Theoretical Framebuilding Part 1- Geometry

Technology alone is a poor substitute for experience.

-Richard Sachs

In my days working in science I had a supervisor that always said – do the experiment on paper FIRST and that means putting down the 5 minutes it takes you to walk to the spectrometer with the samples as well as anything else. He was right. If it didn’t work ‘on paper,’ it was NEVER going to work as you intended in practice aka it would fail miserably. This is the first of three part series on why and what makes a custom fit bicycle frame, different (and better) than ‘off the peg’ one. The posts are intended to shed light behing ‘hand picked, custom butted, special shaped tubing mix’ and all the terms you see and hear mentioned about custom bikes. This article is meant to educate and represents a successful experiment ‘on paper’ by an aspiring framebuilder (me). The science, metallurgy, physics and engineering are all researched, the opionions (as harsh as they may be) are all personal. The articles are ranked by order of importance. Part 1 discusses probably the most important aspect – geometry. Part 2 talks about tube forming and butting and it’s effect on “ride feel.” Part 3 delves in metallurgy and what heat does to metals and explores terms like, heat treatment, alloying, material stiffness and strength.

You might wonder what a thermonuclear weapon or an H-bomb in the header image have to do with custom bicycle framebuilding. Well to this day there is no OFFICIAL and CONFIRMED information on how the weapons work, it is all speculation, though in November 1979 the author of the above article in The Progressive was able to deduce pretty much how it all functioned by just looking at the type of parts etc. ordered by army contractors, in a way he reverse engineered it all. Granted welding or brazing metal tubes in 3 triangles is in a VERY different alley than thermonuclear weapons, Custom Bicycle Framebuilding just like cycling has not been spared by the many myths and legends and the extremely contradicitng information that is prevalent in pretty much everything in our daily lives. This is in a way refreshing since it pretty much points you towards the right direction – see what everybody else is doing and just do the opposite an you will be fine.

The curious individual that I am, I do like riding my bike quite a bit and over time I discovered that what is available out there is highly dysfuncitonal , and that is putting it midly and abstaining from nouns describing pure excrement. Well especially what was offered for somebody (like myself) who is above average proportions and height I was not able to find out why, information was just repeating some old dogmas “that everybody (and/or the PROs) have been doing up to now, so it can’t be wrong.” Needless to say this is what I refer to as a micro$oft/window$ type answer – correct but useless.

Therefore in my search for the most efficient position to make a bike feel like an extension of me, I came across Steve Hogg’s bike fitting website. If you have not seen it, go there after you finish this article, you will not be disappointed. If there ever was a methodical and clearly explained way on how us asymmetrical humans can function efficiently on a symmetrical apparatus that is a bicycle, this is the one stop place.

Through another article getting into the details on what the geometry does to bike handling. As I mentioned earlier I had a bike that I could fit on, though it rode interesting… Brand is not important, though suffice to say, it is one that has been ridden to great victory quite some times by the PROs. It was twitchy and extremely scary to ride at speed, leaned a LOT in corners, regardless of speed, oversteered like crazy, and on anything slightly technical would leave me trailing everybody else. Therefore as much it is not about the bike, the cogs in my head with the information from the above article pointed me that I might be able to really get something out of a custom bike. Here is the place to say that while 10% of people NEED a custom frame due to body proportions/injuries, everybody can benefit from one since off-the-peg items in generally are designed to fit no individual in particular and as many people in general.

Off-the-peg items in generally are designed to fit no individual in particular and as many people in general.

‘On paper’ I came up with a geometry that was supposed to cure all of the above problems. After some going back and forth with Steve Hogg himself, there were some touches here and there and I placed my order. The frame that I receved rode like it was supposed to and oh SO MUCH MORE! I will post a detailed full review on it soon.

Therefore with the next 3 series of articles it aim to dissect what makes a custom built frame special, different and perfectly suited for one person – you, no matter how (different from) average you are.

Bike Geometry Explained

I do send a warning the what I am about to describe are PRINCIPLES what might work for you and/or your clients will most likely be different from my examples here and second, all of the variables work together in combination and cannot be isolated.

Bicycle Frame Geometry. Source: Wikimedia Commons
Bicycle Frame Geometry. Source: Wikimedia Commons

How your body weight related to the wheels of the bike both horizontally (front to rear) and vertically (up and down) is the DEFINING variable that determines how a bike would behave according to its intended purpose (road, mountain, track, etc). To give you an example a 140cm (5’2″) woman, your average Joe rider 175cm/75kg (5’11″/165lbs), a tall lanky person like me 200cm/82kg (6’7″/185lb) and a tall musuclar 100+ kg (220+lbs) frame snapping individual (they exist, I have seen them). Each of those examples has UNIQUE weight distribution when seated on a bicycle, with the short and tall people being the two extremes of the spectrum.

