Gain Ratio: Is it an improvement, and if so, why resist change?

[Chance]

A two-parter question brought up in the “gearing a single-speed-bike” thread. First, why is it better than what we are using today? That was the main question asked.

If we set aside the fact that we’ve been using “inch-gears” in the US for about 100 years, it seems to me that it’s a better system in a couple of ways. It actually isn’t that different or more difficult to use except that it takes crank length into account. And obviously crank length affects total mechanical leverage between a rider’s legs and how far the bike travels. Normally with cranks limited to a range of 170 to 175 mm it’s not a big issue, but if comparing a little lady with 165s and a tall guy with 185s then the difference is significant. And inch-gears simply neglects that difference. Even if we don’t think it adds much value it’s hard to argue that it hurts anything to include crank length.

But more importantly, by cleverly including crank length into overall gearing the result is a number without units, so it applies the same in the US as any part of the world. These may be minor differences but if starting from scratch today they’d be seen by many as improvements, or a better and more comprehensive system.

So, other than you already know what you already know, is it a better system?

[Chance]

To the second question: Why resist change? Maybe it’s as simple as we don’t like to learn new things that we don’t have to. Even if the new “gain ratio” system is deemed incrementally superior to the old “inch-gear” system, the established format works well enough most of the time, so why bother? Foreigners can keep using their own metric system and riders with unusual length cranks can adjust on their own based on trial and error. Most importantly we already have a database of inch-gear numbers floating around in our heads that have meaning to us that would have to be discarded and replaced.

It’s a given we are not changing now or in the foreseeable future so let’s not get hung up on that. That part is clear. My question is whether we are staying with what we have today because it is seen as best available or simply because it’s the path of least resistance.

[witcombusa]

I know what a 70 inch gear feels like. I know that 78 inches fixed will not be fun on my river loop. It's kinda like knowing the languge. Gain ratio, I can translate it into something else and figure out what it means, but in my native tongue I can feel it and dream it.

You tell me a 5.2 gain ratio and I then have to convert it to gear inches before it means a thing to me. Besides crank arm length for 95% of the people is within 10mm (half that for most)......I can get that with a tire pressure adjustment!

It ain't broke, so why fix it?

[Jaq]

I'll admit I'd love to see a change away from anything to do with inches, if only because we ('mericans) absolutely need to be on the metric system.

The problem with gain ratio is that it does take into account crank-arm length (CAL) and claims (as I understand it) that longer CAL increases the length of travel of the drive wheel. But CAL has nothing whatsoever to do with the linear travel length of a drive wheel, because one revolution of the crank arm is one revolution of the crank arm, whether the crank arm is 170mm or 210mm.

One difference, obviously, is that the circumference of the circle described by the 210mm arm (at the pedal axis) through one revolution is greater (2*3.14159*210 vs. 2 *3.14159*170). The other difference would be the amount of force required; a 210mm lever arm requires slightly less force.

What determines the length of travel of the wheel per revolution of the pedals is the radius of chain ring, the radius of the cog, and the radius of the drive wheel.

[Chance]

It ain't broke, so why fix it?
Have never been a great fan of that logic because if applied to too many things we might still be back in the Stone Age. There was nothing wrong with steel frames when titanium was introduced, and nothing wrong with titanium when carbon came along (OK, skipped over aluminum). Nothing wrong with 5-speed freewheels either, and so on. There is still nothing "broke" with steel but we have other options many prefer for certain applications.

My guess from your reply is that having to learn the “significance” of values within the new rating system is the biggest challenge. More so than having to learn how to calculate it or look it up. And that’s expected.

[Chance]

I'll admit I'd love to see a change away from anything to do with inches, if only because we ('mericans) absolutely need to be on the metric system.

The problem with gain ratio is that it does take into account crank-arm length (CAL) and claims (as I understand it) that longer CAL increases the length of travel of the drive wheel. But CAL has nothing whatsoever to do with the linear travel length of a drive wheel, because one revolution of the crank arm is one revolution of the crank arm, whether the crank arm is 170mm or 210mm.

One difference, obviously, is that the circumference of the circle described by the 210mm arm (at the pedal axis) through one revolution is greater (2*3.14159*210 vs. 2 *3.14159*170). The other difference would be the amount of force required; a 210mm lever arm requires slightly less force.

