Predictive power of additional Data provided by Santiment

After doing analysis on the ERC20 tokens now for more than two months - I tend to conclude that additional data fields provided by Santiment (beyond OHLCV) have some predictive power for future trends for day sampled data. Below are my two publication in this regard on steemit

https://steemit.com/crypto/@prameshtyagi/data-provided-by-santiment-improves-trading-profitability

https://steemit.com/santiment/@prameshtyagi/combination-of-santiment-data-and-proprietary-trading-signal-suggests-bear-market-to-continue

I am still working to understand where the value is hidden,if at all, in fine sampled data for similar kind of success. The field used in above study are - Open, close, high, low, volume, marketCap, outInDifference, activeAddresses, burnRate, activity, mentionsCount. I expect that burnrate and activity wont play any role in fine sampled data analysis. Analysis continues. Lets us see what comes out.