Good post, it's not right, but I like it.
Sure a supporter can spot when Anderson improves his passing in one game like the Bolton one, but a 3% increase across the entire season? I'm not going to chastise someone for not picking up on that. Which is the point I was making.
3% is marginal to me because it could be caused by various factors other than improvement. It could even be a sign of regression, he could playing it safe, he could be dropping deeper, exchanging with the centre backs more. There's plenty of contributing factors that could cause a 3% increase that lie outside a basic improvement in passing technique. Put it this way, if it drops by 3% come the end of the season, I won't be citing a regression based on that stat, I'd be looking at how effective he's been, key passes, assists etc.