Goodbye to the Magic Mouse

Today I returned to the Apple Store to exchange the Magic Mouse I bought a week ago for a more prosaic Apple (née Mighty) Mouse. Not because I regretted the extra expense of the Magic Mouse, and not because the Magic Mouse is missing the middle click feature. No, it went back because of two faults I can’t live with:

  1. it scrolls on its own, that is, when I’m not running my finger across its top surface; and
  2. it hurts my wrist to use it.

The first of these problems is the easiest to explain, if not to understand. I first noticed it while looking through a multipage PDF that I was preparing to use in an online presentation to clients. I was treating each page as a slide and had adjusted the window size to exactly match the page size. The idea behind this setup was that each page down keypress would bring me to the next slide in perfect registration with the window. I would use the mouse with Preview’s annotation tools to highlight certain features on the slides as necessary. What I found was that having my hand on the mouse would sometimes cause the slide to scroll up or down, out of registration with the window. At first I thought I was inadvertently touching the top of the mouse and running my finger along it, but no matter how careful I was, the unwanted scrolling persisted.

Later, I had even more trouble with Google Maps. I’ve always thought that Google’s choice to use the scroll wheel to control the zoom (instead of, say scrolling) was a bad design choice. It’s really miserable when your mouse sends spurious scroll signals. Since downward scrolls seemed to be Magic Mouse’s favorite, I kept finding myself zooming out to view an entire state when I’d just been looking at looking a single block.

Ghost scrolling is, I suppose, something that I might have anticipated happening occasionally. I’ve noticed similar behavior with track pads, and the Magic Mouse’s top surface is basically a track pad. It’s happened far too frequently to be acceptable, but it is at least understandable. My sore wrist, though, is just bizarre.

You have to realize that I have been using computer mice since early 1985, when I bought my first Mac. That’s nearly a quarter-century of mousing without any distress. And in that time I’ve used many different kinds of mice: roller ball, optical, one-button, two-button, three-button. Not one of them caused me any pain until the Magic Mouse.

I started using the Magic Mouse in earnest on Monday morning. (I had played with it a bit over the weekend, just to see how it would work with my old iBook, but it was only a few minutes here and there.) My wrist felt a little funny by the end of the workday Monday, but I wouldn’t call it pain. By Wednesday night, though, it was pretty bothersome, like a stretched muscle. I took my wireless Mighty Mouse (a “travel mouse” that goes with my iBook on business trips) in to work on Thursday morning and have been using it the past two days. The bothersome feeling in my wrist is gone.

So what’s wrong with the Magic Mouse? I decided to use some of the equipment here at work to compare it to my wired and wireless Mighty Mice.

Mouse Weight (g)
Magic 105
Mighty wireless 135
Mighty wired (w/cord) 91
Mighty wired (cord supported) 82

The Magic Mouse seemed a bit heavy, and it certainly is heavier than the wired Mighty Mouse; due, no doubt, to the batteries. But it’s not as heavy as the wireless Mighty Mouse.

Mouse Push resistance (lb) Friction coefficient
Magic 0.05 0.22
Mighty wireless 0.07 0.24
Mighty wired 0.04 0.22

A mouse’s resistance to being pushed across the pad is probably a better indication of its effect on my wrist than its weight is. I used a calibrated force gauge to push each of the mice across my mouse pad five times. The results in the table above are the median values. The friction coefficient is the push resistance divided by the weight (in consistent units). I should mention that these results are a bit dicey, as the gauge is not intended to be used to measure forces this low.1 Still, the results match my subjective sense: the heavier the mouse, the harder it is to push it. And it would make sense for Apple to have nearly equal friction coefficients across its line of mice.

The upshot of this testing is that the difficulty of moving the mouse around can’t be the reason the Magic Mouse made my wrist hurt. While it’s harder to move than the wired Mighty Mouse, it’s easier to move than the wireless Mighty Mouse.

I went on to test the click resistance. I determined where my index finger sat on the mouse and used the force gauge to measure the push required to click.

Mouse Click force (lb)
Magic 0.27
Mighty wireless 0.21
Mighty wired 0.21

Again, the results shown are the median of five tests for each mouse. Here we do see a difference that could, in part at least, explain my problem. The Magic Mouse takes about 25% more force to click than do the Mighty Mice. This matches, qualitatively, my subjective impression.

The last measurement I made is the height of the mouse at the point where I click it.

Mouse Height (in)
Magic 0.63
Mighty wireless 0.88
Mighty wired 0.88

Here again we see a distinct difference. The Magic Mouse has a low profile—lower, I suspect, than any mouse I’ve used before.

Another difference between the Magic Mouse and the others is one I wasn’t sure how to measure. It seems to takes more finger movement to scroll a small distance on the Magic Mouse than it does on the Mighty Mouse. Scrolling2 on the Magic Mouse doesn’t start until you’ve dragged your finger a bit, presumably to keep you from inadvertently scrolling with every tiny movement of your finger. Some of your scrolling motion, then, is wasted, which doesn’t happen on the Mighty Mice (or other scrollwheel mice). For long scrolls, this wasted motion is more than made up for by the savings you get with momentum scrolling; but I find myself doing a lot more small adjustments than big flings, so the savings are lost on me. (I also find the lack of immediate response a bit annoying.)

So I’m left with three possible reasons for the dull ache in my wrist:

  1. Higher click force.
  2. Lower profile.
  3. More finger movement when scrolling.

I’m not an orthopedist or biomechanical specialist, so I can’t say with any authority which of these three is the more likely cause. But if I had to bet, I’d put my money on the lower profile and the small but noticeable change it forces in the position of my hand and fingers.

Whatever the cause, the solution was to dump the Magic Mouse for older and less attractive technology.


  1. Yes, I know I shouldn’t report two digits in the friction coefficient when I have only one digit in the resistance. I just didn’t want it to look like the three mice had exactly the same friction coefficient. 

  2. I’m talking here about deliberate scrolling, not the ghost scrolling, which takes no finger movement at all.