The topic of HTML validation has been covered here time and time again - we know Google doesn't downgrade or boost a web site with bad or good HTML validation. We also know that there are cases were bad HTML validation can break structured data - but overall, HTML validation has no impact on rankings in Google.
John Mueller confirmed it again last week on Twitter but added an awesome line that I'd like to share. John said:
HTML validation can be weird, some pages have lots of issues which have little effect at all. Sometimes a single issue will break it across many browsers :). Ranking is nothing without conversion, so making pages work well everywhere is a good goal.
Yep, logical, but "ranking is nothing without conversion," is simple. You can send a billion users to a page, if no one converts, it is worthless.
Here are the tweets:
That post was from the webmaster team (folks making webpages here), not specific to websearch. We don't use valid HTML as a ranking signal. It probably helps with some structured data formats and to make content accessible on more devices though.
— John ☆.o(≧▽≦)o.☆ (@JohnMu) May 17, 2018
HTML validation can be weird, some pages have lots of issues which have little effect at all. Sometimes a single issue will break it across many browsers :). Ranking is nothing without conversion, so making pages work well everywhere is a good goal.
— John ☆.o(≧▽≦)o.☆ (@JohnMu) May 17, 2018
Forum discussion at Twitter.
This post was pre-written and scheduled to be posted today, I am currently offline.