The Coriolis "force" isn't a proper force. It's probably better called the Coriolis effect. It's named after a French mathematician and engineer.
In a rotating frame of reference, like the earth's, when we differentiate a dynamical variable (like the position of an object -- here, Baumgartner) with respect to time we get two terms. The first is just the (infinitesimal) change in position with respect to the (infinitesimal) change in time in the rotating reference system. The second term, not present in an inertial frame (a rotating frame is not inertial (why?)) is proportional to the (instantaneous) angular velocity of the rotating frame with respect to an (instantaneous) inertial frame of the object. The second term is the Coriolis effect and it arises due to the fact that the even when the object isn't moving with respect to the non-inertial frame, it's position in the inertial frame changes due to the fact that the position vector (say, $\mathbf{r}$) changes due to the fact that we chose a moving coordinate system. The relevant equation is
\begin{align}
\frac{d\mathbf{r}}{dt}\Big|_I &= \frac{d\mathbf{r}}{dt}\Big|_N + \boldsymbol{\omega}\times\mathbf{r},
\end{align}
in a hopefully obvious notation. Of course, if you use a reference frame at rest with respect to, say, the center of the earth, you don't have any Coriolis terms in your equations. But you do have to take into account that the target, here the earth, is moving.
If you open a decent undergraduate or graduate classical mechanics book like Davis, Symon, or (egads!) Goldstein you can find a full development. But we can do an order of magnitude calculation, quick like. Baumgartner fell for 4m, 19s. Let's call this 250s, for ease in figuring. The magnitude of the angular velocity of the earth, which we'll take (incorrectly, but this gives an upper bound) as perpendicular to a position vector from the center of the earth to Baumgartner, is 2\pi / 24 hours. Since 24h = 86,400s the magnitude of the Coriolis effect on Baumgarnter results in an additional velocity of about 10^3/10^5 ($\omega$ in $s^{-1}$) * 10 (height in miles), giving about 1/10 of a mile/second. So Baumgartner would have had to correct for this effect. And this can be done with high precision.
I would guess that it would be a little tougher to correct for wind!