Gear Inches and Different Bikes?

Eternally Dirty
Whenever there is mention of a bicycle's gearing, inevitably we bring up gear inches. And when there is a question of what gearing is best on a new bicycle, someone will suggest to calculate the gear inches on a bike the person is already comfortable with and use that as a template. But here is the thing: While I known how to calculate gear inches, I've been finding this mostly useless when setting up a new bike - because, in my understanding, gear inches are bike-specific. In other words, the same gear inches don't feel the same on different bikes. What am I missing or misunderstanding here? 

To quickly summarise for those new to the concept, gear inches are a convenient way to describe the gearing you are in when in different combinations of the front chainring and rear cog (taking into account wheel and tire size and crank length [edited to add: see discussion in comments regarding this; seems that I did misunderstand.]). The lower the number, the easier the gearing. For example: The lowest gearing possible on my Rivendell is 26 gear inches. The lowest gearing possible on the Royal H. Randonneur is 27.6 gear inches. The lowest gearing possible on the loaner Seven is 33 gear inches. 

So, on paper, it appears that the Rivendell is geared easier than the Randonneur, and much easier than the Seven. But in fact the bicycles feel similarly easy to ride in their lowest gearing. Cycling up the same hills, I've determined that 33 gear inches on the Seven feels about the same as 26 gear inches on the Rivendell and 31.6 gear inches on Randonneur. It seems that weight, geometry, positioning, tubing, and a number of other factors play into it and that gear inch calculations are bike-specific. It is not clear to me why some seem to suggest that gear inches are independent figures that one can use to determine the appropriate gearing on any bike. 

Edited to add: My "what am I missing or misunderstanding here?" question has been answered in the comments; please read through them for an interesting discussion. I was mistaken in believing that crank length is factored into gear inch calculations; it is only factored into gain ratio calculations (an alternative way to measure gearing). When comparing gain ratios between the three bikes, the equivalently-experienced gearing on the Seven and on the Randonneur are in fact the same, whereas the equivalently-experienced gearing on the Rivendell is a bit lower. So... while other elusive factors remain, it appears that crank length is a huge one and that calculating gain ratios instead of gear inches allows you to factor it into the equation. Please continue to contribute to the discussion if you feel there is more to it, or that I phrased something incorrectly. I would like all of this to remain here for others' benefit.


  1. I think weight is the big factor here. To a lesser extent, it's going to depend on bike fit and how well you can transfer power from your legs to the pedals on a specific bike.

  2. Specs just don't tell the whole story. And distilling crank effort down to gear inches doesn't tell it anyway. First time I built a single speed, I decided from the 10 speed I was riding that I wanted a 42x18. So that's what I built. Boy! A 42x18 SS was (and is) sure a lot easier to pedal than the 42x18 on my 10-sp! And it isn't JUST the weight of the derailer, extra chain, and that fatter freewheel. It's just more efficient. It takes 6 out of the 10 gears on that 10 speed to get everywhere that my SS can get me.
    Get really good bearings on your BB and hubs, new aligned chain, on the same frame and your gear inches can climb for the same effort. Lighten your rims and tires and it climbs again. But notice that the first improvement didn't give away any comfort, but the second one did. ;-)

  3. I agree it is a moving target, but it is a starting point! A huge factor is the tires and it seems this is more pronounced as the diameter of the wheel goes up. I had to change tires drastically on my mountain bike for it to effct my gearing much, but on my 700C and to a lesser extent my 650B bikes it is much more noticable going from a 42mm tire to a 35mm or even 28mm!! Much of this has to do with weigth, but also tire pressure. Those skinny tires are not only lighter, they have 30+ more pounds of pressure in'em so they roll much easier! Of course as mentioned they are not nearly so comfortable! ;-)


  4. I think the Seven is like 38 x 27 x 700x23c x 175mm so more like 37 gear inches at the low end?

  5. Sorry, I've changed the numbers a bit; got confused with the calculator. Pattern is still the same (but less drastic).

    Interesting about the tires.

