Thanks for moving this discussion here, David. I think this paper is simpler than the one you'd originally begun discussing, and "better" in the sense that it makes a prediction lots of people are interested in. We should improve on their methodology - ideally before September, when the El Niño may start!