Revisiting the Moneyball topic

[CochiseCowbell]

...without bumping an old thread. This just came through "my algorithm."

 

[leeinator]

John Henry was gonna use analytics whether Billy Beane accepted his offer as GM or not. He proved that by hiring Bill James prior to offering Beane. So basically when the rich big market teams took on some level of analytics it became scary in the sense most of them win the World Series than not. So small market teams like Oakland still stay pushed down no matter what.

 

[ZombieKissinger]

John Henry was gonna use analytics whether Billy Bean accepted his offer as GM or not. He proved that by hiring Bill James prior to offering Beane. So basically when the rich big market teams took on some level of analytics it became scary in the sense most of them win the World Series than not. So small market teams like Oakland still stay pushed down no matter what.

Astroball talked about some of the evolution that followed. I still think there’s differentiation in analytics, but it’s a lot harder and might translate into computational power. For example, having AI “watch” games and show matchup-specific advantages. The matchup part already happens with play by play and stat lines, but the video component will be a whole extra element. It could lead to different approaches to roster management than what we see now

 

[POTUS]

Any sport that has discreet events, like plays or pitches, will be more susceptible to having AI reveal advantages. Lot harder in games where ball doesn’t stop like basketball and soccer.

 

[OG Goat Holder]

Any sport that has discreet events, like plays or pitches, will be more susceptible to having AI reveal advantages. Lot harder in games where ball doesn’t stop like basketball and soccer.

Good point. I C U football.

 

[OG Goat Holder]

John Henry was gonna use analytics whether Billy Beane accepted his offer as GM or not. He proved that by hiring Bill James prior to offering Beane. So basically when the rich big market teams took on some level of analytics it became scary in the sense most of them win the World Series than not. So small market teams like Oakland still stay pushed down no matter what.

Yeah but they had an advantage for a few years. That’s all you need to build something. But if you don’t get that title, like Beane said, you are forgotten.

I wish we’d take that approach with offensive football. Just constantly try and stay ahead of the game, or go completely against the tide.

 

[UpTheMiddlex3Punt]

It seems like a lot of teams are doing cargo cult analytics when it comes to payroll. The 2002 As weren't built on a shoestring budget because that's what analytics told them to do. They used analytics because with their shoestring budget it was the only way they could win. If the owners of the As gave Billy Beane an extra $30 million in payroll for that season he would have put together a much better looking team. He wouldn't have turned the money down "because statistics".

 

[POTUS]

Good point. I C U football.

Some genius student at a place like Stanford is probably going to get the All-22 for all the ACC teams and run it through A.I. and give the coaches some big advantage by revealing tendencies by formation, motion or personnel groupings. Can they translate it to the dudes on the field and have them make use of it? I guess that’s the big question.

 

[ZombieKissinger]

Some genius student at a place like Stanford is probably going to get the All-22 for all the ACC teams and run it through A.I. and give the coaches some big advantage by revealing tendencies by formation, motion or personnel groupings. Can they translate it to the dudes on the field and have them make use of it? I guess that’s the big question.

I think it’ll get weird. It’ll involve computational power and quick adaptations between games. Some sports may change roster composition to add flexibility. I also wonder if coaches/players will start messing with AI the same way Romo said Belichick ran a play he normally wouldn’t run in trash time to screw with analytics. I don’t love that sort of thing, but you could see people drop velocity on some passes, show less ball control, etc. You might have a big game 4 weeks away and 2-3 trash games before that. If you ran analytics on yourself and identified matchup strengths and weaknesses, you might intentionally do things to make your strengths look worse to AI, so your opponent doesn’t position itself to neutralize them.

 

[mcdawg22]

Good point. I C U football.

How is AI going to help? You can’t throw a tight spiral with four fingers**