How streamers pick games: the SplatterCat interview

[The GameDiscoverCo game discovery newsletter is written by ‘how people find your game’ expert & GameDiscoverCo founder Simon Carless, and is a regular look at how people discover and buy video games in the 2020s.]

Greetings, my lovelies, and welcome back to a brand new week of informational goodness from this here GameDiscoverCo newsletter. And it’s been a delight to see feedback from the latest newsletter about Steam wishlists and sales predictability.

In particular, Patrick Seibert Tweeted one of our conclusions: “Wishlists are a little bit predictive of how you may do at launch, but not a lot”, and added: “Please repeat this sentence every morning so the wishlist craze can finally end. Thank you.”

And Farbs is the second person to reference something called Goodhart’s Law regarding wishlists - “When a measure becomes a target, it ceases to be a good measure.” And it’s not like you should ignore wishlists. But if you concentrate on game hook, iteration, and using your community to help you improve the game - AND get lots of wishlists - you’ll be in better shape than if you use solely wishlists as a success metric.

[ALERT: only 4 days left on the 20% off deal for new members of our GameDiscoverCo Plus paid subscription, thanks to adding Derek Yu’s Spelunky eBook. Exclusive weekly newsletters, a Discord, our Steam Hype data, a ‘Game Discovery Toolkit’ eBook & more are your prize.]

Interview: SplatterCat on how to pitch, YouTube vs. Twitch, more!

As it happens, a recent newsletter dealt with a ‘high Y1 multiplier’ Steam game that had done well in part because a couple of videos from the notable ‘variety’ YouTuber SplatterCat. And as we said at that time:

“We’ve talked about SplatterCat before, since he plays new games all the time - very rare for a YouTuber. In addition, he has 669k loyal YouTube subscribers, is a fan of systemic & replayable games, and has been a positive factor in success stories for games we’ve covered like Nova Drift.”

Coincidentally, we’d just been introduced to Matt/SplatterCat by a mutual acquaintance, and he was kind enough to answer a bunch of ecosystem and dev-centric questions via Discord DM. Given that all of his videos get 50k-100k views or more from highly engaged players, it’s worth hearing how he sees games - and deals with the Herculean task of releasing a video on a brand new game every day:

You are rare as a YouTuber with a large following who doesn't (often) make multiple videos on the same game, and deliberately favors new games. Why don't many others follow your path? Is it because familiarizing yourself with games takes time, streamers find it difficult to work out which games to play, or other reasons?

I think it's a bit of a case of ‘big fish -> small pond’ in this case. I was an early adopter with indie games back in 2012, where I really fell in love with the creativity of the genre. I noticed that indie games were really breaking away from the tired AAA standard of mass appeal, and focusing on very niche gameplay ideas - because they lacked the massive overhead recoup that AAA development requires.

That being said, indie games are still very much just hitting their stride as a medium, and becoming more popular by the day. Despite that, it's still quite new as a genre and I think the consumer audience is smaller because of that. If you had told me an indie game video on a niche ‘castle defender’ game would pull 100k views 10 years ago, I wouldn't have believed it.

In those days, anything that wasn't Spelunky / Minecraft / Binding of Isaac wasn't really putting up views, because the general public awareness hadn't crystalized yet - so they weren't looking for it. I'm approaching ten years doing this job next year, and it's important to note that I started very small. I was doing Let's Plays which were very much a trend at the time, and that allowed me to gather a solid following that were consuming my content.

Around 2017, YouTube made some fairly sweeping changes to the way the algorithm and site functioned, swapping from a browse / search placement priority to a system which favored highly viewed and highly clicked on content by users. Let's Plays thrived very specifically because of the prior algorithm. So if you could get your content placed high in search by being the first to post, you were gravy and you could finish out the series - while coasting on the views from the first video to pay the bills.

I was warned about this happening through peers and felt this happening, so I pivoted my content to something fresh - which was ‘indie impressions’ videos. My pre-existing audience dissipated a little bit with the swap, but I still had enough left over to get this content pushed to the greater site through the sidebar. It took a long time to pick up speed, and I think that 4-5 year [growth] runway is really what scares a lot of new channels off or burns them out outright.

Indies as a genre are still pretty small compared to the rest of the gaming world, so there's a limited audience searching for it. It's hitting its stride much more expediently nowadays but there's sorta still a lack of viewers to go around. What spaces exist have been filled up until there's another influx of people interested in indies.

