That's only a problem if the noise of wireless is higher than the signal (measured delay difference). And we can see from the data it isn't. Unless you're claim

That's only a problem if the noise of wireless is higher than the signal (measured delay difference). And we can see from the data it isn't. Unless you're claiming it can be consistently different by ~5ms between those two tests... but that's very unlikely. Like, "you should demonstrate it" unlikely.