Captioned Video Accessibility: “Stranger Things” Captions, a Fascinating Case Study


It took me a long time to get around to watching "Stranger Things." That's because I'm not a fan of the horror genre. The show turned out to be a fascinating case study of captions. They contain high-quality captions and went a little overboard on captions for sounds. We can learn a lot from the show's captions.

So, I’m late to the party. Netflix’s “Strangers Things” party, that is. I didn’t tune in until the release of the entire Season 4. Resistance was futile especially when the “Stranger Things” captions became the subject of memes and articles.

I’m not a fan of the horror genre. The hairs on the back of my neck still stand up whenever I think about a movie I should’ve never seen. Why in the world I watched “The Others,” I’ll never understand.

I digress. I watched the entire Netflix series in eight days. The captioning in the first three seasons followed captioning best practices like those in the Caption10 guidelines in the complete guide to captioned videos.

Aside from one video, I never thought I’d see the day when something is over-captioned.

That day has come.

The reason is rare for something to be over-captioned is because captioning takes time. The longer the video, the longer it takes to caption it. Hence, no one would attempt to caption too much.

Until now.

Please know this is about progress over perfection. This isn’t a criticism. It’s education. Learning from the show’s captions and how to improve.

The subtitlers are passionate about captioning. They obsess over word choice. Love it! With the best of intentions, some passionate subtitlers got a little carried away. The show has beautiful captions. It makes for a fantastic case study of both high-quality accessible captions as well as what not to do.

Captioning of sounds? That’s a different story. Before moving on, this does not contain spoilers. #NoSpoilers

Adventures with “Stranger Things” Captions of Sounds

The show’s subtitlers are passionate. They want to reflect the story in their captioning of sound. They decided to get creative. I appreciate that and all their efforts to enhance the viewing experience.

Many articles and memes have popped up in relation to the captions on “Stranger Things.” It’s cool to see captions getting attention. It also prompts this post out of concern that others will want to get creative and overdo it on the captions.

Captions are not the place to exercise creativity. For example, some companies use their brand colors in captions. That’s not a good practice. Captions are also not the place for Easter eggs and big vocabulary words (unless it’s in the dialog). If I notice the captions, then it’s a sign that they need tweaking.

Imagine watching a baseball game and the captions show [thwack] every time someone hits the ball or [blip] when the ball lands in a glove. That would weigh down the viewing of the game.

Every sound does not need captioning. The trick is to describe them in context to the content. The key is to answer: What sound is important to the story that may not be obvious from visuals?

A friend who uses captions but doesn’t depend on them commented they reflected the show’s style of doing things big. Again, captions are not meant to reflect any style.

Captions are meant to capture all audio spoken and sung as well as important sounds and music. However, I found myself constantly looking at the captions and missing the action. A few episodes left me drained. Not because of the wild ride of the show, but from my eyes darting back and forth between the action and the many captions appearing in between dialog.

Let the visuals do the talking

The show’s detail for visuals communicated a lot. And many of the sounds appearing in captions weren’t needed. This would greatly cut down the number of captions and ease the reading load. For example, in one scene that has become a meme.

Eleven is running and hiding. Clearly, she’s panting. The captions showed:

[Eleven pants]

The visuals did a good job of communicating these sounds.

[sobbing] Oh, no.

[screaming] No!

[flames whoosh]

[food plops wetly]

On that last one, if this had happened off-screen, [food plops] would’ve been perfect.

Captioning Key, the best plain language guide on high-quality captions, says do not caption background music that lasts under 5 seconds. This would reduce the number of captions enough to ease the load.

When captioning a video with lots of sounds, don’t rely on the audio alone. The visuals can do the heavy lifting. Also, be descriptive yet concise when describing sounds. “Squelching wetly” can be shortened to “squelching.” That’s because many dictionaries define squelch as a sound involving liquid or mud.

A few episodes had so many captions that I felt like looked at them the entire time. Not all sounds were important or relevant. And sometimes the visuals did the job and didn’t require captions.

Avoid insider knowledge

Also, another challenge is the vocabulary used in the sound captions. Now, my vocabulary game is strong. And yet, there were a few words I did not know.

The show has Dungeons and Dragons (D&D) references. Just because it has these references does not mean the majority of viewers have knowledge about D&D.

In fact, my spouse played a lot of D&D in the 1980s. And our younger son now plays it. Despite conversations with them about D&D, I had no idea what “eldritch” was. Now that I’ve looked it up, it’s a great word but not for sound captions.

I had no clue what [eldritch thrumming] meant or what kind of sound that would be. Thrumming is a fantastic word! It’s “eldritch” that left me scratching my head.

Use plain language

All audio that’s spoken or sung should be captured word for word. But caption information such as sounds and music descriptions should be in plain language that doesn’t exceed 8th-grade reading level.

Some of the words used were at a reading level for high school and beyond. Flesch-Kincaid Grade Level reading score for [Tentacles undulating moistly] is grade 21. This means it would require 21 years of schooling to understand. In an insightful interview with the subtitlers, one wanted to use “susurration.” Thank goodness the other subtitler recommended another option.

