Game Maker's Toolkit - the 2016 audit (Patreon)

So I thought it might be interesting to look back on what 2016 looked like for Game Maker's Toolkit. See what did well, what I achieved, and what I can learn.

This post is really, really long and potentially very boring but I decided that if you don't care you'd skip it whatever the length and if you do care you'd appreciate the level of detail. So fill your boots.

(Hopefully the embedded images work.)

What I made

So in 2016, I released 20 episodes of Game Maker's Toolkit - from Fallout New Vegas in January to Event[0] in December.

That's four off of my target of 24 (two a month). I missed one at the beginning of the year due to illness, and then three at the end because I spent extra time on the Nintendo video and on the seven episodes of Boss Keys. Let's call it even?

I'm proud of basically everything I did. I don't do turkeys and I will can episodes or redo them until I'm happy. Quality is everything to me and I don't care if the video is late or niche or contentious - it's all about fitting in lots of detail, getting my point across clearly, and entertaining people with high production values.

I'm particularly proud of the episodes on DOOM, Jonathan Blow's puzzle design, controllers, immersive sims , Shovel Knight , and Nintendo. Of course none are perfect, I wish I could redo the lot, and I've learned something from everything I made. But that's life!

I also did a good job of hitting my mission statement for the channel. 

To reveal the secret genius of our favourite games (DOOM), to share the philosophies of the best designers (Nintendo), to highlight games that have interesting takes on general systems (Health, Morality), to get people thinking critically (Dark Souls difficulty), to demystify game development (Spelunky), and to "review" games in a balanced manner (Mirror's Edge). Zoinks!

Okay, let's see how that actually translates into traffic.

Views

As I said before, I don't worry too much about views. But this stuff does help me know what people want to watch, and that can guide the balance of the channel.

Let's rank the videos in order of views (thousands).

• Nintendo - Putting Play First - 193
• (Boss Keys) Ocarina of Time - 144
• (Boss Keys) Oracle of Ages and Seasons - 138
• What We Can Learn From Doom  - 135
• (Boss Keys) The Wind Waker - 122
• Anatomy of a Side Quest: Beyond the Beef   - 121
• (Boss Keys) A Link to the Past - 114
• Shovel Knight and Nailing Nostalgia - 114
• How Games Do Health - 113
• Controllers Control Everything - 112
• (Boss Keys) Link's Awakening - 111
• 5 Game Design Innovations from 2016 - 110
• (Boss Keys) Majora's Mask - 110
• Morality in the Mechanics  - 109
• How (and Why) Spelunky Makes its Own Levels - 108
• Hitman, and the Art of Repetition - 105
• What Mirror's Edge Catalyst... - 103
• How Jonathan Blow Designs a Puzzle - 102
• (Boss Keys) The Minish Cap - 102
• The Comeback of the Immersive Sim - 100
• Depth, Mastery, and Vanquish - 96
• Why We Remember Bioshock's Fort Frolic - 95
• Should Dark Souls Have an Easy Mode? - 93
• A Deep Dive into Star Fox Zero's Controls - 86
• Downwell's Dual Purpose Design - 82
• In Praise of Prague (and other small worlds) - 82
• How Event[0] Works - 71

That's about 3 million combined. But I actually got over 5 million last year because old videos continue to do well. My best video of 2016 was actually the episode on Resident Evil 4 which had a whopping 459,000 views! It got a big spike in September (53,000 views in one day) as YouTube recommended it on people's homepages - then lots of residual traffic from there.

Also, the Super Mario 3D World  video continues to succeed (144k views), and to a lesser extent, Game Feel, Dotted Line, and Good Game Design from 2015.

Now, it's tough to get any real lessons out of this data. Nintendo videos do really well... except if it's Star Fox! Indie game videos are duds... unless it's Shovel Knight! There are also lots of other factors like time (Prague and Fort Frolic did way better than Beyond the Beef on their respective launch days), external promotion, YouTube algorithms, and names / thumbnails.

Like, maybe it was colossally stupid of me to not put "Deus Ex: Mankind Divided" in the name / thumbnail on the Prague video. I might fix that...

Then again, Deus Ex wasn't a huge release. And I gotta say, my choice of 2016 games was pretty bad (from a pure views perspective). According to Google, people were searching for Overwatch, Battlefield 1, Dark Souls 3, No Man's Sky, and Infinite Warfare. And I chose... Mirror's Edge Catalyst, Star Fox Zero, Deus Ex: Mankind Divided and Event[0]. Man, I suck at this!

Ultimately, those were the games that I wanted to talk about and I'm glad that I did. And Hitman was a larger game. But I think I could do a lot better at saying something on the more noteworthy releases... No Man's Sky was a treasure trove of game design stuff but I resisted because everyone else was already talking about it. I'll be more assertive this year.

Anyway, here are some things that DO work:

Older, popular games

New Vegas, DOOM, and the Zelda franchise did well. People don't always want what's brand new and I will always be digging up older stuff. The exception here was BioShock. Maybe people were burnt out from all the retrospective articles that came out around the release of the remastered collection.

