Here is my solution. I wouldn't have had a clue about how to go about this challenge without having attended the Machine Learning training at Inspire2019, so thank you very much Guiseppe and Fadi for starting me on the road to Predictive Analytics using Alteryx!
Reposting to add spoiler (tracker is not working correctly without it).
Got the answer based on the data, but since data for CHC is based on only 161 games, I created a second path to project raw number stats (but not percentage stats) for 162 games. Not sure if it is completely valid as some percentage stats would need to be adjusted based on new raw number stats, but alters results to project CHC to win 3 fewer games.
Very good challenge. Going to use this for training people to use the predictive tools.
I spend a significant amount of time looking for a way to "Select from a list".
i.e. once you have your list of top 10 variables how you you use that to select only those fields for the next step.
It's a minor thing but it felt a bit disjointed and undynamic to get to that point and then switch to selecting manually.
Sure there must be an easy way to do it but it is eluding me right now. Answers on a postcard please