You are coaching your own review process during a practice-exam edit pass; the flag notes are below. Use gain × gap ÷ time as the tie-breakable priority signal; do not plan more likely changes than the cap allows. Flag log: Card 1 (Data Insights graph): axis scaling was read too quickly; expected-gain units 11; confidence gap medium (2); review time 5 min; Card 2 (Data Insights sufficiency-style): the prompt asked whether data were enough; expected-gain units 7; confidence gap high (3); review time 2 min; Card 3 (Quant rates): units were converted under time pressure; expected-gain units 6; confidence gap high (3); review time 1 min; Card 4 (Quant algebra): one substitution step was uncertain; expected-gain units 8; confidence gap low (1); review time 2 min; Card 5 (Verbal CR strengthen): causal link was selected quickly; expected-gain units 5; confidence gap medium (2); review time 3 min. You have about 7 minutes and may change at most 2 answers; candidates not in the chosen sequence can be checked only if no change slot has been used. Which review sequence best balances gain, confidence gap, and time cost?