Ever since John Legere took over as CEO of T-Mobile five years ago, the relationship between the “Un-Carrier” and its big rivals has been less than cordial. Verizon and T-Mobile routinely run ads slagging their opponents’ networks, picking on everything from extra fees to data throttling as a reason to switch.

But when T-Mobile put out an ad last year claiming that its LTE network was faster than Verizon, Big Red filed a complaint with the National Advertising Division. Months later, a ruling is here, and no one really won.

The verdict did come out in Verizon’s favor, at least on paper. The NAD “has recommended that T-Mobile USA, Inc., discontinue certain advertising claims made in television, print and internet advertisements,” — but T-Mobile has already discontinued those ads, and the NAD actually did provide some backing to T-Mobile’s speed claims, as well as what it’s been saying about coverage.

T-Mobile had two claims that were really at stake here: that Verizon’s network slowed down (and was slower overall) than T-Mobile in the period when the ad was aired; and that T-Mobile covers 99.7% as many Americans as Verizon.

The network speed claims were what were thrown out, but it was based on one particular objection. T-Mobile based its claims on crowdsourced data from OpenSignal and Ookla, the makers of Speedtest.net. As we’ve covered before, crowdsourced data has some methodological flaws and needs to be interpreted carefully. The data was from the time right after Verizon instituted unlimited data plans with deprioritization thresholds, and Verizon used the same argument it’s used before to challenge the data: customers were running more speedtests than usual to see if they were being throttled, thus corrupting the results and making them unreliable.

The NAD agreed with Verizon on that limited example, but didn’t say that crowdsourced data in general is unreliable and unusable in adverts. Which network is fastest according to that data really depends on how you interpret it, but let’s just agree that both T-Mobile and Verizon are very fast based on that data.

The coverage question is the second issue that the NAD looked at. It ruled that T-Mobile can claim nearly equal coverage as Verizon, based on the number of adults covered, but said that T-Mobile should remove maps of geographical coverage, so as not to confuse geographical coverage with population coverage.

Both companies are trying to spin the decision as a win. Verizon, obviously, is delighted that T-Mobile has received a slap on the wrist (the NAD is a voluntary industry body with no regulatory powers to speak of) for stretching the truth in adverts. T-Mobile, meanwhile, sees the decision as a vindication of the crowdsourced speed tests that consistently label T-Mobile the best network.

Really, the takeaway from the decision is that you should trust carrier adverts about as much as political stump speeches. Network testing is a complicated and nuanced subject, mired in statistics and methodology. The best use of data — be it from crowdsourced tests or repeated road testing — is to compare trends from year to year, not fight about which network is “best” in which location. Everything else is just spin.

View Comments