The mode that exists right now, with the most likely path and outcome chose by the user, and
Another view which would be a binary tree where you can play with the probabilities of each path (not just the one you think is the most likely) to get your estimation of all the possible outcomes
This is low priority but the page could have:
This was previously proposed in https://github.com/PauseAI/pauseai-website/issues/38 but we went with the easier alternative of taking probabilities from the page altogether