Does anyone miss hockey analysis before analytics?

[statswatcher]

number are good. they mean what they say. unless you've got a bad score keeper, they're consistently accurate and verifiably so. you can pay 50 guys to watch hockey all day but none of them would be able to fully replace the pure informational efficiency of the box score. it doesn't give you the full picture, but in most cases it gives you much more of the picture than any individual could. it's in exceptional cases that statistical analysis breaks down precisely because we use statistical analysis to deal with and expect norms and regularities.