What determines the length of travel of the wheel per revolution of the pedals is the radius of chain ring, the radius of the cog, and the radius of the drive wheel.
Jaq, it wasn't meant to imply that. A “gain ratio” is nothing more than how far your bike moves forward compared to how far your feet/pedals move during the same period. It would apply the same whether you calculated it based on spinning the cranks one revolution or 1,000. To me it seems more relatable because it has direct significance. A gain ratio of 5 means my bike will travel 5 times farther than my legs. For every foot, meter or mile my legs travel spinning the pedals the bike will travel 5 feet, 5 meters, or 5 miles, respectively. To me it relates well to leverage in the sense that we give up force to go faster.

On the other hand 66 inch-gear doesn’t mean 66 of anything to me. It’s just a number that doesn’t indicate how far the bike travels in an absolute basis or relative to anything else. Except for another inch-gear number. It’s not like the bike travels 66 inches for every revolution of the cranks, is it? Yeah, 66 inch-gear is less than 88 and more than 44, but beyond that it doesn’t mean much more.

[Don49]

On the other hand 66 inch-gear doesn’t mean 66 of anything to me. It’s just a number that doesn’t indicate how far the bike travels in an absolute basis or relative to anything else. Except for another inch-gear number. It’s not like the bike travels 66 inches for every revolution of the cranks, is it? Yeah, 66 inch-gear is less than 88 and more than 44, but beyond that it doesn’t mean much more.
I thought that gear-inches was a measure of the wheel diameter of the old high wheel bicycle. Perhaps not so relevant today, but certainly not without meaning.

[Chance]

I thought that gear-inches was a measure of the wheel diameter of the old high wheel bicycle. Perhaps not so relevant today, but certainly not without meaning.
You are correct Don, and it still doesn't mean much to me as it relates to how far a bike travels per crank revolution. We know it's equivalent to riding a tricycle with a 66-inch diameter front wheel, but unless we multiply that by Pi it doesn't measure distance covered per revolution. That was the point of my reply to Jaq.

[witcombusa]

Have never been a great fan of that logic because if applied to too many things we might still be back in the Stone Age. There was nothing wrong with steel frames when titanium was introduced, and nothing wrong with titanium when carbon came along (OK, skipped over aluminum). Nothing wrong with 5-speed freewheels either, and so on. There is still nothing "broke" with steel but we have other options many prefer for certain applications.

My guess from your reply is that having to learn the “significance” of values within the new rating system is the biggest challenge. More so than having to learn how to calculate it or look it up. And that’s expected.

That is to my point, there was nothing wrong with steel frames, still isn't.
Alu, Ti and Carbon are simply different materials. All with advantages and pitfalls. Not simply better. I am very happy to ride steel for another 30 years thanks very much.

Gear inches tell me what I need to know when setting up new builds. A "new" systems can't even do that for me without converting it back to inches.

[christian]

22 degrees Celsius is just a number too. But it still tells me what cycling attire I need to wear.

Gain ratio seems like a solution looking for a problem. Anyone who is cycling savvy enough to know about gear inches understands the crank ratio limitation and also understands that 66" is a good fixed gear. The latter is the type of intuited knowledge that makes the current system preferable.

[flydhest]

You are correct Don, and it still doesn't mean much to me as it relates to how far a bike travels per crank revolution. We know it's equivalent to riding a tricycle with a 66-inch diameter front wheel, but unless we multiply that by Pi it doesn't measure distance covered per revolution. That was the point of my reply to Jaq.

err, if you know the diameter of the wheel don't you have to know how far it goes with a revolution of the crank? High wheelers were all fixed gears with the crank and the wheel sharing and axle. One revolution of the wheel is one revolution of the crank. The bike travels the circumference of the wheel for each rotation of the crank.

[Jaq]

Oh, I agree that gain ratio results in a simpler number. But it's a simplistic number as well, and for various reasons isn't terribly useful.

For instance, while using crank-arm length does yield fixed numbers, the variance in numbers from commonly used crank-arm lengths is all but insignificant. Between 170mm and 180mm crank-arm lengths, the final numbers (5.58 and 5.27, respectively) are only about 5% apart, and then only because we're fudging significant figures out to the 100th's place. A 172.5mm crank arm yields a result of 5.52, ~1% different than a 170mm crank arm, again because we're arbitrarily (and incorrectly) fudging sig-figs.

