We present a novel method for designing marine reserves that trades off three important attributes of a conservation plan: habitat condition, habitat representation, and socioeconomic costs. We calculated habitat condition in four ways, using different human impacts as a proxy for condition: all impacts; impacts that cannot be managed with a reserve; land-based impacts; and climate change impacts. We demonstrate our approach in California, where three important tradeoffs emerged. First, reserve systems that have a high chance of protecting good condition habitats cost fishers less than 3.1% of their income. Second, cost to fishers can be reduced by 1/2–2/3 by triaging less than 1/3 of habitats. Finally, increasing the probability of protecting good condition habitats from 50% to 99% costs fishers an additional 1.7% of their income, with roughly 0.3% added costs for each additional 10% confidence. Knowing exactly what the cost of these tradeoffs are informs discussion and potential compromise among stakeholders involved in protected area planning worldwide.