three3d wrote:Audi wrote:So OP posts an “eye opening analytical breakdown” in which an armchair stat head sets out to answer the following (this is verbatim from the guy’s mouth, btw):
“Is Paolo Banchero really one of the most talented rising stars in this league OR is he a chucker who is super inefficient and highly overrated?”
…gee, I wonder why there’s discourse and pushback on a fan forum.
I didn’t make the video nor did I say or ask if Paolo was a rising star or an inefficient chucker, so yea pushback isn’t warranted towards me. I appreciated the statistical comparisons in the video and the clarity into what the numbers really reflected. For me I found the percentiles more helpful than the percentages.
Hearing someone has a 40% fg percentage doesn’t give me the whole picture, now if you tell me his fg percentage of 40% puts him in the 38th percentile I can see how bad that truly is when compared across the entire league.
I know you didn’t make the video - but it’s a great example of most people’s take on analytics - that numbers can be bent to tell you if a player is a star or a bum.
Even thinking you are getting a proper indication of “how bad 40% shooting is” via league percentiles is ill informed. Think about it - what can you really glean about a player shooting 40%, when that number is held against a background of players in varying situations? Bench players? Deep bench? Starters playing alongside solid distributors or alongside other superstars getting them easy looks? “He’s only in the 38th percentile!” Ok, so what does that actually mean when, say, 95% of the players counted in the statistic don’t see anything close to the same defensive attention? When they don’t face doubles and triple teams?
The only eye opening thing about analytics is that you are better off just opening your eyes and watching. People can make numbers say whatever they want them to say. That’s how you got roped into a video titled “the Paolo problem”, posted by the same guy who also has a video titled “the Giannis problem”.