Further, tying the measurement to crank-arm length ignores the variable physiological factors that go into determining correct crank-arm length for any given individual rider, and the ability of that rider to turn the crank. A "5.52 gain ratio" (for a given chainring/cog combination) is arguably a useless number because it means something entirely different to a rider with a 30" inseam than it does to a rider with a 34" inseam, or to a rider who's just coming off a bout of the flu.

Development in Meters is arguably the best system, because it completely factors out the human element, and relies on industry standards. The math is certainly no more complicated than the Gain Ratio system's. A small irony here is that one of Sheldon Brown's complaints about this system that "the resulting value is a less-convenient number to work with, a single digit plus two decimals." is exactly how his own gain-ratio system is expressed.


Jaq, it wasn't meant to imply that. A “gain ratio” is nothing more than how far your bike moves forward compared to how far your feet/pedals move during the same period. It would apply the same whether you calculated it based on spinning the cranks one revolution or 1,000. To me it seems more relatable because it has direct significance. A gain ratio of 5 means my bike will travel 5 times farther than my legs. For every foot, meter or mile my legs travel spinning the pedals the bike will travel 5 feet, 5 meters, or 5 miles, respectively. To me it relates well to leverage in the sense that we give up force to go faster.

On the other hand 66 inch-gear doesn’t mean 66 of anything to me. It’s just a number that doesn’t indicate how far the bike travels in an absolute basis or relative to anything else. Except for another inch-gear number. It’s not like the bike travels 66 inches for every revolution of the cranks, is it? Yeah, 66 inch-gear is less than 88 and more than 44, but beyond that it doesn’t mean much more.

[Chance]

err, if you know the diameter of the wheel don't you have to know how far it goes with a revolution of the crank? High wheelers were all fixed gears with the crank and the wheel sharing and axle. One revolution of the wheel is one revolution of the crank. The bike travels the circumference of the wheel for each rotation of the crank.
The point is that Gain Ratio as defined by Sheldon Brown doesn’t have anything to do with crank revolutions. It’s strictly a measure of bike gearing that happens to be very similar to gear-inches; just that crank diameter (or radius) is also used compared to driven wheel diameter (or radius).

So yes, a big-wheel bike has a gear-inch that is the diameter of the tire. It also has a Gain Ratio too. Because the drive ratio is 1:1 due to being direct driven, the Gain Ratio is tire diameter divided by crank diameter.

By the way, help me understand something here. Gear-inch estimates have nothing to do with crank revolutions either. It’s ring size divided by cog size and that multiplied by driven tire diameter. At no time does crank revolution get input into estimate. So why do revolutions keep coming up unless there is something more that is not obvious? :confused:

[palincss]

Gain ratio gives me nothing extra as all my cranks are the same length, and like other posters in this thread, I've been working with gear inches long enough (going on 40 years now) that the values are meaningful to me. Sheldon's gear calculator is about the only place where I've ever seen gain ratio. There's a selector to let you chose gear inches instead, and that's what I do.

[christian]

Ok, you don't want to discuss, you just want to insist that gain ratio is better? :crap:

[flydhest]

The point is that Gain Ratio as defined by Sheldon Brown doesn’t have anything to do with crank revolutions. It’s strictly a measure of bike gearing that happens to be very similar to gear-inches; just that crank diameter (or radius) is also used compared to driven wheel diameter (or radius).


What I quoted from you was that it doesn't tell you anything about how far the bike moves for a rotation of the crank. I was just pointing out that what you said wasn't right.


So yes, a big-wheel bike has a gear-inch that is the diameter of the tire. It also has a Gain Ratio too. Because the drive ratio is 1:1 due to being direct driven, the Gain Ratio is tire diameter divided by crank diameter.

By the way, help me understand something here. Gear-inch estimates have nothing to do with crank revolutions either. It’s ring size divided by cog size and that multiplied by driven tire diameter. At no time does crank revolution get input into estimate. So why do revolutions keep coming up unless there is something more that is not obvious? :confused:

I think the way gear inches are supposed to work is to be an analog to the measurement from highwheelers, so I think it should be the case that the distance you move for one crank rotation is the gear inches. The gain ratio takes into account that a longer crank arm means the pedal moves a greater distance with one revolution, of course.

