Bruce Tognazzini invented the technique, and wrote about it in his "A Quiz Designed to Give You Fitts" on Ask Tog:
Question 6
What is the bottleneck in hierarchical menus and what techniques could make that bottleneck less of a problem?
The bottleneck is the passage between the first-level menu and the second-level menu. Users first slide the mouse pointer down to the category menu item. Then, they must carefully slide the mouse directly across (horizontally) in order to move the pointer into the secondary menu.
The engineer who originally designed hierarchicals apparently had his forearm mounted on a track so that he could move it perfectly in a horizontal direction without any vertical component. Most of us, however, have our forarms mounted on a pivot we like to call our elbow. That means that moving our hand describes an arc, rather than a straight line. Demanding that pivoted people move a mouse pointer along in a straight line horizontally is just wrong. We are naturally going to slip downward even as we try to slide sideways. When we are not allowed to slip downward, the menu we're after is going to slam shut just before we get there.
The Windows folks tried to overcome the pivot problem with a hack: If they see the user move down into range of the next item on the primary menu, they don't instantly close the second-level menu. Instead, they leave it open for around a half second, so, if users are really quick, they can be inaccurate but still get into the second-level menu before it slams shut. Unfortunately, people's reactions to heightened chance of error is to slow down, rather than speed up, a well-established phenomenon. Therefore, few users will ever figure out that moving faster could solve their problem. Microsoft's solution is exactly wrong.
When I specified the Mac hierarchical menu algorthm in the mid-'80s, I called for a buffer zone shaped like a <, so that users could make an increasingly-greater error as they neared the hierarchical without fear of jumping to an unwanted menu. As long as the user's pointer was moving a few pixels over for every one down, on average, the menu stayed open, no matter how slow they moved. (Cancelling was still really easy; just deliberately move up or down.) Apple hierarchicals were still less efficient than single level menus, because of the added target, but at least they were less challenging than the average video game.
Sadly, the NeXT folks, when coming to Apple, copied Windows, rather than the Mac, in designing the hierarchical menu interface for OSX. Today's Mac hierarchicals are just as difficult to use as those of Windows.
Fitts' law is not just about target size and distance; it's also about the number of targets. The more targets, all else being equal, the longer the task will take. Hierarchicals automatically add one extra target. Making it difficult to enter a second-level menu adds an additional target, the second-level menu itself.
With my hierarchicals on the pre-OSX Macs, in most cases, the user did not even have to think about targetting the second-level menu. The menu opened, and the user simply aimed for the desired item. The only time the user had to consider the second-level menu separately was when there were so many items in the menu that the one the user was after was way up or way down the list, out of range of the allowable pivot for entry. Even then, users would typically arc along a more radical curve to reach their items in a single motion, rather than breaking things down into the jerky Etch-A-Sketch types of moves users typically make with today's hierarchicals.
When designing a user's required motions, reduce the number of motions needed along with both distance and required precision for each motion, then consider how your proposed scheme maps onto a human's ability to make those motions.