Or perhaps, even more importantly, the 10T cog may be what someone needs to enable them to go from a 2x to a 1x drive so even greater aero savings to balance against the frictional loss that will only be in effect for short periods of the race. The aero benefit will be throughout the entire race.

I also wish someone would test the drag on the oversized pulley systems and compare that to the friction reduction. My guess is that they are slower.

---

-------------
Ed O'Malley
www.VeloVetta.com
Founder of VeloVetta Cycling Shoes
Instagram • Facebook

Where I wonder about 1X systems is the "effort" gap when the rear cassette jumps about 14% with each shift whereas on 2X systems it's only a 7% jump when using Sheldon's gear calculator. On a relatively flat or windy course, a 14% jump in gear ratio compared to a 7% difference is putting a significant physical demand on the engine/rider.

Personally I find it less taxing to be in the right gear at the right time rather than me having to adjust to the gear I'm in. With the 1X there will be times when one would be in the right gear for the engine. Yet, there will be times when you're not.

Has this impact on the rider/engine been considered/measured? It's similar to tire inflation and rolling resistance while ignoring what's happening to the rider without some deflection from the road.

PS The 14% and 7% are simply for illustration because up and down the cassette the ratios do differ more or less..

---

Indoor Triathlete - I thought I was right, until I realized I was wrong.

I have a fairly dense spreadsheet where I actually map the gear ratios assuming a comfort level of both 5rpm delta (+/-2.5rpm) and 10rpm (+/-5rpm) around a range of typical "preferred" cadences and find that the "practical" jump is fairly minimal given what is, in my *opinion* (though supported by research on preferred cadence values), a fairly normal cadence range for most riders.

I.e., it's pretty easy to soak up a lot of the jump simply by raising (or lowering) your cadence.

It's also worth noting that, from a practical usability standpoint, a 2x drivetrain doesn't actually have twice as many gears (your 7% v 14% example is mathematically flawed) because of:
- redundant ratios (eg. 53x15 is the exact same ratio as 39x11)
- unusable ratios (39x11 is also nearly unusable because of friction)

Really, "typical" 22-speed drivetrains have somewhere between 14-16 unique & usable ratios. And, within those, you can typically make up the difference by simply making some shifts in cadence.

But I don't really want to go down the physiological road here. This is more about whether or not from a purely mechanical & aerodynamic standpoint, does the 10T cog - which really is, I think, essential to 1X - really fall as short from a usability standpoint as Josh argues. I don't believe it does...

---

"Non est ad astra mollis e terris via." - Seneca | rappstar.com | FB - Rappstar Racing | IG - @jordanrapp

we're not even close to this discussion getting fruitful. first, drive trains are losing friction quickly. so much so we're starting to see something that actually must be considered in 1x, autofeed, because of the mass and inertia of large cassettes and the considerably less frictiony drive trains. (but i'm still a big 1x fan and am building up a 1x bike next week, woo hoo!)

second, the 10t cog sits out there, this object of attention, when in fact whether you've got a 48x10 or a 53x11 on your bike, these are the severe overdrive gears. you ride these gears in emergencies. about as often you split a pair of 8s and hit 21 on both hands. the 10t doesn't exist so you can ride it. it exists so that you can ride other gears. it's the blow up doll if you can't get a date. fixating on the 10t is overscrutinizing the attractiveness of the blow up doll.

---

Dan Empfield
aka Slowman

lol hahaha your analogies are... Agree with you while laughing wholeheartedly. Thanks.

---

Indoor Triathlete - I thought I was right, until I realized I was wrong.

wow, he goes to LA for one day and now it's all blow up dolls and gambling analogies

---

--------------------------------------------
TEAM F3 Undurance

You're right in many ways while I'm only trying to bring up one aspect.

When using Sheldon's gear ration, I got the 14% to 7% delta simply from the difference of going from a 16t to a 17t in the back instead of the 16t to18t. No switching to the smaller chain ring to do that (which is pretty tedious and beyond my ability without losing some watts in the transition).

