int_driver_circuit_era_affinity

One row per (driver_id, circuit_id, era_key).

Part of the Skill family.

One row per (driver_id, circuit_id, era_key). Same shrinkage as int_driver_circuit_affinity but split on the 2022 ground-effect regulation boundary so a driver’s track record is never blended across incomparable car eras (pre2022 = 2018–2021, post2022 = 2022–2024). circuit_id is the physical-circuit identifier, so renamed events and double-headers at one venue pool into a single track record. Within an era the residual is car-removed, so ranking drivers by shrunk_affinity_s ≈ “equal-car track record” at that circuit. Shrunk toward the driver’s within-era global mean (Bayesian, prior = 5 virtual races). Negative = faster than field. Backs the Ghost Car Standings feature.

Lineage

Overview

Property	Value
Layer	intermediate
Family	Skill
Contract enforced	No
Tags	driver_rating, driver_affinity
Upstream	int_driver_race_skill_loro

21 tests guard this model - 18 generic, 3 singular.

Columns

Column	Type	Nullable	Tests	Description
driver_circuit_era_id		Yes	not_null, unique	Surrogate PK = hash(driver_id, circuit_id, era_key)
driver_id		Yes	not_null	Driver identifier
circuit_id		Yes	not_null	Physical-circuit identifier (slug of circuit_name). Collapses event-slug aliases (renamed events, double-headers) onto one venue.
circuit_name		Yes	not_null	Canonical circuit display name for this circuit_id.
era_key		Yes	not_null, accepted_values	Regulation era. pre2022 = 2018–2021, post2022 = 2022–2024.
era_label		Yes	not_null	Human-readable era span (‘2018–2021’ / ‘2022–2024’).
n_obs		Yes	not_null	Number of race-residual observations backing this estimate within the era
seasons_observed_n		Yes	not_null	Count of distinct seasons the driver raced at this circuit within the era
raw_affinity_s		Yes	not_null	Unshrunk observed mean residual at this circuit-era. Negative = faster.
shrunk_affinity_s		Yes	not_null	Bayesian posterior mean. Pulled toward the driver’s within-era global mean.
shrunk_affinity_se_s		Yes	not_null, expect_column_values_to_be_between	Posterior standard error in seconds. Non-negative.
shrunk_affinity_ci_low_s		Yes	not_null	Lower bound of 95% credible interval
shrunk_affinity_ci_high_s		Yes	not_null	Upper bound of 95% credible interval
affinity_confidence		Yes	not_null, expect_column_values_to_be_between	Data fraction of posterior mass. n_obs / (n_obs + 5). [0, 1].