It says those are averaged readings so who knows what they represent. No measurement or listening test is accurate enough to show these headphones to be different. Just look at the two channels in each one of my measurements. Surely those two cups are more different than these two headphones!
This is like arguing if the temperature in your house is 76.25 or 76.16 both by feel and your standard thermometer. You can't go there from here.
I don't understand how headphone users have gotten so married to measurements to these levels. Or hell, with their listening experiences.
Even if the variation within each of the models is so great that there are overlapping regions of their amplitude distributions for any given frequency, the distributions need not be the same, i.e. their means, medians and modes can all differ. To say that the models are generally different, i.e. have a different distribution, is not the same as saying that any unit of one model is different from every unit of the other model. Similarly, to say that one unit of one model is the same as one unit of the other model is not the same as saying that all units of either model are the same or even to say that the models are generally the same, i.e. have the same distribution. Conversely, however, to say that one unit of one model is different from one unit of the other model is not the same as saying that the models are generally different and to say that the models are generally the same is not the same as saying that any unit of one model is the same as every unit of the other model. In short: Specific sameness does not imply general sameness but only difference not being universal and specific difference does not imply general difference but only sameness not being universal. In other words, the question of general sameness or general difference cannot be answered by specific sameness or specific difference.
If you are referring to the units you tested, they may well be indistinguishable - for whatever reason. Per the above, however, this is insufficient to conclude that the models are indistinguishable in general.
Two years ago, oratory1990 has measured the same two HD600 and HD650 over multiple reseats using the same signal voltage. Even with positional variation and driver variation, from about 80 Hz to about 900 Hz, there is at least a 1 dB difference between any measurements and from about 400 Hz to about 800 Hz there is a difference of about 2 dB even between the closest measurements. Apart from the change in tonality, the HD650 will also be louder for the same volume setting.
Source. Revised to exclude measurements later.
Here are the averages of all his measurements of the HD600 and HD650 referenced to about 425 Hz and about 3.3 kHz, respectively. Taking into account the 1-2 dB variation from placement alone, the distributions are likely overlapping across the entire spectrum but still different.