The one aspect that I am noticing in FL on the flat and in the wind is that I'd much rather have my 12-25 than my 11-28 while staying in the big gear. Just a change of one tooth on the back rather than two feels so much better. Yes, I could address it through RPM; yet RPM is one of those efficiency factors that I like to maintain when I'm in a groove.

You've got this much better analyzed than I do and I'm relying more on perception than the data you have. I can agree with you when going uphill or downhill as our engines move through those changes much more quickly so the two tooth jump is usually desirable.

So you did take the engine/rider into consideration and have moved on to friction and aero as being more important. Thank you for your answer.

---

Indoor Triathlete - I thought I was right, until I realized I was wrong.

I broke the gearing calc part of my models and don't have time to fix it right now, otherwise I'd offer some info on the balance between friction and aero. So I'll just note that for an IM the premier tactical testing showed an advantage of 1x of ~40s in a low yaw scenario.
Chatting with a somewhat larger manufacturer a few weeks back they said that their testing had showed a minimal difference so hadn't been planning on a removable FD hanger for their next bike.

I think there are a couple of notable pros to believe they are slower in totality. Fwiw, I used my Berner 13t-15t because it also allowed me to run 1x while the stock Shimano cage did not. So if I needed a longer cage anyway, might as well fill the space with some larger pulleys.

---

---

Save: $50 on Speed Hound Recovery Boots | $20 on Air Relax| $100 on Normatec| 15% on Most Absorbable Magnesium

Blogs: Best CHEAP Zwift / Bike Trainer Desk | Theragun G3 vs $140 Bivi Percussive Massager | Normatec Pulse 2.0 vs Normatec Pulse | Speed Hound vs Normatec | Air Relax vs Normatec | Q1 2018 Blood Test Results | | Why HED JET+ Is The BEST value wheelset

I do 54t front and 12-25 11spd cassette. If it would be outside of that, I’d drop the front ring a tad.

You need a long straight downhill or ripping tailwind for 11 and 10t cogs. 100rpm in a 54 and 12 is 35mph.

Cmon. Just my stinky opinion.

I'm no engineer. Are you suggesting the increased frontal area of the 54t chain ring v 50t chain ring is an issue?

---

***

I am puzzled by this.

Could you clarify how/why/where drivetrains are getting considerably less 'frictiony' ? And also how the mass and/or inertia of large cassettes offer any advantages?

---

Advanced Aero TopTube Storage for Road, Gravel, & Tri...ZeroSlip & Direct-mount, made in the USA.
DarkSpeedWorks.com.....Reviews.....Insta.....Facebook
--

My summary of Josh's statements is:
If you're going 1x, then you could choose (say) 50 x 10-30 or 55 x 11-33 or 60 x 12-36, and you'd have the same gear range.
The first will have higher friction and the last will have higher aero drag (not just due to the ring but also the cassette and derailleur size).
The unknown at this point is where that tipping point is to achieve lowest overall drag.

Personal thoughts: I reckon it's time we* reevaluated 1/2" chain as the standard. Go to say 10mm pitch and you could increase tooth count, decrease articulation, increase contact points (lower stress per tooth=less friction), AND have small frontal area...

*and by we I mean Shimano, after Sram tries and fails first.

ceramic bearings. lighter/better oil. optimized chains and chain/tooth interaction. oversized bearings. you didn't get these memos?

i didn't say that the inertia of large cassettes offered advantages. i said that the combo of low friction drivetrains and high-mass cassettes creates an environment that, combined with higher friction cassette assemblies, may foster autofeed. events that cause the conditions for a cassette to stop, and freewheel, are moving in the direction of the cassette not stopping or, at least, there may be a delay before the cassette stops, causing the chain to sag between the cassette and the chain ring. this is my theory at least. something i'm paying attention to right now.

---

Dan Empfield
aka Slowman

Thanks Rappstar!! I love this, if we're going to argue, it might as well be public! Sorry I'm so late to the party!

I love your point about aero, and the criticality of that feature in making these types of decisions. I also think that your argument would make much more sense if 1x was being done for the purpose of aero, or even with at least an eye toward aero, which just doesn't seem to be the case.. If anybody were making a true aero gruppo with gear ratios optimized around a high speed use case specifically we would have to look at optimizing FA of all the components and their interactions around the gearing you are likely to need/want..

