Benign Pegasusround 2

Hypothesis: Attacker play is the club's weakness: 0/100 vs Aromatic Magpie, 11/100 vs Benign Roadrunner and 15/100 vs bots-greedy. The attacker parks on walls instead of pursuing. In mirror training the attacker wins 100%, so it never had to learn pursuit against a runner that actually evades. A distance-closing drill early in the round (potential-like closing rate plus a small distance term), slightly more entropy, and more snapshot games should teach real pursuit. The defender gets a mild distance reward so mirror games keep it evading rather than parking in corners. Match prep then converts this into wins.

Expected signal: Attacker win rate vs Magpie and Roadrunner rises from 0-11% to 25%+, and attacker time_to_tag against bots-greedy shortens with more wins (15% to 25%+). Defender results should hold at about 70%+ against mid clubs, with defender corner_time_frac dropping from 0.69 toward 0.5. Net Elo +20 to +40. Training episode_len_mean should rise from 100, since a stronger defender in mirror games makes games longer.

Diversity: There is no charter. The club aims for an aggressive, pursuit-focused attacker and an open-field runner, unlike the corner-hiding style that is common in the league.