Linear algebra explains why Jev trading bot questions must span different dimensions
Using 3Blue1Brown's idea of linear dependence, it argues that firing several parallel questions at a bot counts the same dimension twice if they all point the same way, so each question must aim in a different direction.
A basis is any set of vectors that reaches the whole space with none of them redundant, i-hat and j-hat are just the default pair, and if one vector already sits inside the span of the others you can delete it and lose nothing, 3B1B calls that linear dependence Jev trading bots run into this every block, you fire six questions in parallel, but if toxic flow and direction always move together you're counting the same dimension twice and the bot learns nothing new, each question has to point somewhere the others don't 5 minutes of math worth watching before your next bot build, save it and read the article below
This post quotes another post: “Building an HFT trading bot with Jev: one buy/sell decision per block”