The proper timing tools were used to time the engine. During the timing process everything looked good. Apparently, the crankshaft pulley bolt was not tightened enough and during the first startup the crankshaft moved. No other way to explain it.

After thinking about it, it would appear that a cam/crank correlation waveform is useless on these engines.

Looking at the Mazda timing technical document it is clear that during the timing process you are not actually timing the crankshaft, rather, you are actually timing the crankshaft pulley! The crank can still be off even if the pulley is timed correctly to the CKP sensor.

Insane as it seems, as far as the PCM is concerned, as long as it sees that the CKP sensor is generating a signal based off of the timing wheel and that the CMP sensor is also generating a signal off of the cam timing wheel - and they are in sync, then it's happy. So what if the crank is actually shifted and the #1 cylinder isn't actually at TDC!!

Otherwise, How is it possible that scan data can show the timing to be set properly at exactly the specified 8 degrees BTDC?

No timing codes. The only give away is that the thing runs like total crap.

Am I crazy, Or is this a legitimate possibility? Please correct me if I am wrong.