I built a public paper-trading lab for alternative-data trading signals

I’ve wanted to build something at the intersection of data and finance for a long time. About 25 years ago, I was already interested in quant-style thinking, but life pulled me in other directions and I never really made the thing I had in mind. Over time that just sat there in the background as one of those “someday” ideas.

StockArithm started to become real after a recent loss in my life. I spent that day thinking instead of just drifting in grief, and it clarified something for me: I didn’t want to keep postponing the thing I’d been curious about for years. I wanted to actually build it.

So I did.

StockArithm is a public paper-trading lab for alternative-data sector signals. The basic idea is simple: take weird but real data like TSA volume, freight rail, electricity demand, Google Trends, and sentiment/activity proxies, then test whether any of it has a real relationship to market behavior.

The part that matters to me most is not whether every signal works. Most of them don’t. Some look good for a while and then fall apart. Some only work in certain regimes. That’s exactly why I’m keeping the whole board public instead of only showing the winners.

I wanted a place where the failures stay visible too.

See the board (wins and misses both) at stockarithm.com. Free daily signals through June 30, paid tier opens July 1. If you sign up before then, you’re early.