Computation of volatility (I use average true range) has predictive qualities. Expansion and contraction is more predictible than price. Only the direction remains to be seen.
Relative strength, a synthetic result pitting a prevailing price against an index, say the S&P 500 has predictive qualties. RS will fade before the raw price. Neutralizes index influence revealing genuine movment.
Rate of change aka momentum also changes before price, depending on the timeframe one uses. I use 23 days, which approximates a little over a month. Pretty damn good filter for me.
Seasonal tendencies have predictive quatlities. Expiration (assignments) bias and beginning of the month (money inflows) are examples. Again, overall direction remains to be seen. More a matter of fertile ground.
Round numbers, once clearly pierced (subject to interpretation) I would suggest are predictive.
A basic triangle (declining resistance and rising support). Has to resolve. No way around it. Now, that resolution can be a headfake, but it has to resolve. That's predictive. The sooner the break is before the apex, in my opinion, the more reliable.
Fundamentals? LOL. I've discussed this in the past. Too much typing. In a nutshell, 1. what's counted. 2. what's not counted. 3. how what's counted is counted. Cash vs accrual. Smoke, mirrors and bullshit. And some little twat auditor, named Susie, 3 years out of college, instructed NOT to upset the client. It's also the past. The word "predictive" pertains to the future.
Random? Eh.......there are consistent money makers. Their focus is position size and exits. Latter word is plural.
What'ya want, a swami? Some hag with a crystal ball? Have your palm read? You've got the boy part right.
, all I will say is go away and converse with those who think and act like you, for, you have a lot to learn from each other