Single chainring drivetrains have taken a huge bite out of the mountainbike market, and now SRAM, the industry trend’s main backer, is pushing to put 1×11 on a variety of dropbar bicycles as well. As gravel bikes are akin to mtb for a number of reasons, it’s only natural that the growing market segment would be a ripe target for these products. A key element to getting the wide gear range out of a single chainring is SRAM’s innovative XD-style cassette, which allows a cog as small as 10 teeth to be used. Then the biggest cog is pushed to a startling 42 teeth, giving a total range of 420%. But as head Hugger Byron has stated, the 1×11 XD setup has some big jumps between the gears. He doesn’t like it, and said so in context of the experience. I’m a little more open-minded, probably because I’ve had an XX1 group on my MTB for the past two years. I am well-acquainted with how a 1×11 XD drivetrain feels. What I am going to do here is explain the difference in solid, quantitative terms. And that means dry-ass numbers, percentages, and figures. If you’re such a lightweight that you can’t handle some math, or if you’re not into um….deeper reading because you’re surfing the net on your smartphone while taking a dump, feel free to skip everything I’ve written down to the “summary.” No really, go ahead, I’m totally fine with it. Just know that I will punch you in the face if you want to object to my analysis without actually doing the reading. If you ask Mark V a complicated question, he gives you the fucking answer. I don’t have patience for those too lazy to read and think.
Okay, let’s get started. First, let’s define the scope of the question. The topic is drivetrains on gravel bikes. We are talking about product that is currently available as either on complete bikes or what could be assembled from current components. This means 11sp drivetrains with either double or single chainrings. Triple drivetrains are dead as far as the market is concerned, but I will make a minor detour later on, just to compare for fun. But make no mistake, no player in the industry gives a shit how great you think triples are/were. And anyone who even mentions half-step triples is going to get a punch in the face from me. Getting back to gravel bikes, such machines implicitly need a fairly wide range of ratios. Since width of range and the size of the jumps between ratios are directly proportional, we will limit the discussion to 11-28, 11-32, and 10-42 cassettes, cassettes with sufficient range for gravel bikes. No product manager is going to spec an 11-23 straight-block on a gravel bike, so there’s no reason to talk about it.
Next, what are the parameters that we will be comparing? We want to compare the size of the “jumps” between the gears, ie the percentage increase in wheel development (wheel rotation per crank revolution). Smaller changes in ratios between gears are assumed to be better, and the entire history of bicycle drivetrain development over the last 90 years follows that philosophy. Closely spaced ratios allow you to keep closer to a rider’s “ideal” cadence. If the jumps between gears are too wide, a rider’s pedal action may too often either bog down on too high a gear or spin out on too low a gear. The three cassettes in question share some cogs of the same tooth count, but obviously there will be some differences. So we will consider the increase for each shift on a given cassette as well as the average for all ten shifts (between 11 cogs). Further, a rider almost certainly spends more time in some cogs of cassette than others, so we will consider the average changes for the cogs clustered at the top, middle, and bottom of the cassette. In this case the frame of reference is such that “top” of cassette means the cogs that yield the highest ratios, in other words the smallest tooth counts. The percentage changes are calculated as increases from a lower ratio to a higher; that is, a higher tooth count to a smaller tooth count. For example, a shift from the 12T cog to the 11T cog would yield a 9.1% increase in wheel rotations for every complete rotation of the crankset. We will also consider the overall gear development, which is a combination of the cassette, wheel diameter, and chainrings. Gear development is expressed as gear-inches, and is calculated by the following formula:
X= (chainring tooth count) x (wheel diameter) / (cog tooth count)
in all cases wheel diameter is assumed to be 27” for simplicity
50×11 yields 123 gear-inches
(50teeth x 27”) / 11teeth = 123”
Above are two tables of numbers. Depending on the device from which you are reading, you might want to click on the table to link to a more readable version. Sorry, this website wasn’t really developed around itty-bitty charts and tables.