Positive, feel good content

The videos on Nintendo, DOOM, and Shovel Knight did really well because they are full-blooded celebrations of these companies and games. Fans of this stuff can share these around and say "This!" and "I knew I liked Nintendo games for a reason!". It's great to have your love for something validated and I think is part of why "love letter" channels like Movies with Mikey and Kaptain Kristian are so successful. On the flip side, dunking on something crappy like No Man's Sky or Suicide Squad is also fun. But more evenhanded critical stuff is always gonna struggle I think... OH WELL.

Boss Keys

Yeah, this is more popular than I imagined. I don't know if it's just Zelda or if people really like the way these videos work. Following the research process, waiting with anticipation for me to play your favourite game in the series, shouting at me in the comments because Wind Waker's dungeons are actually really good if you understand the LORE or something. Got a few more of these to do, and will then look at other franchises later in the year (Souls, Metroid, Castlevania).

On that note... I want to do videos on the Dead Rising quartet and Prince of Persia's messy history of reboots, and the success of Boss Keys is making me wonder if I should do them as video-per-game type shows, instead of single episodes?

One theme, lots of games

How Games Do Health, Controllers Control Everything, and Morality in the Mechanics are all videos where I look at how one almost universal theme is tackled by lots of different games. Maybe it's a bit like the Top 10 list phenomenon where people are curious about what is included. Either way, I'm gonna do more of these.

Traffic throughout the year

(This pic is also attached to the post if it's too small to read)

Here's a graph to show what 2016's views were like, day by day. It's pretty unremarkable: I got a spike for every video. There's definitely an upward trend, though, where the first day's traffic for each video has largely gone up and even my least popular video, on Event[0], would have been a megaton blast if you moved it to January. 

The only non-video spike was on December 26th, which is when Extra Credits recommended my channel to some 100,000 viewers. Thanks Dan!

Let's drill down into that big Nintendo spike. How did people find that video? 32% of the views came from external sources, and half of those were from that old kingmaker, Reddit. Lots of traffic also came from the YouTube homepage, subscription feeds, and suggested videos. 

It's crazy to me that only 1,800 people saw the video on Twitter. Polygon, NeoGAF, and 4chan got me some views too. 

Anyway, back to the numbers. My average view percentage in 2016 was 61% which is excellent. People could stomach my older videos for longer but they were also much shorter. As I said before, I might do some 5 minute stuff this year.

My videos are overwhelmingly watched by men, which is a bummer. GMTK viewers are mostly American and British, and they're watching them on a computer rather than a phone or tablet. People watch the videos almost exclusively on YouTube itself, rather than embedded on another website.

And while most people watch them with English or no subtitles, a large number of people do watch them with translated subs. This year, GMTK episodes were translated into Spanish, Portuguese, French, German, Indonesian, Chinese, Polish, Dutch, Russian, Vietnamese, Japanese, Korean, Turkish, Swedish, and Italian. Which is ridiculous.

Subscribers grew by over 100,000. I lost 5,000 subs. Bye! It seems like nobody "likes" videos on YouTube - roughly 4 percent of people will hit either like or dislike. My most liked video was Bioshock (about 6% of viewers hit like). The only video that got a sizeable number of dislikes last year was the Resi 4 video with 319 dislikes. But with comments like this, it's not surprising... 

And, hey, not a single video on my channel has ever had fewer than 97% likes so, eat that. Even that video where I said Wind Waker is a bit pants got 99% approval. 

In comments, people enjoyed talking about Resident Evil 4 (764 new comments), Minish Cap (617), Health systems (568), Dark Souls (525), and Nintendo (520). No one really clicks cards, except for one on the Mirror's Edge video where a whopping 70% of people clicked on the card about the video for "Little Dotted Line". I've started playing with YouTube's new end screen system and will let you know how that goes.

Other stats

Patreon grew a lot last year. I started with 239 backers and ended with 1,219. Patreon is basically this crazy race where you lose a bunch of people at the start of each month (totally understood - I appreciate everyone's support for as long as they can give it and have no expectations that your backing is forever!) and so you have to get new people to fill that gap - and hopefully some more on top.

This can be an issue for those with less regular releases but it's has never been a problem for me. I released new stuff every single month in 2016 (which always translates to new backers) and that momentum has kept the chart growing and growing! And if you're curious - yes, there was a new year drop off but I've already regained it.

Average support is down. People were giving $3 in Jan and $2.60 in December. It's not a big enough deal for me to worry and to be honest I'm much more comfortable having lots of people giving a little rather than a few people giving a lot (though: a special thank you to those who are donating beyond the set tiers). And I don't want to segregate this bonus content anymore than I already do.

That's basically about it. I accidentally spent all morning on this. If you have any specific questions let me know and I'll be happy to dig further into stats. I love this nerdy stuff. And, of course, I'm interested in any other thoughts, reccomendations, and takeaways from this data.