I don't see this as a huge debate in any event. All my road bikes have a 53/39 chainring, so all I think about is the front ring-back cog combination. If I ever set up a compact, which I might, I'd look things up once, maybe, to see how low things were, but mostly I'm just thinking of getting a 50/34 to go with my 13-29 cassette for the bike I keep in Colorado and calling it a day.

I don't know how often people have to discuss these things, and then, when they do, all that really seems to matter is having a common frame of reference. I'm not convinced that gain-ratio is that much (if any) of an improvement, but if it is, it is small. The cost to learning something new, also small, so the cost-benefit analysis (worth roughly twice what you paid for it) is a wash. I think that is the answer. The benefits are small. The costs are small. The benefits may well be smaller than the cost, but the effort to figure that out, combined with the cost of switching, exceed the benefit.

[Jaq]

By the way, help me understand something here. Gear-inch estimates have nothing to do with crank revolutions either. It’s ring size divided by cog size and that multiplied by driven tire diameter. At no time does crank revolution get input into estimate. So why do revolutions keep coming up unless there is something more that is not obvious? :confused:

Actually, they do in a somewhat obscure way. Gear inches are the ring/cog ratio times the wheel diameter. If one takes it a step further and multiplies the result by Pi (3.1415), one gets the number of inches the wheel will travel for each revolution of the cranks. The Gear Inch system simply leaves out the Pi step, which in my estimation is really uneccessary.

After all, there's always room for Pi.

[fiamme red]

After all, there's always room for Pi.Mmm... pie. :)

http://lh5.ggpht.com/_JbfKdSBkoWQ/TMMi322OtpI/AAAAAAAACQY/_kEjVn9HlBM/lisa-homer-pi(e)%5B7%5D.jpg

[Ralph]

Nothing new in thinking about gain ratio. I've always thought that way. I'm 5' 10". but like to use 170's where some of you with my leg length like longer crank arms for more so called "Leverage". So I limit my big ring to 52, instead of 53, to "gain" back the difference (and a little more actually) between 170's and 172.5's (as an example).

[Chance]

Ok, you don't want to discuss, you just want to insist that gain ratio is better? :crap:
Christian, don’t know what you mean by this. Was just trying to limit the number of my responses somewhat in that there were too many already. Besides, everyone came out on the opposite side which makes it awkward for me to respond. A large “con” majority was expected but not to this extreme. It’s also obvious that it seems a superior system to me but that can be seen from the OP. Did not hide that in any way. Granted they all work, just some better than others in my opinion.

For what it’s worth; almost responded to your temperature post (copy below) because as useful as a temperature is it can be fine-tuned more by also expressing wind chill, right? Or by stating whether it’s sunny or rainy. The same temperature can feel a little different depending on other variables which can be incorporated into a more comprehensive and informative number (i.e – wind chill).

Also with temperature we have to state whether it’s C or F. Although it’s not possible, wouldn’t it be great if there was one number that worked for everyone whether English or Metric?

To me gain ratio is very similar to above temperature example. It fine tunes inch-gear and also makes it a dimensionless number which seems like a plus to me.

Not trying to convince anyone of anything. Yes, it seems better to me. No, don’t care who uses it. Was just curious about human behavior resisting change in general.

22 degrees Celsius is just a number too. But it still tells me what cycling attire I need to wear.

Gain ratio seems like a solution looking for a problem. Anyone who is cycling savvy enough to know about gear inches understands the crank ratio limitation and also understands that 66" is a good fixed gear. The latter is the type of intuited knowledge that makes the current system preferable.

[witcombusa]

Not trying to convince anyone of anything. Yes, it seems better to me. No, don’t care who uses it. Was just curious about human behavior resisting change in general.


Another way to look at it is "Why do people jump on anything new and different simply because it is"?

I'll tell you that the crankarm length issue does not "give back" enough to learn a new language.

The temp thing is an interesting point because if you tell me it is 20C outside, it also means nothing to me without converting it back to F. There is nothing wrong with familiar. It's been the cycling standard in the english speaking cycling world for over a century because IT WORKS.

