Problem's making a failover vpn connection + static routes. Advice???

Question

So I'm in an interesting situation where I cannot seem to do something as simple as I could in cisco land. The first major issue i've run into in a while with utm. 
 
 Problem: I need a VPN tunnel of some sort as a failover of a privately managed VPN (MPLS) to cover provide outages. This is routed to by a static route on each gateway (BGP in the future). 
 
 Option1/Issue1: Create a site2site IPSEC tunnel. This would not work as you cannot set the ipsec-tunnel metric in Sophos UTM. So the tunnel will always have a lower metric that the static route to the MPLS router. 
 
 Option2/Issue2: Create a site2site Red Tunnel. I had high hopes for this but apparently (for whatever stupid reason) you cannot create a static route for the same destination network (even with a different metric) in Sophos UTM. 
 
 Does anyone have any ideas? At the moment I may be stuck going the IPSec-tunnel route and just turning the main offices profile off unless there is an outage. But i would rather have an option that was live and could be monitored side by side. 
 
 EDIT: Saw this https://community.sophos.com/kb/en-us/120239 but this won't help as the gateways, in theory, would be pingable even if the provider's network was down as the first hoop (the gateway) it is local. An even if we had a user unplug one end, there is no chance we could do it at our main office. 
 Seems very silly that Sophos does not allow us to configure multiple static routes or backup routes. I can understand not setting weight for site2site tunnels but routes, honestly what damage could really be done, and if there were people putting is silly configs then maybe they should not be fiddling with them in the first place. They are likely to do the same damage with 1 route vs multi anyways.

Thiago Palmeira · Answer

No, Caio Silva , the KB doesn't. What it does is to tell you to use Availability Groups as a substitute for routing metrics, which is a standard in every single network/routing device. 
 I'd already stumbled into KB 120239 before posting here; it is definitely NOT an acceptable substitute for routing metrics even though it CAN act as a workaround. Also, as suggested by BAlfson , opened opened a case in Sophos (Case 7784640 ) to understand why Sophos UTM has a Metrics field imbued in the Static Routing configuration if the product itself will completely refrain from it. 
 Answer from Sophos was:

I believe as you stated that this is "normal UTM behaviour" although it may not be what your used to seeing with other vedors and devices. This is how the UTM operates when it comes failover routes. In addition to the KB you provided you can also achieve this functionality by utilizing up-link balancing in conjunction with multi-path rules (for up-link routes). community.sophos.com/.../118457

So, conclusion is (TL;DR verion): 
 
 Sophos wants to re-invent the wheel when it comes to routing standards; 
 Sophos' Static Routing configuration section has a Metrics field with a troll tag (if the solution does NOT use it, why is it there?); 
 Sophos judge themselves righteous above all when it comes to protocol standards.

PS: Just to clarify as even some Sophos Support Engineers didn't get it: Availability Groups are a great functionality overall . It enables you to work with either addresses or FQDNs and does a great job in fallback scenarios in a wide variety of applications. It is jut NOT an acceptable substitute for routing Metrics ; not only because metrics are a standard or that they exist solely for handling route priority, but because they do the job way better.