It’s actually a more helpful analysis for this topic to do a cursory, cosmetic look.
I'm a designer as well, not just a developer, and it's layout is really well done: clean, straightforward, clearly presented, and the data is easy to find. They link to studies and always err on the side of caution in their descriptions.
For me, I've never even heard of the website in question. Why would I, as it's not a field of interest for me. However, if I were to search for it, it would be a search engine result, not a search field on some website I have never knew existed.
Sidebar: examine is one of two sites I regular use the google site filter for: “site:examine.com (supplement name)”. A good search engine would put it first for basically any supplement.
Also, the main page has 2 search boxes if we want to get pedantic about it. Why? I'm assuming as a dev type mindset that they will do the same thing as one is always there while the landing page disappears with use.
And so if you end up on https://examine.com/supplements/creatine/ it pretty much blows away all other pages on creatine.
So at the end of the day, you still end up on useful information.
In my experience, this is often the cause of mysterious drops in search ranking. It's very easy to inadvertently introduce changes that negatively impact your ranking without even noticing.
At a quick glance, I noticed that many articles on examine.com link to hundreds of external references (e.g. more than half the page of https://examine.com/supplements/creatine/). In Internet Archive snapshots from before the redesign, these have the rel="nofollow" attribute, but on the current site, they do not. I'm not saying that's the cause, but it might be worth looking into exactly what changed in the redesign.
Hacker News would never be, for me personally, on the top of the list of websites that inspire confidence by its looks, it's only when you delve into it and realise the content is actually great that you can appreciate it.
Such a shallow evaluation is quite strange coming from technical people who are used to mailing lists and all sorts of badly designed (or at least aesthetically unpleasing/neutral) pages...
Other than the whole unchangeable orange theme, I think it's fairly easy to intuit.
“Topcolor” setting on your profile page
The more something has been “designed”, the more the sight falls into the untrustworthy category, barring substantial evidence to the contrary.
But that has no bearing on whether the site is actually presenting comprehensive or accurate information.
If users act as though it's a dodgy site, how is google supposed to know whether it isn't? You could argue that the wrong heuristics are being used, I'm not sure the technology is there to do it any other way though?
If you are actually doing research on some subject and are actually prepared to read and obtain new knowledge then yes, your cursory first look is going to be about the content and sources/references, and if you're still going to bounce on the looks then either you're not really trying or you are still just looking for a simpler bite sized easy answer (which pretty much do not exist in this field of science).
By that logic, Google could penalize every site that has in-depth knowledge about some subject.
And now that I think about it, these are EXACTLY the kinds of websites I've been missing from the Google search results in the past years. Most people first stop for "in-depth" knowledge would be Wikipedia (try defending THAT one, 6 years ago ...) and if you really want to, maybe that PDF of a publication is not behind a paywall. The web used to be full of pages that just were made by people crazy smart about a subject and they wrote about the thing they love ...
Just for illustration, I went through my old bookmarks, the original link was dead but the page still exists: http://gernot-katzers-spice-pages.com/engl/index.html Just browse a few pages and see what a quality site it is. It even has each page in both Germand and English.
This page used to pop up all the time when you searched for spices way back in the first half of the 2000s. Try googling "fenugreek" now and cry ...
And the majority of this quality content is not even ad supported at all. That is the worst part. So many people seriously argue that you need ads to support the internet. Well, THIS is the internet that I want, the good one, the promised internet. And look at that "fenugreek" search result page again, it's being fucking buried by this shitty ad supported internet of hollow articles about "fenugreek health benefits". THAT is the internet you get, you support by supporting ads. For every starving quality journalist at the news websites that argue they have to serve you megabytes of adtech with a two paragraph article, there are a thousand regurgitated content farm bullshit sites, easily consuming the vast majority of this internet advertising pie ... it's like cheering on mass murder because the obituaries make such nice haikus some times.
Measuring engagement is great for shallow content, if you even can call it that. We see that all across the net.
But high-value, in-depth knowledge, is very often relatively boring.
Examine.com is a very trustworthy website with good research, unbiased information and very good citations that are summarized in a scientific way.
Regardless of the heuristics they use to determine misinformation, they definitely messed up here and it should be re-evaluated.
A few years ago some of the mega UK insurance brands got into major trouble with google. I wont mention any names but directly afterwards they started using cute animals - obviously Sergi had being doing some naughty Black Hat SEO