[Mark McM]

I'm really suprised that nobody has pointed this out already, but the gain ratio adds an additional piece of useful information - while gear inches and development can tell you how far you move with each revolution of the crank, gain ratio tells you how hard you have to push the pedals to overcome a given drag load.

We all know that to go the same speed in a lower gear, you have to move your legs faster, but you don't have to push as hard. But how much less do we have to push? By measuring gearing with gain ratio, we can get a direct measurement of how hard we need to push the pedals. Two gearing setups with differing wheel sizes, sprockets, chainrings and crank lengths but with the same gain ratio requires the rider to push the pedals with exactly the same force for the same output (output being wheel torque, as required for a given terrain and speed).

It's already been pointed out that the common range of crank lengths, there is little difference in mechanical advantage between cranks, so the gain ratio for a given chainring/sprocket combination varies little between different cranks. But if cranks were available in wider range of sizes, the total leverage between drive trains with different cranks could vary quite a bit even with the same gearing combinations so the gain ratio measurement would be quite useful.

In the end, it probably doesn't matter what system is being used, because the biggest variable in the bicycle drivetrain is the powerplant. What really matters most to riders gear sizes relative to their own powerplant, so all they need to know is how much bigger or smaller a gear is (in whatever unit of measure), and not an absolute gear size.

[Dave]

I've never found much value in gear-inches or gain ratio. I know what gear combos I need and knowing that, I use "the equivalent gears formula".

For example, if I used a 30/25 low gear with a triple and wonder what I might need to match that gear ratio with a compact:

25/30 x 34 = 28.33

That tells me that a 34/28 isn't quite as low and a 34/29 is a little lower gear.

Gear-inches can be very misleading. At the top end, 12-13 gear inches is a 10% change, but at the low end, only 3 gear-inches is the same 10% change. People get the idea that 3 gear inches is nothing, but it depends on which end of the spectrum that you're talking about.

I also use percentages when figuring the jumps between cogs and not gear inches. Something in the 8-11% range is a typical shift between two cogs. Once again, at the top end, it only takes a 1-tooth change to make the percentage range, but as the cogs get larger, that 1-tooth difference becomes small and a 2-tooth jump is needed.

[Jason E]

I commend everyone on how civil this has remained.

How many inches until winter is over?

[classtimesailer]

Gear inches doesn't account for the leverage of crank length. Longer cranks mean more travel at the pedal for the same distance over the ground. Larger cog in the rear means more travel at the pedal for the same distance over the ground. From a force/power/work perspective, it makes sense to account for crank length.

I would imagine that shorter cranks for pedaling through turns would require slightly different gearing to keep the watts/inches traveled equal to longer cranks.

Note: I don't subscribe to the notion that crank length is determined by cyclist anatomy except at the extremes.
Jeff

[Chance]

Another way to look at it is "Why do people jump on anything new and different simply because it is"?

Exactly. The “why” we “jump” on something while we avoid something else is more intriguing to me than the things we actually accept or reject. The human factor makes it of greater interest than the item itself.

[Chance]

I'll tell you that the crankarm length issue does not "give back" enough to learn a new language.
For me readjusting was as simple as viewing it as a scale of 1 to 10.

At the very bottom 1 represents when we get off the bike and walk up the hill; used only as reference. Our feet move about the same as travel distance. You’d never want to gear a road bike that low although some have by using MTB components.

Road drivetrains at the low end will go from 2 to 3 to maybe 4 at most.

A gain ratio of 2 (like 1:1 gear ratio with 700 C tires) is very low. Most riders will want low gears around 3 (like 39/26 or 34/23 range). Racers in flat areas may do OK with a low of 4 (42/21).

At the high end numbers in the 8 to 10 range will cover most riders.

A GR of 10 is extremely high and may be used on a tandem for very fast descends (like a 54/11). Most of us will do OK with a high around 9 (50/11). Riders who don’t like to push on downhills might find 8 is high enough (48/12 or similar).

So most all road bikes will be geared between an extremely low 2 and an extremely high 10. And we’ll spend most of our time riding in the middle of that range just like with any other numbering system. For me personally a scale of 2 to 10 is no more difficult than 27 to 135 inches.