So to answer your question I think it's that:

A) it takes a ton of time to get established in any type of content creation. The growth is very slow with variety content, unless you have either stunning critique or a welcoming charismatic presence. I started out with a pre-existing audience from my previous content type. So that helped move the channel strategically to where it is now. Smaller content creators or new ones won't have that initial boost to become visible. It's a very slow burn that tends to gas people out when coupled with the time commitment.

B) On the YouTube half of the business, scouting out games takes a good chunk of my time, and that's with all the connections I've developed over the years. Scrolling lists on Steam and Itch.io can be very daunting. I keep a sticky note on my desktop with dozens and dozens of candidates updated weekly. My Steam wishlist has roughly 150 games on it too. It's a big process. Then you've gotta learn the game and think about what observations / critiques to make before recording and editing, which is more time down.

How about YouTube vs. Twitch for game visibility and how indie games get seen on them? Is there a big difference between the two mediums, and how do you treat them differently?

When it comes to streaming [platforms like Twitch], the issue becomes one of visibility. Because of how Twitch works everything is ‘bubble sorted’ by total viewer count, and the top slots are almost universally taken up by the titans of the gaming industry - Grand Theft Auto, Call of Duty, World of Warcraft et al. Streamers tend to avoid playing the smaller indie titles (in my opinion) because doing so is going to put you waaaay down list and you're not going to have any viewer flowthrough.

There are rare titles like Darkest Dungeon that escape the atmosphere and break into the big leagues but those are… a rarity on Twitch. YouTube is much easier to push niche content on by far. I structure it such that I play a game on my YouTube and if I feel like the 30 minutes didn't fully flesh out the game for the user, it pipelines them to my [Twitch] stream, where I'll play the game for like 4 hours.

That way I can guarantee some viewer flowthrough, making the time expenditure worth it. It has the added benefit of supplementing my coverage on YouTube for those that want a lengthier, beefier chunk of gameplay to make up their minds whether or not a game is for them.

Larger streamers with a 4-5k [regular CCU] viewer audience can hit the smaller games too, because they're large enough to boost the game into the top 25 while they're live which will also guarantee some traffic. But since I'm smaller, it's more feasible to use YouTube as the seeding ground for my stream.

Once again, this is another breaking point where I'll point out that life is pretty unfair when you're a small content creator. Without that oomph it's gonna be tough to push content in a way that both grows your channel AND generates exposure for the title in question. That unfairness I think when it comes to sorting methods and algorithms creates a barrier for entry… [so] the big get bigger and the small sorta burn out, unless they're really in it for the long haul - on top of being incredibly talented.

When devs send you information and keys for a new game or demo, what are the top things they can get right in terms of how they send or present info to you? And what are some things that devs do that you don't find helpful?

So the first big thing is to have a banger presskit with all the materials needed for making thumbnails: HD wallpapers, key art, and logos in various sizes. Screenshots I usually don't use at all, but those are useful for people doing formal journalist articles. Since YouTube functions almost entirely on Clickthrough Rate in the sidebar, having a few keyart pieces that really show action, mystery, and vaguely ask a visual question with the layout are all key.

Thumbnails are incredibly underrated, but 90% of making a video stick is having a rad piece of art to show off. It can mean the difference between 15k views and the algorithm picking it up due to high clickthrough and pushing it out to hundreds of thousands. It really is that impactful.

On the ‘not helpful’ side I'd keep the emails brief, a couple sentences of introduction of both yourself and the game coupled with some teaser images, a link to the presskit, and a Steam link. I can usually eyeball and tell if the game is a good fit from just the trailer + thumbnails / art.

When an email pitching a game becomes overly verbose and takes a long time to read, I usually find myself just scanning for the requisite links/materials, because your game will absolutely speak for itself through the trailers and materials. I'd also urge people to be patient with coverage / followup emails etc. Most YouTubers are getting carpet bombed with emails 24/7.

Having the time to actually formally respond to each and every one is an impossibility while still getting the actual coverage itself finished off. We do our best to get back to you, but if we don't it's absolutely not personal. There's just a ton of stuff going on with our operation at any given time.

You explain your channel in your pinned YouTube video: "I rarely do negative impressions... I filter out the games I wasn't especially stoked about." How do you do filter, practically? How many games do you play for each one you feature, and what are a few of the things you look for?

After doing this for so long, I can do it visually pretty quickly. There are some games that creep on me and surprise me outta nowhere. But from gameplay footage you can usually get a solid feel for what care has been put into a game. I'm typically looking for a unified art direction, [including] animations that don't look stiff and have a fluidity or natural flow to them.

