Yeah. I noticed this as well. Its clear that calibration needs to be done and can be done in the .adx file for ALDL Droid. I just haven't gotten to it. Realistically I dont know which is correct the SM or the ALDL droid maybe they are both correct. Come to think of it, I haven't compared ALDLscan's data - just haven't had time. There may be acceptable tolerances +/- 5% and resolution and with these older cars maybe it all relative and doesn't matter? Or the data sampled by the SM is one sample offset from the sample the ALDLdroid received? Until a calibrated DMM is placed on the O2 with a no kidding value who knows? The data for the SM is the same as that being received by ALDL droid - the BT adapter is just a pass through communication device. Nevertheless, I like seeing more than one to two things at once. I definately agree. This is a good compromise if you don't need high speed resolution, precision or logging. Even more practical if you have an additional Android device laying around.
I have not figured out the differences emperically. My theory is that the values differ due to sample time and processing. Because values are not the same at the time they are observed it does not mean the values are invalid. If you have an electronics background this should make sense if you do not have an electronics background this will take some time to explain.