My experience here was informed primarily by my work on VumaChrono more than 12 years ago now.. That design was such that it could be single or double ring and we wind tunnel tested it as 53/39, 55/42 with front derailleur and also 53 only, 55 only without front derailleur.

In that testing we found something on the order of (at most) 2-4 watts advantage (30mph) to running single ring only without front derailleur, and that was over a pretty narrow range of yaw angles, at anything over 7.5 degrees you just couldn't really see a difference. The 2nd ring and front derailleur loss was way less than the gain of the domed aero crankset.. so if aero was really what people are after, the crankset design is a bigger deal than 1x vs 2x.. but I do give you that single ring w/o derailleur is a marginal advantage

So here are my 2 core issues with the 10t (and the 11t in 2x setups)
1. The 10t cog and really even the 11t cog in most setups is giving you the illusion of something powerful and awesome which you may use 0.5% of the time in exchange for something really great that you would use a lot. 48-10 or 53-11 will both produce just over 38mph at 100rpm (28mm tire 700c) now think about how often you are really putting the power down and going 38mph? 48-11 or 53-12 will do the same speed at 110rpm and save a few watts friction, so for an extra 10 rpm which can be trained.. you can save more watts than a $500 oversized pulley system gets you, AND you can then gain back a cog somewhere in the middle.. when I look at the Velonews chart on friction, the thing that stands out to me other than the friction differences is the sheer size of the gaps in the middle..give me an extra cog in the middle of one of those large gaps so I'm not hunting gears on false flats and moderate climbs and I'll happily find a more aggressive tuck on the downhill in order to hang when things go past 38mph

2. The big focus and advantage of 1x seems to be in the gravel/dirt space and as our tires get larger, these large gear ratios become ever less pushable.. that 48-10 or 53-11 on a 42mm tire is just a hair under 40mph at 100 rpm and nearly 44mph at 110rpm, and we can do the complicated aero math, but my thumbnail calc tells me that isn't happening whether or not the 10t cog is costing you 6 watts of additional friction or whether your front derailleur is costing 2-4 watts in aero. So for gravel/dirt/snirt/groad why not go with like a 50t and 12-36 and gain back a cog in the lower middle of the range? Or, if you aren't Ted King, go with a 48 with 12-34 and gain 2 new really useful gears that you didn't even know you needed and it will be a year of riding before you notice that the 10 and 11 are missing.

Full disclosure for all thoughts included above: I'm stuck in the ways and aesthetics of the old school.. I once used a 28 and have never owned anything larger than a 25, I'd probably choose knee surgery over a giant rear cluster (and am likely on that path already!) and have often muttered things like 'If Andy Hampsten could climb the Stelvio"... you get the idea) AND I have also been a vocal critic of the 11 tooth cog for a long time as well.. don't like it, don't use it, and outside of pro racing, pretty much nobody uses it and even at pro tour level racing it is used far less than people imagine. Lastly, I was in the room for much of the early 1x development stuff at SRAM and am really proud of what those teams have accomplished, but as with any development, doesn't mean I agree with all of the details!! Sadly, nobody would make the 12-23 12 speed straight block that I dream of but I do still have a prototype 13-23 11speed straight block that I still think should be commercialized in a e-tap drivetrain as it is magical!

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Has anyone asked the UK TTers if they've tested the friction/aero balance? Some of these guys run monster 60T and bigger front rings to reduce friction and stay in the middle of their cassette, and yet these are the same guys rocking sub-0.18 CdA. I'm sure Dan Bigham at least thought of cross-testing these two variables (FA vs drivetrain efficiency).

I have my reservation about the claim that the highest gear - whether a 50-10 or 55-11 or 60-12 - is "a blow-up doll" or otherwise an emergency overdrive gear. I use it routinely in a triathlon setting and I'm not that fast. Plenty of courses have long, mild downhills where one might not want to be at 110RPM for long periods of time, and if a tailwind is ripping you could be losing/gaining significant time. In a road racing or pure TT setting I understand the argument - but over a 5 hour ride my cadence preference is a bit lower and I find myself routinely using the 11t (53 front) if it's a rolling course to maintain my self-selected cadence.

