Most projects skip modelling because they think it is about predicting price. It is not.
Modelling is about stress-testing decisions before they cost you. Whether your token is at 4 dollars or 7 dollars in month three does not matter; whether changing your staking APY from 6% to 12% kills your collateral ratio absolutely does.
Time Wonderland's lambo counter assumed everyone would keep staking forever (3,3). One stochastic model would have shown the death spiral when even a small fraction sold; instead, the team found out live, on a bridge they could not unbuild.
Models cost five figures. The mistakes they catch usually cost seven or eight.
Read the full article:
顯示更多