The one time I drove my TL drunk (I don't drive even if I only have one beer) I did the same thing. I was within walking distance of the party and in my own neighborhood and I don't know why I drove but I never will again.

I came out the next day and one tire was on a curb, a couple beer cans had fallen out and were on the ground next to the car and it was about 10' too far back. Never again and I'm sure it was obvious to the neighbors.

This was the first thing I thought of after reading the first post.

 

yes, when working properly, i agree it shouldnt be possible to move. my question is why would it? what could be wrong to allow it to move? one of the 3 controlling shift solenoid valves stuck in some position other than the park control? the pcm not selecting the correct logic? a slightly misaligned shift cable? the parking pawl engaged in a possible location that allowed it to slip, then catch later? youre calling me a drunk and telling me not to get mad? i'm not an idiot and i thought someone here might be helpful. i dont park the car in the middle of parking lots and i dont hand out copies of my keys. its not like it moved a couple feet, it was completely in the middle of the lot. i now park it the same each night and put chock blocks an inch behind the wheels to check if it moved in the morning. but apparently nothing will convince you guys this actually happened.

 

Parking pawl is entirely mechanical. Not likely it rolled that far and then engaged. Think of a ratchet mechanism with at least 10 teeth per revolution of the diff. Factor in the axle ratio and even if the pawl went an entire revolution to engage, it would equate to less than a foot of travel.