I know that this is completely a personal preference, but for me, there's no point to watching most sports events delayed. Sure I'd watch a clip if something re

I know that this is completely a personal preference, but for me, there's no point to watching most sports events delayed. Sure I'd watch a clip if something remarkable happened, but the point for me is to be there (remotely) as it happens. I wouldn't have re-watched the whole women's CAN-USA gold medal game the next day if I hadn't stayed up till 1:30 watching it, I would have at most just watched the highlights. The one personal exception to that is Formula 1, where I'll go to great lengths to avoid spoilers for a race that I couldn't watch live. I don't know much about the background of the author, but I feel that they understate the challenges in doing AI-based scoring for figure skating. They mention being close to the "ideal form", and say that it might be easier than expected to judge the artistic elements, but don't really support this. Figure skating is an art, and judging it in any way where the artistry isn't at a similar importance to the technical aspects defeats the purpose. I do acknowledge the point on needing to be skilled to tell apart tricks. The figure skating pundits would immediately recognize each figure skating trick, while I was left, at best, counting the number of rotations. I just don't know enough to see which edge of the skate the use, or how much toe there is. Overall, I can't agree with a lot of the author's points. Without tripling each event (to gain a larger sample size), and radically changing the sports in play, there isn't much you can change. Sure you could drop all bracket games and do a full double-round-robins, but I feel that that would hurt the games far more than it would help.