Here Bill, I’ve helped point out the obvious you refuse to see:
Genes lost on the tree in common ancestors of multiple depicted species are colored red, gene losses not inherited from common ancestors I’ve colored orange.
As you can see the majority of the pattern (all the red blocks I’ve marked with rectangles) makes sense on a tree as single losses having occurred at varying times, in different common ancestors. And remember, it is entirely possible that numerous of these losses are actually still present in some degraded pseudogene form in many of the species listed on the tree, as the authors themselves state they counted pseudogenes as “lost”, even though technically a highly similar but nonfunctional Wnt DNA sequence might still be present in the species genomes.