I've used a strobe a couple times on tactical building searches, but since no one was actually there, I can't fully testify as to its effectiveness. I have also conducted some force-on-force training using strobe and we found it to be more effective than just the bright light when "getting off the X" and maneuvering against an aggressor who is charging you. Of course, the effectiveness depends on a couple things. First, The defender must already have the light in his/her hand and have a plan on what to do with it. This requires training, but just like any martial art or fighting style, isn't impossible to apply in the real world. Second, the light must be able to "instantly" access strobe. Since I value a high-mode (non-strobe) for most of what I do tactically, most lights won't work for this task. The multi-click strobe lights are too complicated and too easy to skip and accidentally activate another mode like 'low'. The bezel twist lights are too slow to activate or remove the strobe function if you need high.
I recently purchased two Klarus lights (XT11 and XT1C). These are the first lights I've used where I can truly see them being used in a reactive tactical situation due to the dual-tailcap switch allowing instant momentary strobe or high mode. Now the question becomes how effective strobe really is, what situations it may work for, and whether it is worth incorporating into the tactical tool-box. I know there are many who say they have heard of no law enforcement officers who have used strobe. Well, most LE are still carrying incandescent Mag and Streamlights, have no training on the use of strobe, and a built-in aversion to change. We have seen this happen with the introduction of ballistic vests, transition from revolvers to semi-auto pistols, ammunition technology, less lethal weaponry, particularly batons, weapon-mounted lights, and lasers. So just because it hasn't been used doesn't mean it doesn't have potential value.
I do not see strobe as a solution to every problem, or even a majority of problems, but it just might be the right tool for 'some' jobs, and I feel it warrants objective study to see if it works. The key is training and testing in a force-on-force environment, using lights designed to do what we need the strobe to do, and then seeing if it works in the real world. So my jury is still out, but I'm looking into it and won't dismiss it yet.