Part One explained most of the frame dimensions.
Part Two explained trail.
Part Three explained how changing components can affect ride qualities.
Knowing the size bike you ride is pretty important when you want to climb aboard another bike. But the information isn’t necessarily informative. And even if it is informative, it only has value if the bike your basing your fit decisions on is one you feel fits you well.
As with anything made in multiple sizes, there are sizing conventions for bicycles. Just like you typically know your shoe size before you set foot in a shoe store to try on shoes, you know a representation of your bike size before trying to fit on another one. Sometimes, that size is a number, like 55, sometimes it’s a letter, like M, sometimes it’s a word, like Medium.
There are two common size-naming conventions in the bike world. The first is seat tube length; the second is relative position in the size run, sometimes referred to as “t-shirt sizing,” because it utilizes XS/S/M/ML/L/XL, etc. Seat tube measurement is either from the center of the bottom bracket shell to the center of the set lug or cluster (aka center-to-center, c-c) or from the center of the bottom bracket shell to the top of the seat tube (aka center-to-top, c-t). The measurement used to be in inches on cheap road bikes and most mountain bikes. The measurement used to be, and largely still is, in centimeters on higher-quality road bikes. This measurement was a pretty good way to compare one bike to another when top tubes were parallel to the ground and stems were also parallel to the ground. In this era, road bikes had a seat tube angle between 72- and 75-degrees. When all builders designed for parallel top tubes, it was fairly reasonable to expect that the corresponding top tube would be within a narrow length range and the top of the headset was in a narrow height range relative to the bottom bracket.
Once compact frames came into vogue, people started discussing what virtual top tube length they needed, as the seat tube length was shorter by an unknown amount and the actual top tube length was impossible to compare from bike to bike. Virtual top tube length, sometimes called effective top tube length, is the length the top tube would be from the head tube to the seat tube if the top tube was parallel to the ground.
But that presented other problems, like head tube length was not necessarily tied to any other frame dimension. And that presents a problem. To locate where your position would be on a virtual bicycle, you need to know where your saddle will be and where your bars will be relative to the bottom bracket.
As the changes to bike design were going on, a triathlete bike designer was looking for a better way to describe frame sizing. In the 1990s, Dan Empfield, then of Quintana Roo, wanted a better way. In 2003, he proposed such a way. He suggested measuring bikes via two dimensions relative to the bottom bracket that would tell you where the top of the head tube was in relation to the bottom bracket. Those two dimensions were stack, the vertical distance from the center of the bottom bracket shell to the center of the top of the head tube and reach, the horizontal distance from the center of the bottom bracket shell to the center of the top of the head tube. This way, seat tube length didn’t matter, virtual top tube length didn’t matter, head tube length didn’t matter. If you know what the stack and reach are on your current bike, know how many spacers are under the stem, and know the length of the stem, you can figure out pretty well whether or not you can replicate your position on nearly any bike you’re looking to ride.
Stack and Reach in Action
Here’s an example. One bike frame that is listed as a 56 has a 58.5cm seat tube length center-to-top. This 56 name comes from 56cm being the seat tube center-to-center measurement a la classic sizing numbers. The stack on this bike is 56.7cm. And the reach is 39.3cm. A rider gets his fit right by installing a 120mm -6 degree stem on the bike and has the stem slammed (aka sitting directly on top of a flat headset cap, with no spacers between the two). To achieve the same position on another bike, it would be easiest to find another bike with the exact same stack and reach, but as that’s unlikely, the thing to look for is another bike with the same stack, or up to 2.4cm more stack (the bike will need a 125mm -17 degree stem to replicate the position) or up to 3cm less stack with 30mm of spacers under the stem (30mm is a safe limit for spacers under the stem) and the same stem. Reach can vary by up to 2cm less, which would necessitate a 140mm stem, about the longest length commonly available, or up to 4cm more, which would result in needing an 80mm stem, which is about as short as commonly available.
Slowtwitch has a calculator that one can use to normalize stem position. It’s found here. Yojimg has a nifty calculator for determining how two stems of varying angle, length, and stack compare. You can start with the specs of one and it will help you determine the theoretical stem that will do the same thing but at another angle, or with spacers, and so on. It’s found here.
Thankfully, stack and reach numbers can be found on most geometry charts these days. Here’s a calculator if you can’t find stack and reach but know other measurements of your frame. And Slowtwitch has a table with lots of time trial/tri bike models as well as a table with lots of road models.
An increasing number of frame designers are now prioritizing stack and reach to determine the frame sizes they make and the geometries thereof. The thinking is that if they can have a steady, or steady-ish, progression of stack and reach numbers from the smallest frame in a size run to the tallest, they can fit more people with fewer frame sizes because each frame size will be distinct. This means that they try to have the stack and reach increase by fairly consistent amounts, for example, have stack increase ~20mm per size and reach by ~10mm per size. Thanks to computers being able to crunch so many calculations at once, it is relatively easy to plug in certain values and have a spreadsheet derive the rest.