---

ZONE3 - We Last Longer

Interesting discussion - though probably differences are tiny compared to other things.

Two things I'd like to point out that seems to have been overlooked:

1) When you have a 10T as the smallest cog, you're unlikely going to only have increased friction in that one cog. You're like running something like a 10-28 instead of a 11-32 and then using a smaller chain ring. This will mean that you will have increased friction in every gear, not only when in the 10T. So the aero difference might always be there, but so is the friction increase.

2) I have only seen very little data on this, but the aero difference between e.g. a 11-32 vs. 10-28 cassette seems tiny (when mounted on a wheel in a frame). The same for a smaller chain ring - in fact, I believe it was Damon Rinard who said that a bigger (aero) chain ring could actually be more aero than a smaller one. I guess if anyone has data it might be Josh, and would love to hear if he can share anything on this point (aero of casssettes and chain rings).

As a side note, I would definitely also be interested to hear or see data on larger RD cages/oversized pulleys. And what about RD's - better with the newer, smaller Shadow RD's that are closer to the frame/wheel, or the slightly bigger older ones, that are further from the frame/wheel, but maybe more sheltered by the leg/foot?

That's not really an answer. It's like saying that it's easy to make up for slow tires by simply pedaling harder. Sure, it's doable, but it's what people are trying to avoid in the first place.

Small-small on modern road drivetrains isn't an advisable place to spend a whole lot of time, but we're talking about a few watts. "Nearly unusable" is a pretty hyperbolic assessment.

On my Emonda (Shimano 50-34 11-12-13-14-15-17-19-21-23-25-28), I occasionally plunge to the 34-11 as a solution to the 16-tooth cog problem. I don't do it very often, but there are times when an effort is being held steady for a bit and both the 50-15 and 50-17 are less comfortable. Splits those gears nicely.

Alternately, you can use more pulleys!

Sometimes it's about being able to keep cadence low when you don't need to put very much power down. It's more relaxing to do push 100W at 40mph at 110rpm than 120rpm, and sometimes you need to keep the power flowing if someone is pushing the pace hard on a shallow downhill (either because they're pedaling hard, or because your group ride is graced with the presence of a tandem or a recumbent).

But yeah, for the most part road bikes are geared pointlessly high for the vast majority of riders. I think some of it is an issue of marketing... nearly all modern cassettes have small cogs of 11T or less by default, and cranksets with big rings smaller than 50T are traditionally held to be "not road cranksets", so road drivetrains on bikes with 25-622 tires implicitly have top gears of 120" and up.

To a large degree, a lot of folks just don't really understand the gearing math. Sometimes the conversations get plain weird.
I had a friend a while back who built up a very sweet 1x road bike, but he didn't want to use a cassette wider than an 11-32, and his crank had a 50T chainring on it. Now, consider the following:
1-I'm a lot stronger than him.
2-We self-select very similar cadences.
3-We ride the same hills as each other.
4-In those road hills, I get loads of use out of gears in the low-30" range (like 34-28).
One day, we were chatting after some group riding; he'd been riding in the slower pack, me in the faster. Unsurprisingly, he said that his 50-32 low gear was too high. Now, I knew from his ride data that he rarely had any need to pedal above the mid-20s mph range, so I suggested swapping to a much small chainring, at least down to a 44T but maybe a 42T or even smaller.
He was very surprised, insisting that a 42T chainring is far too small to do any kind of road riding on.
Now, on that day, I was riding my 1979 Fuji America, which currently has a high gear of 52-14. I pointed out to him that this is roughly equivalent to a 41-11, and that I'd only used the highest gear a little bit that day despite riding in a much much faster pack. I also pointed out that his highest cogs were beautifully shiny and clean. But he just couldn't wrap his head around the idea of a road bike having such a small big ring. (He finally tried it out a few months later, and said he didn't miss any of the gears he'd lost.)

