Back to the stable nature theodicy:
i) To take a comparison, it's like healing and prayer. If God always healed in answer to prayer, then medical science would be pointless–and if God never healed in answer to prayer, then prayer (for healing) would be superfluous.
Occasional miraculous healing in answer to prayer doesn't make medical science useless. You don't know in advance which will do the trick, or whether either one will do the trick. Sometimes we pray for healing because medical science failed.
The dilemma for the stable environment theodicy is that it can't explain why God intervenes in some cases rather than others. So that must be supplemented by skeptical theism.
ii) I doubt it's possible to even guess at how often God prevents some natural evils. Physical events leave physical evidence in their wake, but nonevents leave no trace evidence of their nonoccurence. So what's the evidence that something didn't happen because God preempted it?
To take a comparison, consider those time-travel scenarios in which a Jewish scientist goes back in time to kill Hitler's granddad, thereby erasing Adolf from the space-time continuum. If successful, there will be no evidence that Adolf ever existed, because changing that one variable changes a host of affected variables. To be consistent, there must be corresponding adjustments.
Of course we know that's unrealistic: hence time-travel antinomies. But I'm just using that an an analogy to illustrate a point.
In the case of divine intervention to preempt a natural evil, that doesn't change the past, but prevent that past from happening in the first place–in which case, there's no empirical evidence that God intervened. We have no basis of comparison. We just have what actually happened.
It's not as if there's a gap or hole in the historical record or natural record when God prevents a natural evil. So in that sense, there's no direct evidence for divine preemption. Not like a missing folders in the filing cabinet between the As and the Cs where the Bs ought to be. All the "space" is filled.
So, from what I can see, there's no estimating the frequency of divine interventions in that respect. For all we know, divine intervention to prevent natural evils might be commonplace. It's imponderable.
I'm not saying it's never possible to identify divine preemption. In some cases you have plausible answers to prayer. But in other cases, no testimonial evidence will be available.