It might be appropriate to keep in mind that the design and operational rules, inter alia, are based on rational statistical history relating to accidents and incidents
The rules (and, by inference, the SOPs which, largely, are based on the rules) do not provide anything like an every day guarantee that an aircraft heading off to the holding point will, in fact, successfully find its way to the other end of the flight planned for the day. History certainly indicates that the rules get it right a very high proportion of the time so they should be viewed as being not too bad overall - but not infallibly correct and appropriate.
Even with the bit of padding and fudge factoring which is embodied in the rules, if circumstances on the day get a little too far away from the presumed story, then one has to accept that, on such days, one really ought not to have got out of bed. On flights for which such applies, the rules don't really give us much comfort and the captain, yet again, finds that when he (she) looks to the left to seek guidance, there is only the well-known face reflected in the side window - once again, the captain realises that he (she) can be in a very lonely place. Sometimes, the captain is called on to take decisions for which some of the rationale can only be guesswork - that's the nature of command, unfortunately.
Windshear is one of those considerations for which the rules can be found wanting. If a pilot cannot accept the reality that flying is not a perfect activity and that risk cannot be regulated totally out of the equation then, perhaps, that pilot ought to seek alternative employment.
As a side comment, knowing who folks such as safetypee and mad (flt) scientist are, might I suggest that their (very experienced and competent) technical/operational expertise warrants that their counsel be considered ...
I think safetypee summed it up, if I might be permitted to quote - "The critical decision in potential wind-shear conditions is in the choice to takeoff, opposed to delay."