I can only assume that the model has a hard time with the underlying metrics of VGK.
To be fair, if you look at them, we’re pretty middle of the pack on almost everything. But, at the top of the west in points. Average PP, Average PK, no 30 goal scorer, no 70 point scorer, no consistency in goal (in terms of who is playing – I’m sure the model can’t easily weight a bummer 1 goal who only had 10 regular season starts) etc. I’m not sure about the 5v5 numbers.
Either way, we definitely have a “sum is greater than the whole of the parts” team (which I think is best!) That I think challenges the algorithm.
1 Comment
I can only assume that the model has a hard time with the underlying metrics of VGK.
To be fair, if you look at them, we’re pretty middle of the pack on almost everything. But, at the top of the west in points. Average PP, Average PK, no 30 goal scorer, no 70 point scorer, no consistency in goal (in terms of who is playing – I’m sure the model can’t easily weight a bummer 1 goal who only had 10 regular season starts) etc. I’m not sure about the 5v5 numbers.
Either way, we definitely have a “sum is greater than the whole of the parts” team (which I think is best!) That I think challenges the algorithm.