So the group involved veterans from like Knight and DRW and stuff: we understood the model of combining lots of small signals with a low-latency regression. We

So the group involved veterans from like Knight and DRW and stuff: we understood the model of combining lots of small signals with a low-latency regression. We were trying to learn those signals as opposed to sweat-shop them. But the broader point holds: signal isn’t alpha.