The upper table compares 11-32, 11-28, and 10-42 cassettes, the 10-42 cassette being the XD-style that would be used as the basis for a 1×11 drivetrain. As you can see the average increase for all shifts on the XD cassette is 15.9% versus 10.6% for the 11-32 cassette. The 11-28 of course has smaller jumps still, averaging 9.8%. But with the common 50/34 compact double crankset, the low gear is only about 33 gear-inches, which is probably not low enough for steeper climbs on a gravel bike with big tyres. Is a 15.9% jump inherently impossible for a rider? Probably not, since there’s a similar jump within the 11-32 cassette (going from the 22T to the 19T). But when we compare an 11-32 cassette to the 10-42 XD cassette, you can see that the XD cassette has fairly constant jumps throughout the range, whereas the 11-32 has tighter jumps in the middle and top end. Since the 11-32 cassette would be used with a double chainring, the rider can toggle between chainrings to keep the chain in the middle of the cassette as much as possible, where the middle five upshifts for the 2×11 setup is 11% versus 15.4% on the XD used by the 1×11 drivetrain. However, a long steep climb would force a rider into the bottom of the cassette, whether he was on a 2×11 or a 1×11. There the score is 13.9% for 2×11 versus 16.1% for 1×11. The difference is real, but whether it is unacceptable to the typical consumer is perhaps debatable. What is clear is that at the top of the range the difference between 2×11 and 1×11 is huge (8.1% versus 15.9% for the top four upshifts).
Now let’s look at the lower table of figures, which compares the range of several drivetrains as expressed in gear-inches. The typical 2×11 drivetrain on a gravel bike would have the 50/34 compact crankset with the 11-32 cassette; we’ll call this a compact double GS (GS for grand-sport, or a medium cage derailleur). You can see that the high gear on the compact double-GS is 123 gear-inches and that the lowest gear is 29 gear-inches. In reality, 29” isn’t that low if you have really long climbs and a bike with a load on it. If you needed lower options, SRAM allows you to use an 11-36 cassette, but that does push the jumps closer to the 1×11 XD drivetrain. Mathematically you could shift the drivetrain range lower if you could reduce the tooth count of the chainrings, but while the 50 tooth ring is probably way too big for the application, 34 tooth is the lower limit that common road cranks allow due to the 110mm bolt-circle-diameter. Indeed, component manufacturers have also noticed this issue, and led by FSA there should be a number of cranksets to hit the market in 2017 that eschew the 110mm bcd standard to allow 48/32 and 46/30 chainring combinations (see this earlier post about gravel-optimized cranksets). But at the moment, 1×11 XD drivetrains certainly offer the most convenient option to lower the gear range; you just swap out chainrings with no front derailleur to adjust. SRAM makes 110bcd ring for 1×11 in 38 to 46 tooth counts. With a 38t ring, the low would be 24”, and even though high end would be just 103”, arguably that would be more than tall enough for a lumbering gravel bike on unpaved surfaces. Understand that changing the size of the chainring does nothing to change the size of the jumps between the gears.
So what do we conclude? 1×11 XD drivetrains consistently have larger jumps between the gears when compared to the 11-32 cassette one would expect to used on a 2×11 gravel bike drivetrain. However, if you consider the range as distinct clusters of cogs within a cassette, the two cassettes differ most dramatically on the high-end and mid-range of the cassettes, whereas the jumps are actually not hugely different at the low-end. Given this information, one would conclude that 2×11 drivetrains would be superior to 1×11 on flat terrain where a rider could choose the either big or small chainring of the double to keep the chain in the middle of the cassette, which would keep the jumps small. It is worth mentioning that the middle of the cassette should also have the least drivetrain friction because the chainline should be as straight as possible, though this is true for 1×11 drivetrains as well. On fast sections, drafting in packs, or downhills, the 2×11 should have a distinct advantage. Considering the usage profile typical of a gravel bike, perhaps this isn’t an advantage that can often be exploited. If the consumer needs an especially low gear, 1×11 is the hands down winner, due to the flexible chainring options. But two or three hours of constant climbing can alter your perception of what is an acceptable gap between gears.