If viewers have to look up a word from the sound captions that aren’t in the dialog), then it’s a sign of a problem. Captions are meant to be plain, boring, and easy to read.

Captions are like the stagehands in a play. You know they’re there, but you never notice them unless you look for them. Stagehands often wear black clothes. The job of captions is to stay out of the way and be there when needed. The black background with white text makes that possible.

Captioning Key and Caption10 guidelines recommend captions be one or two lines. And each line can be up to 32 characters long. This lets viewers quickly scan them while limiting cognitive overload. In other words, you can glance at them for a split second and absorb it without making your brain work harder.

My eyes spent too much time on the captions and not enough on the action. The video should be the star of the show, not the captions. Aside from these, the show’s captions are superb and do the job well.

Awesome Things About “Stranger Things” Captions

I need to point out the many wonderful things they did right. Everything followed Captioning Key, Caption10 guidelines, and general captioning best practices. Subtitlers can still get creative with the sounds that need captioning. The keys are to ensure the words don’t go past 8th-grade reading level and don’t require insider knowledge.

Uses mixed case captions

The captions in “Stranger Things” never use uppercase to identify speakers. It uses mixed case and puts the names in brackets.

[Will] Oh, I got it.

Some captioning companies do it like this:

WILL: Oh, I got it.

While the name in uppercase helps it stand out, uppercase captions are harder to read. That’s because uppercase letters are rectangle blocks.

Let’s look at an example with my name.



All caps show zero shape variation. MERYL has a flat top edge and a flat bottom edge. That’s it. No change in shape. It’s like a rectangle block.

Meryl shows three variations on the top:

  • Once for the M
  • Once for e, r, y,
  • Once for the l (little L).

Meryl also has three variations on the bottom:

  • Once before the y
  • Once for the y itself
  • Once after the y

Visual differences in mixed case captions boost readability. The lack of visual differences in uppercase adds friction to the reading experience. Mixed case helps you scan faster. I wish more captioning companies would follow the show’s example in using mixed case captions for speaker identification.

The show correctly used ALL CAPS once or twice. The uppercase words communicated someone getting louder. The captions started with exclamation marks. The person got louder and louder, hence the uppercase.

Screaming! Screaming! SCREAMING!

Thankfully, the captions also rarely contained italics. The only time I recall seeing italics was on a phone call between someone on screen and a voice on the other end. The italics represented the unseen voice. Captioning Key has a few instances where it recommends italics and the show follows these recommendations.

Notes voice changes and short dialogs

Here’s a conversation between two emotionally distraught people where the sounds matter. It reveals what they’re feeling as facial expressions don’t always tell a complete story.

– Nothing. [sniffles]

[hoarsely] No. Oh, no.

Sometimes dialog is so short that you caption both speakers at the same time. This works well in helping the captions stay on long enough.

– Wait, did you see something?
– Yeah

Follows guidelines on line lengths and breaks

Two of the hardest things to get right are line length and breaking points. The subtitlers did a great job in following best practices on breaking points.

It’s hard to create guidelines for breaking points because there are many variables. I go based on instinct from watching captions since 1983. For guidelines on line breaking points also called line division, I recommend Captioning Key’s Line Division guidelines.

It’s amazing how much of a difference length and good breaking points make to the viewing experience. When they’re not done right, it’s harder to scan and absorb the captions.

Identifies songs and music

Every time a song played, the captions showed the song title and the singer. And they also captioned the lyrics. However, not all of the lyrics sung on the show contained captions. I noticed someone singing and no lyrics a few times. It surprised me considering the thoroughness of the captions.

When it comes to music, the important thing is to capture the mood. Here are two examples.

[Jaunty music]

[Foreboding music]

Amazing what a difference an adjective makes, right?

The subtitlers earned an A+ on the rest of the caption guidelines such as accuracy, synchronization, speaker identification, and voice changes. The captions appeared on the bottom and never hid any of the on-screen text or credits. Viewers could easily see both.

As I mentioned before, I don’t like horror. A few friends who feel the same said they loved the show. We all enjoy the many ‘80s references. I’m glad I finally got on the “Stranger Things” bandwagon. Although, I could’ve done without a few of the cringy scenes! The gripping story and characters make it all worth it. If you’re looking for something captivating to watch and an intriguing example of captions, watch “Stranger Things.”

Thank you “Stranger Things” subtitlers-extraordinaire!

Level up Your DEI

Many companies have diversity, equity, and inclusion program. However, they need help leveling up on people with disabilities and accessibility. Take the next step with my help through training and speaking. Want to get to know me first? Of course. Join the mailing list, check out my writings on LinkedIn, order progress over perfection swag, or read the bio.

Image credit: QuentinMqt

3 thoughts on “Captioned Video Accessibility: “Stranger Things” Captions, a Fascinating Case Study”

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.