I think Mark's clarification of those two issues is really helpful. Personally, i think there is a 3rd relevant issue, which is how well the card is able to simulate what should be an unpredictable event.
I agree that how far a unit is moved is not the critical factor.
At first I thought that the suggestion of playing the card at the end of the turn instead of the beginning was a good solution and I do think it might improve the situation. But, there remains a problem, in that a player may still exploit the card in a way that feels very artificial. Knowing that they hold the card to play at the end of the turn, a player might order a unit to make a sacrificial attack, then 'teleport' the unit to safety. Probably better than the other way round, but still very gamey.
A further suggestion, which might also address my issue 3, would be to make the card a play instantly. Since there are occasions when both players collect Tactics cards at the end of the turn, whether it was effectively played at the beginning or the end of the turn would depend on who picked it up (although it would be mainly the end). With the remainder of the card unchanged, this would still clearly allow both forms of exploitation of the card mentioned BUT it would be impossible to plan for it. Yes, someone would get an advantage, but it would be much closer to being a random event and, to my mind, much more exciting!
[Obviously, it could also appear in a starting hand, in which case why not simply play it before the first turn - a unit simply isn't where it should have been - lost en route, wrong orders...]
If everyone made war only according to his own convictions, there would be no war.