Below is our response to editors at Bike Radar regarding the editorial feature titled “Best aero road helmets: tunnel- and road-tested.” In addition to our response, we understand that Faster (the facility where the testing was conducted) has also submitted a response to the editors.
Giro appreciates the time and investment Bike Radar put into this story. It’s fantastic that the media is paying attention to aerodynamics and we welcome outside testing of our helmets. However, test results need to be based on a rigorous scientific protocol and attention to detail in order for riders to make choices based on the most accurate information possible.
After reviewing the editorial, we are concerned that important details may not have been given critical consideration:
1 The set-up was not consistent from Helmet to Helmet. The test pictures show that each helmet had a different taping over the straps. Testing has shown that strapping can affect the aero scores dramatically.
2 No Margin of Error was mentioned. This is very important. When testing helmets in a wind tunnel, two main components of error come from the operator assembling the helmet to the head form and the error that the tunnel registers. Multiple set-ups and tunnel runs with the same helmet at different times in the day need to be done and evaluated to understand the baseline and confirm the overall error. Giro’s extensive testing over the years, including testing at the Faster wind tunnel facility, has found that total tunnel data error is approximately 10g of drag; which means that any helmets whose drag is within 10g of each other are statistically equivalent. Therefore, Bike Radar’s 5 point scale is misleading. For example, the LG Course and Protone get a 3, the Giant gets a 2 and the Air Attack & Synthe get a 1 – yet all five helmets are within 9 grams of each other and therefore statistically equivalent. If the calculated error is much higher than 10g, it is difficult to make any valid aero claims.
3 The Head Form size may not be compatible with the Test Helmet Size in all cases. We’ve tested on this headform at Faster and know that it is on the large end of a Medium size for most helmet manufacturers. Most of these helmets appear to be Small size helmets based on the test pictures. We believe this to be the case as there is an extra, higher brow mark that was not on the head form previously, and the helmets clearly don’t fit correctly. Since the helmets do not fit the head form properly, they sit much higher on the head form than where they would actually be worn by a consumer. This could have a massive effect on the aero performance of each helmet versus if it was fitted to the correct size head form.
4 Averaged Drag is significantly different than Wind Averaged Drag. Simply averaging the results from 4 angles does not accurately represent real-world conditions. Riders statically encounter yaw angles of 5º more often than 20º . This is why Wind Averaged Drag, which is a weighted calculation, is used in our testing.
5 Testing at a single Head Angle. This is too limiting compared to real world scenarios. Rigorous testing requires at least two head angles to properly evaluate performance.
Again, we applaud the effort to perform verifiable testing. We at Giro hope Bike Radar can be part of our ongoing discussions with media and other manufacturers toward the end of designing a universal, repeatable, accurate protocol so that riders can make informed decisions when choosing their next helmet – or helmets!