Speaker measurements differences

These are different speakers. The port tuning looks different to me - you can see this in the impedance/phase plot.

Wilson makes a variety of different sounding speakers and this shows that that the same model may vary even from one year to the next. I don't think it undermines stereophile or soundstage - they just published what they measured - if anything it raises a question about consistency of design philosphy at Wilson. They make many absolutely exceptional speakers but like B&W I think they feel compelled to make changes constantly from year to year...this is probably more marketing related than anything else but it inevitably leads to this kind of result when testing - inconsistent response.

As some have pointed out - your room will probably make a bigger difference on the overall result. Nevertheless, I think this is the likely outcome of a marketing policy that seeks to roll out new models continuously. Cars have the same problems as they are tweaked from year to year...

To amplify what Newbee says: Get a sound pressure meter and a test recording and make a few static measurements with white and pink noise from several locations in your room. Do not be alarmed when your results show big, and I mean BIG, swings in response below 500 HZ.If you speakers were top drawer these results would not change much. A more sophisticated test unit, can make dynamic test measurements and do the math necessary to evaluate them. Multiple 10-20 db swings in room are usual.
Either trust you ears, or trust the people who can do math, but do not make buying decisions on stuff you cannot understand without some considerable technical background.

Usually differences in measured results are more likely to be the result of methodology than anything else. Were both sets of measurements taken the same distance from the speakers? If not, were the measurements taken from further away where the room effects might have influenced the results or the blending of the drivers in the speaker?

Everything else being equal I would accept the Stereophile reults over most any other reviewers. They have been doing it a long time and their fairly sophisticated.