    The way I understand it, gear inch calculations are helpful when:
    a. changing the drivetrain setup on a bike, or
    b. setting up a new bike that is near-identical in every way to the bike you're using as a template

    But I don't understand how people can say that such and such a gear inch value should be suitable for a bike without having experience with the bike.

  6. Anon 12:44 - The small chainring on the Seven is a 34t. It is probably this crankset.

  7. Velouria said...
    " In other words, the same gear inches don't feel the same on different bikes. What am I missing or misunderstanding here? "

    You are not misunderstanding anything at all. :^)

    No two bikes can be EXACTLY alike no matter how hard we try to make them alike. It's called "Material Variations" that is the range that all products flex within during manufacture.

    If you look VERY closely not all bikes (and everything else for that matter) will look alike exactly.

  8. Gear inches are a useful way of normalizing what happens as you alter freewheel cogs, chainwheel sizes, wheel diameter, and tire size. The calculation makes it possible to compare different drive trains on bikes, including lowest gear, highest gear, and size of steps between gears. However, what a particular gear inch "feels like" while riding also depends on both the load you are carrying and the terrain you are riding over. The same 30 gear inches that feels ridiculously easy on a flat road, may be a welcome relief on a steep climb. Now try the same steep climb with 80 pound of camping gear on your bike, and the same 30 inch gear may no longer feel like a relief. On that bike, with that load, you now want gears down in the 20's. So yes, the gear inches you will feel you need do depend on the load you are carrying. The Rivendell and the Randonneur are probably 10 to 15 pounds heavier than the Seven, including a lot of weight difference in the wheels themselves. That's not the same as 80 pounds of camping gear, but the principles are the same: carrying more weight makes you feel like you need lower gear inches to maintain similar efforts.

  9. I disagree that one needs experience with a bike before gear-inch calculations can be handy.

    I've always felt that you don't necessarily have to "know your bike" before you set up your drivetrain based on a gear-inch range. I find it tremendously useful when setting up a new bike. The trick is to understand how that bike will be used. Then the choice of gear-inch range makes more sense. You need to have a starting point when designing a drivetrain for a bike with a particular purpose in mind. For example, if you are designing a loaded tourer that you plan to use to climb steep hills with 50 lbs of extra gear, you'd better design it to have a lower gear-inch range than a lightweight road bike that you intend to use for paceline training rides. You also have to factor in what kind of cadence you want to have with the bike. For a fast road bike, you will want a faster cadence, with gears that are spaced such that you can maintain a constant cadence as the speed of the bike changed. For a bike that will climb steep hills, you don't need as fast a cadence. But how you build these bikes initially if you don't have an understanding of gearing? You have to start somewhere, and it makes sense to start off with something predictive and quantitative.

    A lot of it is predictive-- as long as you know how you intend to use the bike (and assuming the bike is built in accordance with that intended use), coming up with a gear-inch plan before ever riding the bike is actually quite useful.

  10. Just as a followup example: I knew how I would ride my Jeunet and what my riding position would be like when designing its 1x5 drivetrain. I knew that it would be an all-purpose city bike that needed to have just a little lower gearing than a typical city bike so that I could get up the steep hill to my house (a hill which you are familiar with). I knew that the gears would be widely spaced and that a "perfect" cadence could never be achieved with this setup, but I also wasn't trying to create a perfect road bike. I used a gear calculator to come up with the appropriate front ring and 5-speed freewheel. I still consider this setup about perfect for my needs, but the key here is that I knew how I wanted to use the bike, what my body geometry would be, and what my local riding environment would be, before I decided to come up with a gear plan that would work with it. But I had never ridden the bike before coming up with that plan and had nothing to compare it with. I just came up with a plan that made sense based on my understanding of gearing. I knew I wanted a gear inch range of about 40-90, and that's what I designed.

  11. Masmojo— A smaller tire lowers gear inches. I don't think that the 400 grams saved (if that) going from 41mm down to 32 makes the biggest difference. The change of 1-2" is what is likely doing it. For example: 48x21, 700x23c is 60.1". Change to a 38mm tire and this goes up to 62.4". Now that may not seem huge, but it will make a difference. Also, if that skinny tire is more supple that the fat one, lowering rolling resistance, the bicycle will go a little faster at the same effort.


