops all the Ultimates. I am a little surprised that White Queen’s Ultimate variant outperforms her Artgerm variant, though.
Okay, Some Competitive Value
Time to get down to brass tacks!
I have found one collection-based metric that is actually relevant to competitive play. I think it’s most useful for informing decisions about what cards to purchase with your tokens (or which free Series 3 card to claim next).
What I’ve got is like the meta share metric you can see on our card stats page. Except instead of just looking at how often a card is in the player’s deck across all games, we’re gonna look at how often a card is in the player’s deck in games where it was possible. In other words, we’re gonna require that the player owns the card.
When you’re considering what cards you want next, it can be useful to know how popular a rarer card is among the players who own it. A card’s rarity doesn’t make it worse. Shuri was still a monster when she was in Series 5 — she just didn’t become completely meta-defining until she dropped to Series 4 and became more accessible. Sometimes, a card benefits from being rare because it’s not common enough for the meta to try countering it.
Owned By % represents the percentage of players who own the card.
Usage by Owners represents the percentage of games played by a card’s owners that they brought that card in their decks.
|1||Jeff the Baby Land Shark||13.77%||43.99%||5|
|9||Howard the Duck||1.71%||21.53%||5|
Of course, new cards are always gonna see a boost in usage while they’re new and being tested out by players.
But notice that Stegron and Snowguard are nowhere to be seen. Their shininess has worn off already, and they weren’t good enough in the current meta to keep their usage up even if you limit the scope to players who own them. Stegron has received mixed reviews and Snowguard has been universally panned.
The “ooh, shiny” effect can be seen with the great numbers Howard and Kitty have posted. They both made the cut even though they were available for a little less than half of the time covered by the dataset (I accounted for ownership but not release dates). If I ran this again covering just the time since their release, I expect their numbers to be even higher.
What the heck — let’s do it.
Indeed, Kitty’s ownership rose to 100% and her usage rose to 19.28%. Howard’s usage by players who own him is at 24.39%.
It might seem weird that Howard’s usage is higher, but consider how each card is acquired. Everybody got Kitty for free while most Howard owners paid 6,000 tokens for the privilege. Players with Howard are a bit more invested. 🤏
Back to our bangers. Jeff and Nebula have been out long enough that their continued success on this chart is meaningful. They represent a very flexible 2-drop and the new premier 1-drop. Iron Lad is newer so it’s harder to say definitively, but his numbers look well beyond what you get from a simple novelty boost.
I believe some popularity can come from a card having a well-loved character (like the adorable Jeff), but for the most part, most people build decks and develop card preferences based on a card’s function.
I think Jeff, Iron Lad, and Nebula are all great Series 5 purchases. Which is best? I lean towards Nebula. Ownership is an important metric to consider, too. To me, Nebula’s 37.51% usage by owners is more impressive than Jeff’s 43.99% usage by owners because she is owned by 4 ½ times as many players. Jeff and Iron Lad may be better in specific situations, but Nebula is better in general.
Speaking of their ownership, I am surprised by how low Jeff’s ownership is based on my own gameplay experience. Yet his 13.77% is probably about as high as a series 5 card will ever reach without being a big bad or season pass reward.
Comparing Jeff’s ownership to Nebula’s (62.38%) and Hit Monkey’s (65.14%) is very telling. I personally think ~14% is way too low of an acquisition rate for a popular Series 5 release. Like so many others, I’m not a fan of many of the recently announced changes to card acquisition, but I am glad that more cards will be released directly to Series 4.
Look, I’m two cards short of a complete collection. I’ve obviously put hundreds of dollars into the game. I don’t ever want that to be why I win. I don’t want card acquisition to be a problem for anybody! I want to see card accessibility go way up, especially for new cards.
The data explored in this article supports the idea that while players appreciate great cosmetics and rare variants, a card’s functionality is the primary factor influencing how often players use it and how many resources they pour into it.
However, it also shows us definitively that cute sells. Jeff and Baby Groot prove it. There’s no doubt Second Dinner looks at these kind of metrics. Which makes it a safe bet that similarly adorable characters and cosmetics are on the horizon.
I think it’s important that all of us keep collections in mind as we discuss the competitive side of Marvel Snap. Card acquisition is a hot button issue for a good reason. All players benefit when creators cover Series 5 cards — even if you don’t have them, your opponent might — but it’s good for creators to share cheaper alternatives when possible (many are already great at this). In addition, gauging the popularity and success of a card while it is in series 4 or 5 can provide insight into its long-term value and the way the meta might evolve as the card becomes more accessible over time. Of course, not every card will significantly impact the meta.
I hope most of you found it as fun or interesting as I did to dig into some game data that’s not relevant to competitive success. Snapalytics will typically focus on competitive data, but I would like to dig into fluffier data every now and then! Let me know what you think about that in the comments or hit me up on Twitter. 💜
This analysis was powered by the data from our Marvel Snap Tracker. If you play on PC or Android, consider giving it a download! It has many features including a live overlay, the ability to sync your collection to our site, letting you review your personal stats, and powering useful tools like the Meta Snapshot.