The path is ideation, hypothesis, specification, backtest, iteration, walk-forward, paper, live. Most of the elapsed time is the loop between specification and backtest, and most ideas leave through the bottom of that loop rather than out the far end.
Walk-forward is where the honest ones separate from the flattering ones. Decisions made inside a training window are frozen before test sessions are scored, so a parameter chosen because it worked cannot then be credited for working. Anything that only survives when that freeze is relaxed did not survive.
Paper is the gate people underrate. It is not a rehearsal in a separate program — it leaves through the same order-management door as live, with the same routing, the same timeouts and the same fan-out, and only the terminal fill is simulated. That means a strategy which clears paper has already exercised the machinery it will use on capital. It also means paper catches an entire class of problem a backtest structurally cannot: the ones that only exist because the order has to travel.
Eight strategies cleared every gate into live capital — intraday index options and overnight stock options. The number that mattered more was how many did not.