What's the cut-off for GP you are using?
Saros is 11th, and Ullmark is 6th with no filter.
I did 30 games as that is a #1 goalie most likely since there are about 60 games played.
Samsonov is in the same group as Saros right now. Ullmark is by himself.
The confusion comes because Mess stated that he was showing the top 5 in GSAx per 60, but he was actually showing the top 5 in total GSAx.
In total GSAx, Ullmark is #1, Saros is #3, Vasilevsky is #5, and Samsonov is #14.
In GSAx per 60 (15+ GP), Ullmark is #1, Saros is #4, Vasilevsky is #8, and Samsonov is #12.
While Vasilevsky has a much longer history at this level, and beyond, this season by GSAx per 60, Vasilevsky is saving about 1 extra goal every 7 games compared to Samsonov.
Saros is one of the best goalies in the world, but Samsonov is also having a very good season, and:
1. Even the best goalies have significant fluctuations in 4-7 game samples. You're not guaranteeing anything.
2. Saros was just reported to be untouchable, so not an option anyway.
Though it should be noted that Samsonov and Saros have the exact same SV%, so I'm glad we have an acknowledgement from a certain someone that SV% can be misleading.