To follow on from Jelly’s comment - it does sounds like a rollup issue on the group itself. Check the group IIS-Cluster in SCOM if you want peace of mind it’s not SquaredUp having a quirk.
Easiest way I’ve found is to create a new Dashboard in SCOM, then add in a widget. Call the widget “Groups” or something friendly, and add in the IIS-Cluster Group as an object. It should give you a friendly looking tile that will look green or red and you can use Health Explorer on it.
If I understand your issue correctly, could it be that the health roll up monitor is configured to a percentage of nodes having to be down instead of the best of/worst of algorithm?