The truth about a lot of the big-name fantasy sports touts is that they argue incessantly in the comments section. Right now on the Tweeters there are two major projectionists -- actually three because two of them are fighting over the interpretation of the third projectionist -- about how much of a sample size you need before making definitive statements about improvement. The short answer: it's inconclusive. The long answer (which is me summarizing a decade of data collection): about 400 IP worth of pitching. This is why you see the ol' standards like Max Scherzer and Gerrit Cole always appearing at the top of the Streamonator and other Razzball tools -- we know how to expect Scherzer and Cole to perform, but we don't know how to expect Nestor Cortes to perform. You simply can't know the wild card pitchers when there's no data on them. Like I pointed out last week, if you based your team on Week 1 SP returns in 2021, you would have abandoned Max Scherzer after his 4 HR allowed outing and dropped him for Kyle Gibson, the eventual SP4 for most non-winning fantasy teams.