  12. I agree about the importance of having a starting point - after all, we can't choose the gearing at random. But then we also have to be prepared to switch out the cog/cassette/crankset if the gearing does not meet our preconceived notions. Even setting up ss bikes - which should be simple as it requires getting just one gear inch measurement right - I've had to replace the rear cog from what I initially thought it ought to be every time.

    As for predictive, I don't know to what extent. Maybe with production vintage roadbikes, which seem to have been more similar to one another than not. But with modern bikes, where the tubing and geometry is all over the place, it seems trickier. Would you have predicted such a difference between the Randonneur and the Rivendell? They weigh very similarly, have the same wheels and tires, are both to be used for light touring, etc.

  13. Two things. That crank length thing might be coming into play. Also, this is a great example of why it can be a great thing to have a couple of different chainrings sitting around at the same BCD. A small (or large) change in gearing can be made easy.


  14. Crank length does not factor into gear inches. It's a somewhat related issue, but is not part of the equation (number of chainring teeth divided by number of cog teeth multiplied by rear wheel diameter, including the tyre, of course).

    More importantly, this whole conversation is missing the point of gear inches: Multiply gear inches by Pi and that is exactly how far you will travel with one rotation of the cranks. A 30" low gear on one bike is the same as a 30" gear on a different bike. Again, in a 30" gear, the rider will travel 30"xPi for each rotation of a crank, so gear inches do inform us exactly how far one travels in any given gear (and thus, from this, we can infer "how hard" it will be to pedal in that gear, depending on the terrain).

    All this other stuff you're sensing and discussing is about positioning, bike fit, bike and body weight, type of tyre and pressure, etc. Some bikes do, in fact, "climb better" than others while handle rollers or flat roads better. Quantifying that is the hard, if not impossible, part, but gear inches are simple, straightforward, and not open to interpretation.

    - Chris Kostman

  15. I do think that gear-inch calculations can be fairly predictive, but it's also normal to realize that we use our bikes differently than we intended. You have mentioned that you use your RH mixte almost exclusively for transportation cycling, which is not how you envisioned it to be used when designing it. I also didn't intend to do as much trail riding with my Shogun 650B as I've been doing. I initially thought I would do more road riding with it. And while I can't answer your question about the similar feel of your Riv @ 26 gear inches versus the RH randonneur at 31.6 (but have some ideas, such as more aggressive riding position on the RH, based on the photos you've published, and perhaps that mysterious 'planing' concept being more at play with the RH), I can say that the gearing that I designed for my bikes has mostly worked out as predicted. Of course, I do want to make changes and tweaks here and there (for example, I no longer feel like I need the stump-pulling 21 gear inches on the Shogun for really steep climbs, and will probably switch out the cassette for one that gives me a low of 25 gear inches, in return for tighter spacing and better cadence control).

  16. Chris Kostman said...
    "Crank length does not factor into gear inches. "

    Garbiel said...
    "That crank length thing might be coming into play. "

    Awesome : )

    Chris - If it's not part of the equation, why is it included in the calculator?

    Gabriel - it is already accounted for in my comparison.

  17. Crank length does not factor into gear inches. It's a somewhat related issue, but is not part of the equation.

    Correct, that crank length does not change gear inch calculations. However, it does change the feel. Longer crank arms give more mechanical advantage, increasing the amount of torque about the crank for a given pedal force. Therefore on two otherwise identicla bikes, the one with longer cranks may allow the rider to climb the same hills with a slightly higher gear. [Velouria- what crank arm length does the RH have? If longer than the Riv's, then that could be the reason for the difference in you describe].

  18. Longer crank arms = more torque generated = lower cadence needed = higher gear to do the job

    Shorter crank arms = less torque generated = higher cadence needed = lower gear to do the job

  19. "Quantifying [other factors] is the hard, if not impossible, part, but gear inches are simple, straightforward, and not open to interpretation"

    I understand that gear inches themselves are not open to interpretation. But what I am saying is that in a real world scenario, setting up Bike A by using the gear inches on Bike B will often not work, precisely because there are so many other factors. Do you disagree?

