In the race to deploy AI, organizations risk repeating mistakes from gaming, specifically the focus on actions per minute (APM) as a proxy for skill, which became a target and lost its value per Goodhart’s Law. A 2022 study of StarCraft players found no statistically significant difference in APM between expert and novice players, while a 2014 analysis showed older players compensated for slower reaction times with better judgement. Google DeepMind’s AI system AlphaStar performed worse when allowed to click more and averaged substantially lower APM than professional human players yet still won. The article warns that the same error is occurring in AI with "tokenmaxxing," the idea that more tokens is necessarily better. Anthropic has found that agents typically consume 4X more tokens than chat interactions, while multi-agent systems consume 15x more than standard chat, with users and companies bragging about token usage similarly to gamers.