Can a bot wallets behavior analysis be done to reverse engineer their strategy on meme coin markets?

External question — Cross Validated Stack Exchange Author: Tookie Original post: https://stats.stackexchange.com/questions/677206 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. Say there are thousands of trade pair of a bot wallet with all the contextual trade and not a single missing data or order. Every decision by the said expert is done using the same information available to every one. Meme coin markets are short lived and algorithmically depending on outside information is highly unlikely. How much can we understand from those trades data? The exits are highly state dependent. No clustering of any TP/SL. Neither any time constraint except a hard cut off timer around 20 minutes if the algo doesnt trigger the sell. What kind of techniques i can implement? Can we figure out what they are optimizing for? There seems to be too many variables, despite the same observable raw features, what lookback or transformation of that feature, what prediction are they optimizing for almost felt like black box to me. With so many data of their distinct behavior can approximating/knowing what they are optimizing for, what drives the entry what drives the exit is so difficult? Any guidance will be highly appreciated.