Go Kart Mozart

Remember folks, verifications show that AI has its greatest value beyond 4 days. I suspect the physics-based models will win and win big here.

Keep an eye on the Navgem, that holds the key.

A little bit less WOR, a tad better Emass. Predecessor looking decent.

It's up to Reggie, if that capitulates, we're in big trouble.

Deepest darkest New England?

NAM tapped back to the pitcher, while Reggie cracked a double. Batting third, the GFS.

It happened on his first birthday?

Does anyone remember Mike Doran and electrics....maybe 15+ years ago? Now those were some whacky posts.

Didn't you already get scolded by 40/70 for posting this?

Good points. I do think AI currently has solid value beyond just "tame patterns". In any case, just wait until we have a blend of AI, physical inputs, and quantum computing. In a way, I hope modelling doesn't get too good, it will take the fun out this. Like watching the UConn women play Xavier or some team like that.

There are many online articles and verification charts that show just the opposite, with AI outperforming beyond 4 days...at least at h500. It does have lots of flaws, such as black swan events with minimal historical corollaries. It will probably not be good for something like December 92, for instance.

I have never met Ray (40/70), nor have I ever heard him speak. But in my mind, he speaks with the Seth MacFarlane voice of Ted. If Ted were a friend of yours, he would insult you but you would just laugh him off. Or you might rip the stuffing out him...who knows.

AIs have proven to be better outside of 96 hours, but the physics-based models are better inside of that time frame. It makes perfect sense really, the "math equation" is solvable in the shorter time frames, but too chaotic to be useful in the longer range. Pretty much can be said for ensembles vs. operational. The value of pattern recognition is greatest in the longer (mid, really) ranges.

Rock bottom. When I see a post about the 84 Reggie...and I dive in to take a look. I need an intervention.

Unfortunately, we have reached the time frame where the physics-based models outperform the AIs.