Whether you check against theoretical or real blocks, you find the same.
If you randomly pick a point in time over the last year and see what the time between blocks was at that point, you get number around 2 minutes on average. Even though the average blocktime is only one minute.
I understand this as:
1. Take a point in time over some very long time period.
2. For that point, take the time difference between the last and the next blocks created (unless the randomly selected point in time is exactly a block creation time, in which case pick again)
3. If you repeat 1 and 2 some large number of times and take the average time between blocks, it will be about 2 minutes.
This has to be wrong, so have I just totally misunderstood you?