The traditional way to come up with frame sizing is to start with the seat tube and then increase numbers from there, as seat tube angle and length were seen as the foundation for the rest of the frame, with both the rear and front triangles dependent on it. In the steel era, it was fairly common to find a pro bike model size run that had a frameset per 1cm increase in seat tube length that ran from 50-60cm, with a few smaller and larger sizes as well. Nowadays, some companies do a size every 1.5-2cm of seat tube length; some vary the size run with 1cm differences in the middle of the size run and 1.5-3cm differences on the tall and small ends.
Even though seat tube length is how frames are still nominally measured, actual seat tube length doesn’t matter all that much these days as seat posts are fairly long and thus offer a wide range of adjustment, and the length of the seat tube doesn’t have to join the top tube at the same spot, and the top tube can slope not at all, a little, or a lot depending how the designer wants the frame to fit, perform, and/or look.
Some companies prefer to use a number, like 55, some utilize XS/S/M/ML/L/XL etc., but the result is the same. The choice for using a size word rather than a number might be to make sizing seem simpler, someone who is of medium height might assume they ride a “medium” frame. But once a rider has figured out how her bike should fit, the name can be confusing, especially for women, who, as a group, are several inches shorter than men. A male of medium height is around 5’9” in the United States, and that person used to typically fit a 54-56cm seat tube length frame, which is pretty close to many “medium” frames. It’s generally larger than what an average height female in the US (5’4”) could fit. But these days, the seat tube lengths are frequently shorter, and thus going by seat tube length could be confusing. There is more potential for confusion, as people with long torsos relative to their height might want to size up to get the reach they want, while people with short torsos relative to their height might do the reverse. The odd-looking ML size, in fairly widespread use, seems to acknowledge there are limitations to the naming scheme, but that’s a separate issue.
Prioritize Stack or Reach
What’s more important, to have the stack exactly where you want or the reach? There is no easy answer, not least because getting either exactly right on another bike is hard. Do you want to have the stem height a certain distance from the upper headset race first, or do you want to have a stem length that is more to your liking?
Empfield believes in prioritizing reach. For him, the choice is fairly easy. “I usually fit people to reach. There’s so much more you can do to change the elevation of the bike. You can add spacers under the stem, and you can change the pitch of the stem, and so height is highly fungible. The only time I take a hard look at the stack of the frame is that there’s no way I can get low enough.” He also admits to having a personal preference for stem length, one that he tries to stick with regardless of the road bike he’s riding. Road bike traditionalists don’t quite see reach as everything. Yes, they often prioritize longer stems, but they also seem to believe people seem to prefer how a smaller bike feels underneath.
And keep in mind that once the handlebar shape is changed on a road bike, the reach is effectively changed, as the handlebar reach has been changed.
When looking at stack and reach of two different sized same-model frames, realize it might not be one or the other. Things like bottom bracket drop and trail can change for a bike model from one size to the next. The preference might be for the bottom bracket drop or trail number that is more to one’s liking, so check all the geometry numbers, not just stack and reach, when trying to determine frame size. The one thing to keep in mind is that while the weight of the frame might vary a few grams from one size to the next, the variances that come with frame construction and paint account for more of the difference, and, unless the seat post is getting cut down to minimum length, the post weight is the same regardless of how much is inside or outside the frame.
Getting even closer than stack and reach
Stack and reach dimensions can do a pretty good job of getting you very close to being able to step off of one bike and onto another with an identical position. But handlebars dimensions, and for triathletes, pad position, are even more precise. Pad position for tri/time trial bikes is a good bit more complicated, especially if trying to fit to proprietary handlebar systems.
Build in some room for adjustment
If you’re setting up a new bike, it’s probably not wise to push the limits of gear with that new ride. By this, going with the longest- or shortest-possible stem, or with a -17-degree stem slammed on the upper headset race, or with 35mm of spacers under the stem. If the bike is a little off, unlikely with carbon bikes out of molds, or if you change handlebars, or your position changes due to another fit, or injury, or age, you could suddenly go from fitting the new bike to not being able to fit it at all. It’s better to provide a little room for adjustment.
For the rider, there isn’t one better way for designers to size bikes. There are enough variations in body geometry and flexibility, that there isn’t one sizing scheme that will fit everyone perfectly and riders shouldn’t expect that every bike model will fit the way the rider wants it to. What matters for the rider is having a system of measurements that make it easy to determine sizing without having to build up or try multiple bikes before determining if the model exists within a rider’s fit parameters.
Stack and reach are valuable measurements for comparing one bike to the next when looking at fit. But they’re only valuable if you’re happy with the starting point. And getting the fit right is only half of the equation, though arguably the more valuable half. The bike still has to do it’s thing, but once you’ve found your position, it’s easier to experience what that thing is.
I want to thank Edwin Bull of Van Dessel, Brad DeVaney of Litespeed, Dan Empfield of Slowtwitch, Steve Fairchild of Fuji, Tom Kellogg of Spectrum Cycles, Damon Rinard of Cannondale, and Scott Warren for their time and insights. And Bikecad.ca for the drawings.