I am brainstorming here and, depending on your strut/spring dimensions, this might be impossible but...
High spring rates might not matter if the spring wire diameter, free height and number of turns allow them to coil bind when they really clout the curbs. Put small tie wraps around the coil wire on both sides--if they get "snipped" off during a session, then the spring is collapsing to the point that the spring rate is going to "holy crap".
Also, since I seem to remember some sage suggesting that smart strut manufacturers use dimensions that bottom the external bits before the tender internals, you might be able to do the tie wrap trick on the shaft as well. If it gets squashed, then you are over-travelling the strut. If the strut (sans spring) can be compressed to the point that it stops with shaft still not inside the strut body (heck, or even if not), good bump rubbers are strongly advised. Some folks even think of these as "helper springs" that can provide an increased rate when things get rough. You should be able to hunt up a variety of lengths and rates, probably from your strut mfgr.
It would also be interesting to look closely at the break and see what it might have to say. If it is a uniform, rough surface, then it is likely a catastrophic failure. If it has two different surfaces (one rough like sandpaper and the other smoother, perhaps with "beach marks" that look like contour lines on a map, then it is spending some time as a crack before it gives up. This might suggest the failure mode and/or if you can catch the problem with routine magnafluxing. Point of interest--those two busted stub axles that that I mentioned earlier failed in exactly the same way, with cracks propagating to the center of the diameter and then big spontaneous failure, within less than an hour of rally stage time of each other!
If that is worth anything, remembering that I am working on a PhD in Education--NOT Engineering!
Best wishes,
Kirk