<result> <column_name>previous_points</column_name> <family>additive</family> <subtype>intensive_additive</subtype> <confidence>0.85</confidence> <requires_user_confirmation>false</requires_user_confirmation> <sum_meaningful>false</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>mean</primary_reducer> <valid_reducers>mean, min, max, median, distribution</valid_reducers> <excluded_reducers>sum (summing point levels is meaningless), product (not a ratio)</excluded_reducers> <average_semantics>arithmetic_mean</average_semantics> </reducers> <inner_outer_measure> <inner_measure_reducer>min</inner_measure_reducer> <outer_measure_reducer>max</outer_measure_reducer> </inner_outer_measure> <multiplicative_partner_derivable>false</multiplicative_partner_derivable> <reasoning> Gate prior: DIMENSION table with no sequence axis makes multiplicative highly implausible. The column name "previous_points" indicates a snapshot level (a score or rating at a prior time), not a ratio or rate. The values are large absolute magnitudes (763-1837) centered far from both 0.0 and 1.0 (mean=1220, median=1183), ruling out ratio_multiplicative. These are intensive levels - summing "1004.75 + 1011.91 points" across entities is meaningless, but averaging them is sensible. Classified as intensive_additive with mean as the primary reducer. </reasoning> <table_type_conflict>false</table_type_conflict> </result>