The problem is that if you have a job that runs on a day which is marked to create a synthetic full it will be marked as completed successfully when the retry runs even if the synthetic full was never created. There are two problems here:
1) The retry ignores the fact that the job required a synthetic full and doesn't retry it
2) The reporting doesn't see the failed synthetic full as a critical backup error.
Behold:

Nothing to see here move along...

Additionally: These backups specifically are being sent offsite. Due to the length of time it takes to complete a synthetic full, the requirement for Veeam to keep the connection open while the synthetic full takes place and Veeam's IMHO over sensitivity to connectivity problems I do on occasion have synthetic full jobs fail mid-way at other sites as well due to abritrary "connectivity" issues. Here too I queried support on being able to set the time-outs and sensitivity but was told this is hard-coded. Hopefully it's on the feature request list.