I usually avoid anything monochrome as a stylistic choice, just because a lack of color doesn't fly on youtube both in video and thumbnail format, regardless of the game's quality. I usually pause videos and look at the screenshots to quickly appraise the UI and figure out if they're going for something more artistic or minimalist. If I feel like I can't get a good feel from the assets, we go hands on for 30 minutes to an hour.

I look at every single Steam link that I get, and I've got it optimized fairly well at this point to get rid of wasted time picking through things that are clearly lacking in various regards. I probably play about 3:1 not picked vs picked on a normal week, and for a really busy week like a festival on Steam it probably swings up to 5:1.

The whole thing is a very delicate act of finding A) something that matches what my audience wants to see B) something that I think is high quality and needs the attention C) something that works for propagation / pushing out on YouTube (Note that this one I wish was not a factor, but the platform demands what it demands - despite our displeasure) and D) something that I personally find really fun. This listing isn't in any particular order. All those criteria are really important, and you've gotta have that right mix. Every choice is a pragmatic one trying to balance all the factors.

What trends in games have you seen recently that you like? Are there particular genres that are 'coming back', are some styles of game massively oversupplied, and are there any genres/subgenres of games you would like to see more of? What do you find most rewarding about taking an 'all new games, all the time' approach, and what is most draining or stressful about it?

I'm really into anything with carried progression. I like the idea of the actual game completion being locked behind skills that you slowly accumulate through failure. I think it provides a nice ramp for accessibility as well, since eventually you'll be able to stomp regardless of skill.

There's a ton of card games right now. I think in general most of us are feeling some level of card game fatigue lately. RTSes are on a big comeback, as well as Stronghold-style castle builder / defenders which I’m very excited about, since that's a genre that always gripped me and has been sorely lacking lately.

As far ‘as new games all the time’, it keeps things fresh. Work is never boring, and I never get trapped in that loop of playing a "main game" all the time that burns so many folks out. What I do is really a blast and I really look forward to going to work most days - which is a blessing.

Are there any other messages you'd like to give to developers (or publishers) about how they should interact with or treat their friends, the streamers?

Hang in there. In a lot of ways, content creators for YouTube / Twitch and developers have a ton in common. We're both ruled by algorithms we have no control over and sink or swim a lot of the time through forces largely out of our control.

As far as interactions go, I like to keep it professional, I can be stand-offish in a lot of ways through interactions just because, in the past, you get really close with a developer and it comes back to bite you in terms of favors asked, etc. I usually prefer to keep things strictly business-like nowadays, to avoid awkward conversations later on down the line.

The vast majority of developers I interact with are very kind and easy going people. Don't be worried about approaching for coverage - we're just normal folks. It never really set in that my channel had gotten larger for me mentally. I still do things simply like I did in the early days, and after I get done I'm probably gonna go take out the trash and do some dishes just like everybody else.

Where’s the Hype? Upcoming Steam card battlers!

Something we’re going to try to do more often is expose our ‘Steam Hype’ data to free newsletter subscribers. (For those who don’t know, we grab data on Steam wishlist ranking, followers, forum chatter & more to rank all unreleased Steam games!)

In this case, we thought we’d take a look at the ‘Card Battle’ tag, but for all unreleased Steam games - in part because of SplatterCat’s mention above that he feels like card games might be getting overplayed.

Looking at ‘Card Battle’ specifically - and card-based games have fragmented tag coverage, since there are other tags like Deckbuilding, Card Game & more - we still found some interesting things (above), as follows:

  • Looks like there are only six titles with more than 3,000 Steam followers (so, what, 30,000-ish Steam wishlists?) in this particular tag, out of 130 in total. It compares a bit less well to tags like Metroidvania (11 unreleased games with >3k followers out of 270 total) or the big daddy, Colony Sim (a whopping 28 games with >3k Steam followers, out of 144 total!) But still seems promising to us.

  • It’s always interesting to me to note the number of low-profile games (broad publicity-wise!) doing well in these charts. There’s higher-profile stuff in here - you’ve all heard of Firaxis/2K’s Marvel’s Midnight Suns. And then Voice of Cards: The Isle Dragon Roars is the new Yoko Taro (Nier) title which has shot up the charts since a surprise announce, of course.

  • But further down, were you paying attention to Daniel Mullins/Devolver’s Inscryption? Or how about John Bell/Critical Reflex’s tower defense/card mashup ORX? Or Thing Trunk’s co-operative deckbuilder Hellcard? One thing that we find these tag rankings helpful for is delineating the sheer depth of releases - even in these very specific tags.

