On: FrontierCode: An eval to measure whether you would actually merge the code

[tedsanders]

Makes sense, thanks. I suppose error bars are tricky if trying to handle problem-to-problem variance, rubric-to-rubric variance, and run-to-run variance all at once.