RChung

So, first a bit of historical background. Neither Tom Anhalt, nor Al Morrison, nor Andy Coggan, nor Jens Heycke, nor I work in the bicycle or the tire industry. We all started out because we raced and wanted to find the best tires for our own use. Since none of us worked in the industry and we weren't selling anything, we didn't need to keep anything secret -- but it also meant that we didn't need to produce papers for publication. I'm a professor in a completely unrelated academic field; papers on bicycles or bike racing or how to measure rolling or aerodynamic drag for a bike don't add to my day job and, to the extent they take away time from it, are actually a distraction. So, back then, we were *exactly* occupied with this question of the agreement or disagreement between roller and field tests. We were 100% concerned that roller tests were not realistic, and we were hoping to use field tests to prove that roller tests were inadequate. Most of this stuff was just presented and discussed and analyzed and shared on the Slowtwitch Forum. Here's the part you're not gonna like, and I don't really like it either: if you go to Slowtwitch and search the forum for posts by any of us with subject headings like "roller Crr" or "roller vs. field test" or variants of that, you'll see the discussions. So, you can look up the discussions -- they're still there, including the descriptions of what we found and discussion and arguments, but you'll have to read through a bunch of scattered posts to find where we concluded that *we were wrong* and the rank ordering of roller and field tests is almost always preserved. Look in the time frame around 2008 or 2009 or so.

Second, more to your purposes, we were racing, so almost all of our tests (except for a couple of tests of training tires) were for racing tires, and those were tires that we either personally owned or were sent to us for testing -- remember, I said none of us work in the industry. So, altogether, there were maybe a a dozen of these joint tests until we got tired (!) of doing them -- field testing is a lot more time intensive than roller testing and none of us were getting paid for it (cuz we're not in the industry) so once we saw that relative rankings hardly ever changed, we stopped that unless there was a really good reason. I was happy and relieved for that.

Third, one of the really good reasons to continue field tests, even though they're time-consuming, was because of what is presently called "impedance." Once again, if you go onto Slowtwitch and search for it, you'll find in what was then real-time, this puzzle that although the roller and field test rank orderings didn't change, we were interested in finding out why the raw Crr estimates differed. (If you listen to Josh's recent podcast on Models and Modeling, you'll see that we were still trying to understand the tire resistance model). So we were trying a bunch of things, including different temperature tests, adding texture to rollers, seeking out freshly paved vs. older asphalt vs. chipseal vs. cobbled pave' (I was living in Europe and had cobblestones outside my door), and changing tire pressure. It was during this phase of testing that Tom observed what we now call a "breakpoint" inflection with tire pressure. This was unexpected. So at first we (we including Tom) thought his experiments were in error. So he repeated them, and we were shooting emails back-and-forth trying to diagnose what he'd done wrong. Once again, the main point is that we were trying to disprove what we were seeing, not prove something we hypothesized the existence of.

Fourth, what I didn't know at the time was that people in the bike and tire industry were following what we were doing. Among those people was Josh (but there are others). It wasn't until years later that Josh started talking publicly about the tire and inflation testing they were doing for Paris-Roubaix. Tom still does gravel racing so he's been expanding his tests to gravel tires, but I'm still riding on the road so I only care about road tires.

This is a long way of saying that the evidence is really there, but not collected in a single paper, or a single post. We don't make any money off this so we don't have an incentive to collect the data into a paper. The people who do get money from this have an incentive not to give away their research. You have to paw through a hundred posts spread out over 3 or 4 years to get the full picture -- but it's there.