Gary Allen could probably answer this one better, but my intuitive answer is "compression".

The reason for this is that we're dealing with highly supersonic velocities, with intense and broad shockwaves that heat the surrounding atmosphere. Most of the trail you see is hot atmosphere, not meteorite debris. You can tell this from the characteristic excitation of atmospheric ions, rather than rocky material - this is why meteor trails at their best have colours that change with altitude/time as the atmosphere changes with depth.

The bolide experiences drag due to the shocks generating pressure gradients across it, rather than mere surface friction.

OTOH, the transfer of heat to the bolide is probably more by conduction, radiation, and friction (which all act to ablate the surface) until a critical density of the surrounding medium is reached.

A rule of thumb that I find useful for determining the outcome of reentry events is that a bolide will disintegrate when it has swept through a volume of the target that includes the same mass at itself. Thus, dense bodies dive deeper than light ones. Small bodies explode in the atmosphere, while larger ones impact the ocean/land and burrow to some depth before the final detonation.

In essence, at this point, the pressure gradients across the body exceed its strength and it tears apart into smaller bodies which each experience larger gradients and themselves fragment - a runaway process. So one also expects a loose aggregate of material to break-up higher than a well-bound body.

Different rules apply to spacecraft since they are specially shaped and engineered to beat some of the "rules of thumb", and enter on unusual trajectories that minimise drag.