Has anyone found a real directional edge using open interest and options volume?

I’m trying to figure out whether open interest and options volume can be used in a genuinely repeatable way to forecast short-term stock direction.

My working assumption is that there should be some signal in the options market beyond simple sentiment narratives. For example:

• put/call OI imbalance

• day-over-day change in OI

• abnormal options volume

• volume-to-OI ratios

• strike concentration around ATM vs OTM

• positioning by expiry, especially near-dated contracts

What I’m looking for is not a discretionary “read the chain” approach, but something more systematic:

• a formula

• a factor

• a scoring model

• or a tested setup that produces a directional bias with at least some robustness

Has anyone here tested this seriously and found a framework that actually adds predictive value?

Thx