  20. What an awesome topic; I would never learn about this stuff anywhere else.

    When you were working with *Bella Ciao,* did you discuss/alter things like gear inches or only components/accessories? . . . what are the gear inches on the lowest/easiest gear on the Superba?

  21. somervillain - The Riv has 165mm crank arms; the Royal H has 170mm crank arms. I assumed that of course this matters, but I also assumed it was getting factored in, as it's one of the values the calculator asks for?

  22. Velouria, the following statement:

    gear inches are a convenient way to describe the gearing you are in when in different combinations of the front chainring and rear cog (taking into account wheel and tire size and crank length).

    is not completely correct. Gear inches are not affected by crank arm length. Try changing the lengths in the calculator, the gear inches do remain the same. The reason you can enter crank arm length is for if you're calculating gain ratios. This gets back to my earlier comment about longer crank arms giving you more torque. It's the ratio of the velocity of your pedal stroke to the velocity of the bike. This does change with crank arm length.

    So, Chris and Gabriel are both correct, and I think that the difference between 165 and 170 crank arm length at least partially accounts for the RH and Riv bikes feeling the same with different gear inches.

    Try this: plug both bikes in but instead of selecting gear inches as the output, select gain ratios. The gain ratios might seem more equivalent.

  23. The place where gear inch comparisons are most useful are on the velodrome w/ track bikes. Generally speaking...

    for "warm-up" pacelines 48/16 => 81 inches
    for races of a few laps => 90+ inches
    for match sprint or keirin (explosive and brief) => 100+ inches

    If you visit a velodrome riders will frequently be swapping out cogs and chainrings depending on how they're feeling and what their event is. It makes a big difference in terms of efficiency and performance.

  24. Gear inches actually gives you the size of the front wheel of a high-wheeler with direct drive.

    So if you have a 27" rear wheel, and a gear ratio of 52-26 (2:1), then your gear is 2 x 27 = 54". It's that simple. If you go to a 52-13 gear (4:1), then you get a 4 x 27" = 108" gear.

    So you can see that the wheel diameter plays a role, but crank length does not.

    As somebody pointed out, all you need to do is multiply by pi, and you get the actual "rollout" - how far the bike travels for each crank revolution. That is how Europeans usually measure gearing, but in meters, not inches.

  25. Julia - Well the gear inches are directly related to the components - namely rear hub and rear cog, since on a bike like the Bella Ciao you can't switch the chainring due to chaincase issues. The rear hub I chose (Sturmey Archer 3 speed coaster) is different from the one that was on the bike originally (Shimano 3 speed coaster), which made the range a bit narrower and (in my view) improved the spacing of the gears. The rear cog was kept the same. I do not remember what the actual gear inches are, but when I get my bike back (it's at the rack-maker's) I will count.

  26. somervillain & others - Interesting, I see what the source of my misunderstanding is now. I think a lot of people are confused about this, because the gear inch calculator appears to factor in crank length.

    "Try this: plug both bikes in but instead of selecting gear inches as the output, select gain ratios. The gain ratios might seem more equivalent."

    Yes, let me try that with all three bikes.

  27. What calculator includes reference to crank length in determining gear inches? I have never seen nor heard of such a thing.

    Of course crank length affects torque / leverage, not to mention what your knee experiences, but there are only three variables in the gear inch equation: front teeth, rear teeth, and wheel diameter.

    I completely agree that different bikes can feel quite different, even with the same gearing. For that matter, when I raced the Race Across America I had three different road bikes, all set up with the same gearing, saddle height, reach to the bars, bar width, shifter placement, pedals, tires and wheels, etc. However, one was a Holdsworth built with Reynolds and sport-touring geometry, another was a Ron Stout built with Columbus and road racing geomettry, and the third was an Alan Carbonio that had a very tight design and whippy tubing. Even though they all "fit" the same, they rode and felt differently, which is why I changed bikes every 18-24 hours on average. It just felt good to switch things up.

    - Chris Kostman

  28. Crank length does not impact "gear inches". However, it does impact how a particular gear combo will feel. Two identical riders riding identical bikes but with different crank lengths will travel the same distance with each revolution of the pedals. The only difference is that the one with SHORTER crank arms has to push harder to maintain the same cadence as the one with longer crank arms (of course, the circumference is proportionally smaller, so they're doing the same amount of "work" for each revolution).

