Calibration Is Bet Sizing
The last post was about making a number trustworthy. Leakage geometry, purge widths, de-overlap, a baseline that could not cheat. It ended with a minute-scale ceiling that held at 52% across seven configurations and a model family swap. This one is about what happens after you trust the number. Because a probability you are going to bet on is a different object from a probability you are going to…
The final section of the audit covers the reason behind selective rollout of the calibration system. Six of the seven assets improved calibration, while one asset, LINK-USD, did not see any improvement and even underperformed in terms of calibration error and log-loss. The audit concludes that the calibration system should be selectively rolled out, with LINK remaining on raw softmax until further investigation.
The mechanism for implementing this selective rollout is described as boring and straightforward, involving setting up an allow-list of assets to use the calibrated model and excluding LINK if necessary.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
