Sorry T-Mobile, AT&T and Sprint. Verizon has beaten you handily in an extensive series of tests of wireless network performance.\nRootMetrics, an independent testing company based in Seattle, performed thousands of tests in 125 metropolitan areas across the U.S., measuring speed, quality and reliability of voice calls, texting and data downloads. When the tests \u2013 all conducted in the last half of 2016 -- were tallied, Verizon came out on top or tied for first 658 times. The closest to Verizon was AT&T with 372 wins and ties, followed by T-Mobile with 270 and Sprint with 246.\n\u00a0Although Sprint fared poorly in many tests, the number-4 carrier earned high marks when it came to call reliability, a key measure of quality for consumers. Indeed, RootMetrics said Sprint was \u201coutstanding\u201d when it came to call reliability and earned an \u201cexcellent\u201d rating for that metric in all 125 metro areas it tested.\nMore wireless wisdom than the crowd\n\u00a0While no one\u2019s testing is perfect, RootMetrics now has millions of data points collected over the years, making its assessments far more valuable than the crowd-sourced data often presented by Verizon\u2019s rivals.\n\n[ The votes are in: Which mobile data provider is best? ]\n\n\u00a0When the results were released early this morning, the carriers quickly took to social media and email to tout \u2013 or quibble with \u2013 the results. Seizing on Sprint\u2019s improved scores -- coming in first is called an award \u2013 Sprint CTO John Saw wrote in a blog post: \u201cWe\u2019re excited to have achieved our best-ever performance in cities across the country, setting a new Sprint company record for highest number of Metro RootScore awards to-date.\u201d\n\u00a0T-Mobile CTO Neville Ray took a wack at Verizon, claiming that RootMetrics\u2019 tests are essentially obsolete, and that crowd-sourced data is far more relevant. \u201cThe network studies Verizon touts are like using CD sales to rank today\u2019s hit songs,\u201d he said in a blog post. \nSpeed kills and results vary\n\u00a0RootMetrics also looked at performance on a state-by-state level and found that, \u201cAT&T and Sprint are making gains in network performance.\u201d The mobile analytics company had some kind words for T-Mobile as well, saying that it \u201cdelivers fast speeds and strong reliability at the metro level.\u201d\n\u00a0Speed, however, was another story. Verizon left its rivals in the dust, delivering a median download speed of 20 Mpbs or more in 81 metro areas, compared to 52 for T-Mobile, 31 for AT&T and six for Sprint.\n\n[ How to save on mobile plans: Your guide to 16 no-contract carriers ]\n\n\u00a0There\u2019s another factor to consider. If you live in San Francisco, for example, it doesn\u2019t matter to you how well a carrier performs in Chicago or Boston. Fortunately, RootMetrics lets you drill down in the data and see how each carrier performed in specific metro areas.\n\u00a0And even that can be a bit misleading because metro areas cover many square miles and performance can vary quite a bit from neighborhood to neighborhood and even block to block. In San Francisco, where I live, Verizon is rated the best on network speed and when measured across the country RootMetrics found that it delivered a median download speed of 10 Mbps or faster in more markets than any of its rivals\n\u00a0But I\u2019m not close to a cell tower, so when I tested the download speed on Verizon\u2019s cellular network at home with the WiFi turned off it registered an anemic 4.22 Mbps. That\u2019s why I always advise people thinking of switching to another network to have someone who is a customer of that company to come to wherever they use their phone the most and try it out. If Verizon, for example, performs poorly in your home or workplace it makes no difference to you that it has won a boatload of awards across the country.\n\u00a0RootMetrics has divided its data into three reports you can access: metro, state and national. I think the metro results are the most useful for a consumer, which is why I focused on them, but there\u2019s interesting data in the other two as well.