    Usually, folks with long legs ride longer crank arms: 175mm and up. Sprinters and folks with shorter legs often prefer the shorter crank arms: 165mm and less.

  29. As usual, Sheldon's explanation that accompanies his on-line calculator provides a nice clear summary of both gear inches (not affected by crank length), and gain ratio (is affected by crank length):

  30. Wow, okay.

    So, when I select gain ratios and factor in crank length, the ones for the gear combinations that feel equivalent are:

    2.4 for the Seven
    2.4 for the Randonneur
    2.0 for the Rivendell

  31. As others have pointed out, crank length is not factored into gear inches, but it has a potentially large effect on the effort required to turn the cranks. The late, great Sheldon Brown devised a measure to take crank length into account; he called it "gain ratio," and the gearing calculator on his website allows one to calculate it. The difference in effort required to turn 170mm cranks as opposed to 175mm cranks isn't all that big (though it might well be noticeable), but if you go to really short cranks (e.g., 155mm or even 145mm) as many recumbent riders do, it is sometimes necessary to install smaller chainrings in order to compensate. (BTW, within the recumbent community, the debate over whether or not this is really "necessary" occasionally approaches the helmet debate in intensity.) Of course, as others have also pointed out, there are many other factors that also influence the effort required to pedal a bicycle. But these, unlike gear inches and gain ratio, are much harder to quantify in any way that is directly related.

  32. Here's a good calculator...

    Note that crank length has no impact on gear inches. It influences what they call "gain ratio."

    Optimum crank length is "fit" parameter that depends on how long your legs are.

  33. To clarify, I've used gain ratios before as well and still encountered bike-to-bike inconsistencies. But I definitely misunderstood the role of crank length in gear inches, thanks to all for explaining this.

  34. I modified the gearing on my bike and used bike speed at a given cadence to compare the different gearing options. I find the overall gearing range and differences between all the possible gears to be interesting. This shows the calculations and a graph comparing a 52/39-crank 13/26 cassette, with my new 48/34 crank-11/28 cassette. The benefit (if my calculations are good), is I get both higher and lower gearing:

  35. oh, it's interesting. I think it's an irrelevant subject in the bicycle world. :P What counts most is the ride and that is always different, just like all the days in life.

  36. To add some vintage comments, the crank length does vary on some of the old Raleighs. The kids bikes had much shorter cranks (with lower gear inches, to match the lower leverage of the cranks).

    I've read that some of the older DL1s came with slightly longer cranks than the Sports (more leverage to match the bigger wheel), although I'm not sure I've actually noticed this. I have found that my DL1s all seemed to come with small sprockets (16T) and large chainrings (48T), for insanely high gearing. I changed to larger sprockets. The Dutch roadsters had usable gearing.


  37. I like this topic. The discussion is interesting and I think there is more to it. To summarize what is my understanding so far:

    Gear inches: are a measure of how much the bike will advance with one crank rotation and are independent of the crank size.

    Crank size: is a measure of how much leverage is available to move the cranks, and therefore a measure of how easy it will be to rotate the cranks.

    Gain ratios: are calculated combining the gear inches with the crank size, resulting in a measure of how hard it will be to make the bike move.

    I find the gear inches more intuitive, and, since their calculation doesn't involve the crank size, they are the same for every rider regardless of the height. For the gain ratios, we can have the following example of 2 bikes on the same lowest gear of 24 gear inches, with 175mm and 145mm cranks and yelding gain ratios of 1.73 and 2.08 respectively. This means that the 175mm crank bike needs 17% less force on the pedals to move in the same gear.
    I doubt that someone who is 5' tall would have an easiar life riding the 175mm crank bike, and that's why I think the gain ratios will only provide an accurate indication when comparing crank sizes that are close to each other, but in this case the gear inches method would give similar results.

    Other important topics closely related to gearing are power and efficiency, as was hinted by Phill Miller 12:01pm and Masmojo 12:29pm.
    From my limited experience I agree that the rolling resistance of the tires can make a huge difference on how "heavy" the bike feels.

