I was just looking at
Moneypuck's SV% above expected statistic, and one thing that jumped out at me was that that the top-5 starting goaltenders in the stat are all starting goaltenders on teams that are top-6 in points percentage. The one goaltender whose team is in the top-6 in points percentage, but who also isn't in the top-5 in this state, is Martin Jones. 6th in the stat is Jimmy Howard, which is almost too funny of a coincidence to be true.
So, I went ahead and took a closer look at things. I looked at every starting goaltender's points percentage in decisions and tried to see if there was a correlation between their save% above expected and their winning percentage.
Indeed, there was a strong correlation; on average, for every 1% by which save percentage above expected increases, the goaltenders's winning percentage increases by 13%. I included a chart because it's easier to visualize things that way.
The x-axis is SV% above expected and the y-axis is points percentage. So, hypothetically, top-right is good goalies on good teams, bottom-left is bad goalies on bad teams, top-left is bad goalies on bad teams, and bottom-right is good goalies on bad teams. I highlighted Martin Jones.