There are obviously inaccuracies/weirdnesses in using data like this - overlapping tag types, user-created anomalous tags, wishlist intangibility vs. sales, etcetera. But it’s still a good & interesting lens to look at the amount and type of certain games in the Steam pipeline. And GameDiscoverCo will be working on some tools to better rank unreleased game tags, in the medium-term.

(Reminder: all the above ‘Hype’ info on unreleased Steam games is available to all paid GameDiscoverCo Plus subscribers - you can sort by any tag you like, export as CSV, and more.)

The game discovery news round-up..

Finishing up for this particular newsletter, we’ve got a whole bunch of notable game discovery news zooming into your line of vision since the last GameDiscoverCo newsletter, last Wednesday. So let’s take you through it, in this precise order:

  • So yes, Apple filed an appeal in the Epic lawsuit, asking for a stay on the injunction that lets developers add in-app links to payment websites.” This is basically a ‘give us more time to sort this out’ thing. Apple piquantly points out: “Epic will suffer no harm from a stay because, as authorized by the Court’s decision, Apple recently rejected Epic’s request to reinstate its developer program account; Epic has no live apps on the App Store and thus no standing to enforce the injunction.” Youch, plenty of spice left in this one.

  • The folks at popular YouTube channel Ask Gamedev did a video on the top platforms for marketing data, including websites like Gaming Analytics, SteamSpy, SteamDB, Game Data Crunch, TwitchTracker and Steam Likes. These names are perhaps not surprising if you read this newsletter, but useful context for many.

  • There’s an interesting ‘fireside chat’ transcript with Sony’s Jim Ryan over at GI.biz. Particularly: “Right now success with the current console model, a really great PlayStation hit you're talking ten or 20 million people being able to play that game… I would love to see a world where hundreds of millions of people can enjoy those games.” Sony could do more cloud, subscription, mobile and PC expansion.. and maybe will?

  • I respect N3twork’s Ethan Levy (a veteran dev/monetization designer), and his take on why you can be bullish about blockchain gaming even from a game design perspective is pretty intriguing: “For me, the potential of blockchain gaming is not about the money, it's not about the promise of my players getting rich quick, it's about how ownership can change the way a game makes them feel.”

  • Nintendo sometimes lets us behind the curtain in unexpected ways, as in this Q&A with the Switch OLED hardware designers, in which they come close to admitting the Joy-Con has wear issues, and explain: “For example car tires wear out as the car moves, as they are in constant friction with the ground to rotate. So with that same premise, we asked ourselves how we can improve durability, and not only that, but how can both operability and durability coexist?” So the Switch OLED likely has improvements to the Joy-Con form, if not changes to the function?

  • Kudos to Eurogamer for quizzing EA’s Chris Bruzzo on FIFA and loot boxes. Complex moral questions here, and Bruzzo says the right things: “I'll just speak for myself: when I hear about an individual who has spent too much time or too much money in FIFA, it pains me. It drives us, actually. We don't want that. We're not building the game for that.” But we also know it’s possible and engineered for - like most F2P-driven companies, EA want to drive players up in spend. Meh.

  • Like me, MauroNL has noticed that a bunch of the Xbox Game Pass marketing still says “Immerse yourself in a library of over 100 high-quality games” when: “Meanwhile on October 10th [the service has] 419 Console Games ; 374 PC Games ; 307 Cloud Games ; 116 Day One Games.” To be fair, 419 is certainly ‘over 100’. Just goes to show how quickly Xbox is scaling its ambitions for available games.

  • Microlinks: ah, I guess a senior Apple Arcade exec left and is now running PlayStation’s mobile efforts; there’s some more ‘App Store is anti-competitive’ rulings, this time in the Netherlands; some more ‘how to have a hypercasual hit’ eye-openings - “I tested close to 20 prototypes before Run of Life.

Finally, NPD’s Mat Piscatella Tweeted this intriguing graph which shows the company’s GamePulse service tracking a swiftly increasing release count by year on consoles:

Of course, you’re dealing with pandemic disruption, which may explain some of the increase. And in addition, I checked with Mat and “each platform is counted as a separate release.” So for those PlayStation and Xbox graphs, you’ll see intentional double-counting of the same unique game which has both a PlayStation 4 and PlayStation 5 version. (Something that may be getting more common?)

Nonetheless, the 34% Switch increase has to be unique games, and I’m sure the unique number of new games on PlayStation/Xbox is well into the double digits. Pretty interesting stuff - thanks for doing all the math(s), NPD!

[We’re GameDiscoverCo, a new agency based around one simple issue: how do players find, buy and enjoy your premium PC or console game? We run the newsletter you’re reading, and provide consulting services for publishers, funds, and other smart game industry folks.]