Using Machine Learning to predict court decisions ... profit!

[Jurgis]

I think this could be a great side-way for investing returns:

 

http://www.sciencemag.org/news/2017/05/artificial-intelligence-prevails-predicting-supreme-court-decisions

 

Assuming the technology can be applied to more accurately predict patent litigation (techs/biotechs), or other financial litigation (Fannie, BK/post-BK cases), there should be a yuge opportunity for outsized returns, since I think most people invest in these areas by guesswork that has yuge biases. Instead you could have a more accurate probability of win and use more precise Kelly for investing.

 

I won't work on this, since it's way outside my area of expertise, but someone who does could make $$$$$.

 

You've heard it here first.  8)

[Jurgis]

BTW, it should be possible to apply similar techniques to merger outcome prediction and get better results in merger arbitrage. However, IMO the legal prediction has fatter returns if you can do it.