WebP is awesome at lossless and way better than even PNG.
It's because WebP has a special encoding pipeline for lossless pictures (just like PNG) while AVIF is basically just asking a lossy encoder originally designed for video content to stop losing detail. Since it's not designed for that it's terrible for the job, taking lots of time and resources to produce a worse result.
Lossless webp is actually quite good, especially on text heavy images, e.g. screenshots of a terminal with `cwebp -z9` are usually smaller than `jxl -d 0 -e 9` in my experience.
Lossless AVIF is just really quite bad. Notice that how for photographic content, it is barely better than PNG, and for non-photographic content, it is far worse than PNG.
It has lossless just to check a box in terms of supported features. A bit like how JPEG XL supports animation just to have feature parity. But in most cases, you'll be better off using a video codec for animation, and an image format for images.
There are some user-level differences between an animated image and a video, which haven't really been satisfactorily resolved since the abandonment of GIF-the-format. An animated image should pause when clicked, and start again on another click, with setting separate from video autoplay to control the default. It should not have visible controls of any sort, that's the whole interface. It should save and display on the computer/filesystem as an image, and degrade to the display frame when sent along a channel which supports images but not animated ones. It doesn't need sound, or CC, or subtitles. I should be able to add it to the photo roll on my phone if I want.
There are a lot of little considerations like this, and it would be well if the industry consolidated around an animated-image standard, one which was an image, and not a video embedded in a way which looks like an image.
I believe it is more fundamental. I like to think that AV1 entropy coding just becomes ineffective for large values. Large values are dominantly present in high quality photography and in lossless coding. Large values are repeatedly prefix coded and this makes effective adaptation of the statistics difficult for large integers. This is a fundamental difference and not a minor difference in focus.
Usually the issue is not using the YCgCo-R colorspace. I do not see enough details in the article to know if that is the case here. There are politics around getting the codepoint included: https://github.com/AOMediaCodec/av1-avif/issues/129