In these days 60T isn't considered big... Marcin Białobłocki's rig:

https://dgtzuqphqg23d.cloudfront.net/...XUBApK-g-768x576.jpg


I believe that the ideal friction vs. drag scenario differs a fair bit depending on the race scenario. In a TT/Tri, you'll spend a whole lot of time at high speeds facing the wind. That being said, you're never holding back; so both drivetrain friction and FA will play a key part in saving watts.

Road Races are a different story though, as you're spending a lot more time protected in the pack and at a lower power output. There will be more and higher power surges, but the average power output is a lot lower than in a TT.

When comparing 1X and 2X, there's aloso the factor of added friction due to cross-chaining on hillier courses; so finding the ideal drivetrain set-up gets even trickier. My guess is that there's more to gain from 1X in a TT than a RR.

As we're not spoiled with better PM accuracy than +/- 1.5% in most cases, it's really tricky to measure small friction differences in real world scenarios. It would be interesting to see the aero cost of bigger chainrings and sprockets in the wind tunnel though. My bet is that the bigger sprockets won't add much drag in anything but zero yaw scenarios, especially if you're running a read disc wheel (TT/Tri bike) or have conventional bottles on a road bike.

And if you are spinning out your 53/12 on a downhill, it's probably time to stop pedalling, tuck and enjoy the rest!

I would love this.
Also forgive me phone posting but someone mentioned a 8mm pitch chain system? Well Shimano did try 10mm.
https://progettopistavintage.blogspot.com/...mm-pitch-series.html

I have something similar going on on my TT bike. In races, I use mostly the 52x19, 52x17 and 52x15. The jump from the 19 to the 17 cog is no biggy but for some reason the jump from 17 to 15 feels quite heavy. Feels like I'm grinding too much even though I'm still spinning at 90-95 rpm cadence, so I end up shifting back into the 17 cog spinning at 105rpm.. A 14-28 cassette seems perfect here. However when you start to think of 1X, on a hillier course or a flat course with the occasional short but steep climb, a 36x23 can be handy too.. So that would require a 32 or 34 cog for the 52t chain ring, meaning I would have to mix match a 14-28 and an 11-32 cassette or something similar..

As with all wind tunnel testing the more data you have, the more data you wish you had!!

My data on VumaChrono shows 4-6 watts advantage to the domed aero design depending on yaw angle and direction compared to the DA crank of the era: 2008/2009 when development was happening. Removing FD and inner ring could be another 2-4 at low yaw and doesn't seem to matter at higher yaw. 53 vs 55 was undetectable in what I have.

What I don't have is anything on derailleur cage length, cassette size, oversized pulleys, etc.. I do have one data point on the early Berner oversize cage with the wing shaped cage design. This looks to be ~10-15 grams of drag better than a standard SRAM Red cage of 2010, and I speak in grams of drag as 9grams is considered the floor of repeatability of that tunnel.. so it appears to be significant enough to call it an improvement, but just barely.

With this cage being about 30% longer but more aero in form than a standard cage netting you ~1-1.5 watt aero savings I would imagine that making the un-aero cage form 30% larger is probably costing you a similar amount, but hard to say without testing it.

---

http://www.SILCA.cc
Check out my podcast, inside stories from more than 20 years of product and tech innovation from inside the Pro Peloton and Pro Triathlon worlds!
http://www.marginalgainspodcast.cc

Thanks for the input, Josh. Your data seems to pass the smell test and jives with what little I've seen or heard in this regard.

I assume your numbers are all without a rider on board and I'd wager that the differences are unlikely to increase with a rider on board (and probably rather decrease, at least for the stuff downstream like RD, cassette, and cage). I remember Steve Hed mentioning results changed when testing disc wheels with vs. without a cassette, but the aero hit of a cassette is likely quite a bit smaller in a bike with a rider on vs. a stand-alone wheel (the cassette seems to be fairly sheltered in many frames).

Sounds like oversized pulleys might be close to a wash, depending on speed, yaw, power output, chain, and whatnot. I've thought about testing it for some time, but the truth is it would be rather time-consuming trying to work out what might be 1-2W faster or slower everything considered so something more important always takes precedent. That Berner cage is pretty nice - but probably quite a hassle to keep the pulleys clean!