    The power needed to overcome rolling resistance, air drag and the internal mechanical losses is not easily calculated as it happens with the power to overcome gravity but all of those things combined can have a big impact on how heavy the bike feels.
    Also, the riding position on different bikes can result in different power outputs for the rider.
    Given the complexity of those calculations, the only easy and reliable thing would be to determine them experimentally using the following instrument:

    Velouria, with something like this you would be able to find out why your Gazelle climbs better that the Pashley even being heavier.
    I would like to have one even just to see how much power is lost by adding fenders to a bike, but these things are too expensive.
    I hope some reviewers will have the idea of adding efficiency measurements to their reviews when comparing different bikes.

  38. Jazz, your basic understanding of Gear Inches and Gain Ratios are incorrect. I think this is a good indication of how much misinformation there is in this article and the comments.

    In my experience, the only people who legitimately concern themselves with gear inches are track racers. For everyone else, changes in terrain, physical condition, wind and gear carried will effect the usefulness of any gear in any situation. Calculating your gear to the inch is pointless. Trying to transfer it between bikes is nearly as pointless. Relax and ride the bike. Changing a cog is not uncommon, but it also does not require anywhere near this amount of consideration.

  39. Jazz' post is an accurate summary of gear inches and gain ratio. In my experience, knowing them helps in predicting if a bike will suit for one use or another. I know I want a minimum of 35 gear inches for my commute to work. Don't want to get sweaty on the way to the office.

    And I know I want 25 gear inches for the return trip on a looong recreational ride where I have exceeded my comfort zone, like the little, but frequent hills of Truro - yesterday. :-)

    Sure, the factors mentioned by Kyle are important too. But this post and the comments are valuable for those that want some insight into how their bikes work.


  40. My Waterford requires 39 x 26 to do the hills I did yesterday.
    Rickert did them in 42 x 24. Rickert is also 4 pounds heavier and has 170 cranks rather than 175. Same tires. Both built to a race bike model.
    Go figure.
    I'm just glad I don't need to dig up vintage freewheels for the old bike.

  41. Anonymous Angelo at 11:23.

    Medium gear is 48 x 18. The sun rises in the east. Any gear larger than 48 x 18 is a big gear. Any gear smaller than 48 x 18 is a small gear.
    If you have DL-1s with 48 teeth you have some nice old DL-1s. The 16 sprocket you mention is someone's mistake, Raleigh didn't fit that.

  42. Gear Inches are an absolute measure for what they are worth but they definitely don’t tell you the whole story if you use them to compare one bike to another. If comparing different ratios on the same bike, they are much more use.

    Here’s my own proof: I have ridden 2 very different bikes up the same hill – a mass market 21 speed hybrid and an 8 speed hub gear Big Heavy Dutch Bike. On the hybrid I used to commute up this hill, I know it all too well.

    Gear Inch comparisons are tempting because in theory the lowest gear on the BHDB isn't that much higher than the lowest gear on the standard hybrid. So on paper you might conclude that there's little difference. But in practice you really do feel the difference because of the different posture you adopt.

    On the BHDB for most of the climb I stayed on and pedalled whilst seated, but along the steepest half mile it really was easier and more sensible to walk up and push. On the hybrid I never had to do this. Getting off and walking wasn't a problem for me -- I wasn't in a big rush and it probably only took a few minutes longer anyway. I wouldn't have wanted to do it after dark without a sidewalk, though.

    The different handlebar set up on the Dutch bikes doesn't make it easy to ride "standing up" on the steep bits -- the front end feels very light and twitchy, and your arms are too far back. It's simply not what the bikes were designed for. Even leaning forward whilst seated is harder.

    My verdict is that if you simply compare Gear Inches between two radically different bikes, you can have an objective but almost meaningless measure. This is because of the other (sometimes unquantifiable) factors that need consideration but which aren’t included in the measure.

  43. A way of paraphrasing my previous post...

    Gear Inches are very useful if comparing different ratios on the same bike.

    Gear Inches tell you much less if you try to compare the same ratio on different bikes.


Post a Comment