Notable Point · 26:38-28:08

What an AI-run battle can and cannot test

AI control on both sides reduces player bias, while tactical choices, chance and imperfect ship modelling still limit the comparison.

Notes

Both squadrons remain under AI control so the comparison does not depend on a human player outperforming the computer. A human British commander would probably charge in line abreast to exploit nine forward-firing guns and present a narrow target angle. A lucky hit, compared here with the loss of HMS Hood, could also overturn the expected balance. Repeated engine damage raises a separate question about whether the simulated stern and machinery protection are adequate. These limits make the battle useful for design discussion without turning it into a literal historical experiment.