Hacker News Reader

Comment by darcys22 | Hacker News Reader

I used to wear a fitbit when playing squash but after a while heart rate data says the same boring stuff and doesnt help much.

Unsure if the tech is there but id be more interested if they tracked the players movements in detail using some form of motion capture. Squash is confined in a small room so its probably easier to do this than other sports. Have a fully digitised replay available for each game.

Then we could start analysing the data to show to amateurs how they should have moved and reacted within their own games compared to how the top players have reacted in similar situations.

View on HN

This is a neat idea. PSA currently shows a movement heat map over the course of each game but it isn’t interesting either - they are all just very similar looking blobs.

I would love to see slow-motion breakdowns, but that might only be interesting to fans who play seriously. Then again, maybe that’s most fans since it is a more obscure sport?

I’m just glad they finally figured out that dark blue courts and a white ball is the only combination that gives people a hope of seeing the ball on TV. High def viewing helps too, of course, but the old color schemes used before 8-10 years ago made it impossible to see the ball.

Yeah, the blue/white combo has indeed revolutionized the TV viewing experience. We need more characters in the commentary box too though. Joey Barrington is good at what he does, but others are "Meh" so far.

A spot in the summer olympics would go a long way to improve everything, of course.

What is the deal with the commentators emphasizing when the players “fist pump” and pronouncing it strangely? Such a silly gimmick.

Considered carefully, heart rate data measures your tiredness (probably even better: how quickly it returns to baseline).

For pro athletes, comparing heart rate data to the player's typical patterns could tell you how hard they are being pushed and potentially give a very good indicator of who was going to win.

You could sell the data to gamblers, as well.

Not sure if squash is popular in the same markets as sports gambling, but there's at least some opportunity there.

There's a bunch of apps now sort of like this for golf: swing, posture, ball flight, etc.

Given something like this old post about tensorflow[1], you'd think this isn't even the worst hobby project if you can get some footage that a model likes. I've never really thought to try running NBA clips through it...

[1]https://medium.com/tensorflow/real-time-human-pose-estimatio...

PSA has done this in terms of heatmaps with sportsdatalabs. Personally I would invest into a video AI for let/no let system, as well as better camera systems for out of bounds calls ( like in tennis ). Also, switch to a lime ball on a blue or even black glass surface would make for even better watchability imo.

An AI system that makes let calls would revolutionise the sport!!

Accelerometers on the wrists, ankles, hips, and ... back of the head? And that would be enough to pretty much fully reconstruct the player's movements.

Reply on news.ycombinator.com