I'm not certain how to interpret this information. It is super helpful to know more about what's going on in our peer groups, but i'm not sure how to read this information as presented.
Thanks, Matt. It took a little thinking but I was finally able to ascertain that there are two ranges for each value and the mean. I actually appreciate seeing the range because it allowed us to have internal discussions about "how much value do we put on this metric" with regard to our own model health standards.
I think the community metrics are incredibly insightful data this information has helped us trim some metrics we felt weren't really "moving the needle". I would love to know if there's information about the "# of points" assigned to each metric. I like the check engine score as it allows us to weight metrics but there's very little guidance and documentation when you start getting into the weeds of model metrics.
Still think this is a really helpful resource, it's beneficial to see not just our number internally but to get a feel for where we are in the the peer group.
Dan,
In the above graphic, you have set your tolerances for "0 to 400mb" is "green", "400-800mb" is "yellow", 800mb and higher is "red".
For the "green/yellow" split, the community values range from 20 to 750, with an average of 258mb.
For the "yellow/red" split, the community values range from 100 to 9000, with an average of 715mb.
(I will say that - for the record, I don't really understand the high/low values there - so perhaps we shouldn't supply those and only supply the median/mean values).
