But you're not espousing a neutral view. You're claiming that the "face value" of "175m users streamed a show" isn't subject to distorting effects like very low views per person etc. to any significant degree. That's already a particular interpretation (and the implicit exclusion of other intepretations) that you somehow purport to be "neutral". That's the bias.
And by the way: "face value" means nothing. It doesn't somehow give more legitimacy to something, or make data more accurate, or exclude problematic interpretations. It's a meaningless red herring from the get-go.
"Impressive" is a value judgement, which in turn means it's intrinsically contingent on the underlying data. It's impressive IF AND ONLY IF the data is interpreted in a certain way; as proven by my earlier example that if you interpret it differently, it ceases to be impressive. Your claim of it simply being impressive as a default obscures that contingency. That's why it's a bias - it holds true only for certain interpretations, which are NOT derived via objectively justifiable conclusions from the given data.
Your mistake is simply the constant return to it being "impressive" as though that was self-evident fact, when the whole point I'm making is that it can't be evaluated as such. Not with the given data.
That depends entirely on context. "Growth" is another very vague metric. It's used in shorthand, or summaries, or simplified statements. It's MEANT to sound impressive given the kind of document it appeared in, but the absolute number means very little without crucial details.
What's especially ludicrous in your example is that those 70% are in reference TO THE DATA ALREADY IN QUESTION, which means we have no idea how to interpret this.
What if of those 70% extra "stream hits" 60% tried watching for 5 minutes, then never came back? They'd show up the same way in this data, and it'd be treated as "growth", but what that would actually MEAN is that the company massively FAILED in conversion rate because 86% of the people who gave it a shot turned it off immediately. Of course I don't think that's the case, but it's at least a POSSIBLE interpretation of the same data that you cannot exclude, and that has relevance even at much lower degrees of realization. So, again - I don't KNOW what this means, because I can't tell from the data provided how good of a thing this is or isn't.
That's sort of the point? I'm saying the statement is too simple to allow for a value judgement. Your response seems to be "lol look at this guy saying this simple data is simple!" which only reinforces my impression that you are simply completely out of your depth here. You just don't know how data evaluation works. Many people don't. Statistical (and stochastic) illiteracy (or innumeracy) is one of the biggest shortcomings of most contemporary education systems. It's WHY companies push simple numbers so much - it's easy to create any number of impressions with them, without ever actually telling lies.