As a generalisation, the most dramatic examples are because of continental drift, or tectonic activity if you like. Have a look at the Andes and Rocky Mountains. These are where the Western-moving plates are being forced up and over the subducted plates, the parts of the crust that are being dragged down into the mantle. 

Not all examples of rugged and mountainous topography are caused by subduction, or are on the coast. The Alps, the Atlas Mountains, the Himalayas and some other regions resulted from say Indian or Italian bad driving, and the inevitable collisions.

In the past though, before the days of Pangaea, a lot of the directions of tectonic activity were radically different from today's, so personally I should say that there is nothing special about those directions, apart from incidental current major trends. 

Much of any tendency that you see in our modern atlases is to a great degree coincidental.

On reflection however, there could be another factor.

This would be on far less of a grand scale than plate tectonics, but it might have something to do with the process of erosion at certain latitudes, particularly in temperate regions.

The thing is that our planet rotates from East to West.

Much of our air movement at lower altitudes is from polar regions towards the equator. As a result the Coriolis forces direct the moving air towards the East, in other words in the form of strong westerly winds. It goes beyond that of course, leading into the formation of cyclones and so on, though that is less relevant here.

Well, the upshot would be that in most of the regions in question most of the storms blow from the West and bring their erosive forces to bear from the West. Not only would they carve picturesque glens into the mountains, but marine wave action also would come from the West, undercutting headlands and creating cliffs and the like.

I am uncomfortable with the idea of taking this principle too far; it depends on a lot of factors. For example there would have to be mountains or at any rate raised topography to carve in the first place. However, it seems reasonable to me that there could be a significant effect and that the effect could produce a marked bias.

And don't forget ice-age glaciers. The Atlantic's north-eastern seaboard was heavily carved in the last ice-age.

But the west coast of Africa doesn't really fit the generalisation, does it?

Some of this may just be an apparent affect - what seem to be the most "jagged bits" of Africa are in the North (Atlas mountains) and the West (the Rift Valley); in Australia it's the West (the Snowy Mountains?) and Asia, Europe and the Antarctic don't appear to be be able to make their minds up.

Well P&W, first the niggle; the Snowies are in the east. You are right about the Atlas mountains though, but like the Alps and Himalayas (and of course, the mighty Gibraltar), they are comparatively fresh products (mostly a few tens of MY) of tectonic collisions. If you look at Italy where the Apennnines (bloody stupid spelling!!!) join the Alps you can see that the alps are a classic case of tectonic orogeny. Same for Atlas-Gibraltar, but the latter has been eroded through by probably the biggest waterfall on our planetary history, and probably more than once at that. Italy and central Europe have been arguing ever since about whose lousy driving caused the collision. The Gibraltar situation suggests that for once the Italians are innocent, that they were parked in the Med, minding their own business, when along came the Europeans! But then some malicious parties would have it that the Italian and African plates were moving North!

Isn't traffic management amazing?

Similar situation with India and the Himalayas, except that there it is obviously the Indians' fault. You can see the skid-marks on the ocean floor for thousands of km.

But the point I should have been making is that tectonic collisions and ruggedness go together more consistently than East-West influences.