Which colours dominate movie posters and why? | Hacker News Reader

Which colours dominate movie posters and why?

stephenfollows.com

I feel like me and the author of this piece have different opinions about where to put the line that separates orange from yellow.

If you showed me many of the posters shown here in isolation, and asked me to describe the palette, I would have said "that's a yellow poster": https://stephenfollows.com/i/171004131/orange-the-mvp-of-the... - with some exceptions such as Lorax and Unbelievers that I would have said are orange.

EDIT: I changed from "most of the posters" to "many of the posters" as there are some that feel to me decidedly orange.

What we really need is an article "Which colours dominate bike sheds and why?".

I suppose from reading the article they are doing this programmatically, so it should be the code / algorithm / library used that is describing yellow as orange.

One interesting thing is - are these images encoded in the RGB color space or in a color space that allows all the colors human can see. If the latter and the code to analyze assumed RGB there may be things that look more yellow to us that would end up being interpreted as Orange. But that is a long shot.

You can go all the way through the data provenance:

- do you have all the posters for all movies? Probably not.

- do you have well-preserved examples of those posters? Many are going to be sun- or age- faded.

- do you have good scans of the posters that pre-date digital originals?

and so on.

To me it's the distinction between orange and brown (since they're the same colour).

Raises the question if the author could have or should have included grey in the analyses.

s/gray/saturation/

I think the author's line between pink and purple is also not very clear. For example, teen spirit appears in both.

That particular poster has pink on the left and purple on the right.

Obligatory xkcd (from 2010!) - https://blog.xkcd.com/2010/05/03/color-survey-results/

“ Basically, women were slightly more liberal with the modifiers, but otherwise they generally agreed (and some of the differences may be sampling noise). The results were similar across the survey—men and women tended on average to call colors the same names.”

On a tangent, “more liberal” in this context confounds an insight I had on reading it, that while more words are used to describe the range of colors in the one group, more colors are described by each word in the other. Which is the more conservative?

Interesting. I was wondering if the orange/blue thing was going to extend to posters, and it looks like it does. For reference: https://priceonomics.com/why-every-movie-looks-sort-of-orang...

I thought there was a clear smoking gun reason it has become popular but it doesn't seem entirely clear cut. Except that the blue/teal makes orange (and skin) "pop" as the colors are complementary, and color swapping/enhancing has become much easier in recent years. (And I think, somewhat cynically, because both hues are both pretty far from the default green screen color...)

A natural reason teal and orange works - it's generally understood that the most flatteing lighting is at golden hour, when the sun sets but is still very bright. The sun takes on a more reddish hue (going to orange) and the sky tends towards teal. The stricking image of a sunset you can picture in your head.

In real life the cokor of both the sun and sky will strongly affect the color of what you see, meaning you actually see more "teal and orange" at that time, although much less intense than when color graded ( and you're losing light, so you're not used to bright teal and orange).

Unsubstantiated, but I also think your brain feels the manufactured lighting (which is flatteing to the skin) is similar to golden hour, and that then the teal and orange can make it "make sense" more, unconsciously.

Look out for sparks as well. There was a period when everything had sparks. It seems to have calmed down a bit but you still see it on action posters. Orange and teal is fascinating. Once you know it's a thing, you can't unsee it. Especially when the set dressers are in on it and characters have orange and teal books, light fittings and clutter.

https://www.theshiznit.co.uk/feature/how-to-design-a-hollywo...

Orange and blue are complimentary in subtractive color which explains why this combination is everywhere. Just like red/green and yellow/purple.

https://en.wikipedia.org/wiki/Complementary_colors

Fun post! I like reading blog posts like this.

I did wonder about one thing:

> The first measure I tracked was a ‘Colourfulness score’. This measures how intense and varied the colours are in a poster. Higher values mean the image contains more saturated and diverse colours, while lower values indicate a more muted or monochrome palette.

As the description suggests, it seems to me these are actually two things: one is how saturated the colors are and one is how diverse they are. In other words one is how much color there is and one is how many colors there are. The examples given for each color later on are ones where a single color predominates, but it's not clear from the article how common this pattern is.

Personally I'm intrigued by posters that use a wider palette. Subjectively it feels to me that movie posters have moved towards a minimalist design over the past 10-20 years (as have book covers), making single colors more central, but I'm not sure whether that hunch would be borne out by data.

Although it's an interesting attempt, the first genre I thought of doesn't seem to fit the analysis. The predominant color scheme of classic horror was green and purple. TFA notes how green can connote the weird, but doesn't recognize the prominent role of purple in classic horror.

This might be because of two color processes (e.g. two strip technicolor or cinemascope) which were cheaper and excluded either yellows or blue hues. Purple (or magenta for bright spots) also is complementary to green so they tend to be used together.

Personally I always associated sickly yellow and greens with horror, but that seems to have been a trend perhaps.

Where do you draw the "classic" line?

I feel red/green/black was the most common scheme in the horror I saw as a kid ('70s-'80s), tho perhaps it was purple and not black..

Classic Horror generally means into the early 1950s. The genre changed a lot after that. So yeah, 70s/80s horror aesthetics were quite different.

Isn't '30s to '50s usually get called the Golden Age, with Classic being until the late '80s?

Huh, that's a new one to me. If some people are using it that way, I need to be careful about using "classic". In my reading, both Classic and Golden Age have been used to mean the era dominated by Universal Studios.

Well done! I was expecting less, to be honest. I liked how the article noted that colors can carry antithetical qualities, depending on the context. Since the context is cultural, it would be interesting to extend the study to other countries.

Looking at the "Brightness Contrast Score", clearly dark posters are on the left - even those with high contrast like 2nd, 4th, 5th poster, whereas the "high contrast" is often given to a very bright poster that happens to have a black background footer.

Interesting idea! Would be nice to see:

* How the colors were picked and assigned to each category and (e.g. at what point is red pink and no longer red)

* An indication of distribution in charts, they have different scales on the y-axis.

* The author likely sampled posters with mostly the same color above a given threshold for each category, would that (together with the lack of methodology and error bars) heavily skews the reader's presentation of the data analysis?

There is no actual data-driven reasoning behind colors here, its just unsubstantiated supposition with charts nearby.

It's not science, it's a perception-based theory, a criticism that is a form of art. No need for data-driven reasoning at all.

I couldn't disagree.

Seems like a pretty blurry line between orange and brown, which makes sense as they are the same color.

Is there some kind of correlation between color and imdb rating?

I'd bet you can find some correlation due to very bad and niche movies having very low budget and very predictable, bland posters. Moreover, bad movies might turn to colored posters later than good movies with high budget, so historically there may be a correlation here as well. Overall I wouldn't expect more than some slight statistical difference.

I wonder to what extent the color palettes of movie posters match the color palettes of the movies themselves (for those that have one), or is this mostly unrelated ?

In the red section they speak about red on white in comedy, but then have a separate section for posters with white backgrounds, some with red text overlay.

I also wonder what color filters are applied to movies themselves and what colors dominate there.

I'm guessing a lot of green and blue.

Any guess which dataset OP used for this?

I don't watch modern action movies precisely because they lack colors.

We went from black-and-white to color to blue-and-orange cinema.

Reply on news.ycombinator.com