And now as promised, I will briefly talk about triples, the VHS player of road drivetrains. I chose Shimano’s Ultegra 6703, the last quality road triple of real relevance (which shows you what I think of Campagnolo Athena). The 10sp Ultegra triple had 52/39/30 rings, and the widest stock cassette was 11-28 (though a 12-30 was introduced at the end of the 10sp era after Shimano had started to phase out triples). The 10sp triple’s 11-28 cassette wins handily in the mid-range over the 11sp double’s 11-32 cassette, with just 9% jumps in the middle four shifts versus 11% (middle five shifts for the 11sp cassette). And the 10sp is waaaay better than 1×11 XD (9% versus 15.4%). In terms of overall gear range, the 29” low and 127” high of the 10sp Ultegra triple seems a bit top-end biased and ill-suited to gravel. Admittedly, 6703 Ultegra predates the gravel bike trend by several product cycles, so that demographic was not the target of Shimano’s engineers and marketers during development. But if five years ago there were a number of hacks that consumers and customer builders could pursue to widen or lower the gear range, that was a different era of product. And this isn’t a discussion about DIY kludges or eBay as source of vintage parts. Triples are dead and buried; get over it.
Shimano abandoned triples because compact double cranks fulfilled the needs of two-thirds of what was formerly the triple market. When SRAM and Shimano started making mid-cage derailleurs that could fit 11-32 cassettes, that allowed compact doubles to swallow probably 80% or more of the triple demographic. And the triple-loving outcasts leftover can get stuffed, because they weren’t spending enough money for Shimano to justify developing triple-specific integrated levers, derailleurs, cranks, and chainrings. The truth is harsh. And now SRAM is pushing 1×11 because that’s the turf they made their own, a way to pull the OEM rug out from under Shimano’s feet. And frame manufacturers are attracted to 1×11 because it allows them more design freedom when there’s no need to mount a front derailleur and only one chainring to crowd the right chainstay. It’s too soon to say that the front derailleur is an endangered species on road/gravel bikes, but 1×11 is no fluke. The gravel bike is very much an interesting realm of product development, and we can expect cool things in the near-future.
Other considerations? Weight may as well be mentioned. In not-so specifc numbers, SRAM’s long-cage X-HORIZON rear derailleurs are heavy when comparing examples of equal quality, and then they are somewhat more expensive. The XD cassettes are somewhat heavy and fairly expensive at the high-end. But the lack of shifting bits on the left lever and omission of front derailleur paints 1×11 as a win in both cost and weight.
In my opinion, 2×11 makes the most sense for the majority of the gravel bike category. At the custom frameshop where I help with the bike spec and do the actual assemblies on the finished framsets, there have been several bike designs that were tailored for 1×11, trading the front derailleur for big tyre clearance. However, if double cranksets better suited to gravel had been available (smaller chainrings but not so small as the typical MTB fare), the framebuilder and client together might have chosen 2×11. A light bike with 1×11 can be a lot of fun on choppy, variable terrain when you’ll be working rapidly across the cassette, but long, steady stretches into the wind or up a long pitch begs for finer steps between the gears.
For gravel bikes, 1×11 drivetrains have bigger jumps between the gears compared to 2×11 drivetrains, and this is most distinct in the highest gears and in the mid-range of the cassette. Theoretically, the 2×11 should be superior shifting (ie have closer shifts) mainly on flat and fast terrain where the top of the 11-32 cassette has much closer ratios…. or the choice of two chainrings allows the rider to stay in the middle of the cassette more often. On extended climbs, one would expect to stay at the bottom of the cassette where the difference between 2×11 and 1×11 XD drivetrains is smaller, but the longer the climb, the more you’ll want tighter jumps between the gears. However, when having a gear that is low enough becomes critical, the 1×11 drivetrain can easily accommodate a lower low gear by swapping the chainring to one of a lower tooth count, though at the expense of top end. There is no convenient option to do this with 2×11, until cranksets targeted at the gravel market become available with something smaller than a 34tooth inner ring. A 1×11 drivetrain may allow bike manufacturers to optimize the frame design in ways that would otherwise be restricted by the prerequisites of accommodating front derailleurs and multiple chainrings. These possibilities may sway a consumer’s decision to choose 1×11.
If you like simplicity of use, structure, and maintenance, 1×11 with SRAM’s 10-42 XD cassette is for you. If you know you’ll need a hella low gear and can give up some high-end gear, go 1×11…or maybe you should train and get stronger. If you like tightly spaced shifting, then choose 2×11…or you could shut up and learn to pedal with some suppleness.