That weight distribution is a direct result of body proportions AND functionality.

Body Proportions, Weight and Stability

According to bike fitter, Colby Pearce, bikes exist on a spectrum on how the rider’s position relates to the machine. On one extreme end you have time trial bikes and on the other end you have downhill mountain bikes. Priority is aerodynamics and power production with almost no rider weight changes on one side versus super low seat heights giving the the ability to drastically shift bodyweight to maneuver (at speed) over extremely varied terrain. In between you find the road bicycle and I don’t necessarily mean a racing one. Since the principles outlined below apply across the board to all types of bikes, for simplicity I will focus on the road bike since a lot of riding such as cx/gravel, touring, etc. originates from there.

The more weight you have towards the rear of the bike, it will result is a bike that that wants to ‘wheelie’ when climbing and in off road situations (MTB) one that you cannot steer on loose surface. Conversely too much weight at the front can cause real wheel slip and unresponsive/sluggish steering. You want balance that is achieved either by shifting back and forth, up and down as is the case of downhill MTBs or by having your center of gravity at the right spot. Road bikes achieve that balance by placing 45/55% of rider weight front/rear. Off road/MTB requires dynamic weight shifting to match the terrain, i.e. more weight/grip at the front tyre, etc.

Center of Gravity

This is not a new idea. The seat tube angle is probably the most important measurement on a bike since this puts the rider at a balance point from where other positions originate – out of the saddle climbing and sprinting for example. The idea is that since the bottom bracket and pedals don’t change position, out of saddle efforts would always place the rider at similar position relative to the BB, however if you are sitting too far back too far forward, out of saddle positions require a massive shift of your center of gravity. Anything from 72-74 degrees is reasonable , going into extremely like 69 (super slack) and 77 (super steep) creatives massive compromises i.e. hitting the bars with your knees when out of the saddle and/or having ridiculously long reach to the handlebars.

Here we come to the balance again.

All movement is a balance between two muscle systems – phasic and postural. In short, there must be a stable platform for movement to occur. Postural muscle provide that platform for the phasic muscles. Since all movement starts with commands from the brain, it is no wonder evolutionary the postural muscles ALWAYS get priority. What does that imply for pedaling a bike?

In short (full article by Steve Hogg) in order to pedal a bike efficiently you should be in a stable enough position to be able to cantilever your torso forward WITHOUT the use of upper body muscles for support; your arms should NOT be holding you upright. Why?

Remember postural muscles get priority so energy is NOT going into pedaling, but trying to keep you upright. In addition a good number of the upper body postural muscles are involved in breathing – you don’t want to be limiting that. Under load the position should be self-supporting. Steve Hogg refers to this as the balance test. Warm up and pedal at about 80-90rpm at 80-90% effort (hard but not maximum). At the right saddle setback you should be able to swing your arms backwards. Also bike use/power produces plays a role into it (racing vs touring for example)

Balance Test
Balance Test

Ideally a custom frame should be build that your saddle of choice at the right setback sits in more or less the middle of the rails on your seat post of choice. Hence we get back again to center of gravity and seat tube angle.

The balance test is where it gets tricky. A lot of ideas on position are based on this body part over that frame bit and/or the flavor/pro of the week (slammed stem, my handlebar drop is bigger than yours, etc) coupled with the fact that most people spend 40+h of the week sitting at a desk, creating massive amounts of muscle imbalances and overall dysfunctions (i.e. the runner that has a gait with the butt backwards and upper body leaned forward is one example).

While we all “love” the interview question: where do you see yourself in 5 years, it gives the first clue to how the frame should be built. If you are flexible and can functionally!!! ride an aggressive position, it’s an easy answer, if however, you have some limitations and lack range of motion/can’t bend well at the hips, do you see yourself improving down the road? The answer to the last question can make today’s made to measure frame, tomorrow’s poor fitting one. My advise to all aspiring frame builders and also towards cyclists getting a made to measure frame is – Don’t compromise based on the latest fads. There is no substitute for experience. My father told me on multiple occasions that all compromises are like a boomerang that comes back and hits you in the head when you don’t expect it or eloquently put:

Folks don’t come to you to get your version of what is sold at the mall. That wasn’t even the case when steel was ubiquitous. Clients call and arrive at your doorstep because you have abilities and experiences that enable you to translate their information and construct a frame of higher quality and fine design, material is NOT part of the equation.

-Richard Sachs

The takeaway point here is that by going with body dimensions and stuff like knee over pedal spindle (KOPS) is an extremely limited approach that results in (sometimes quite significant) compromises which became super obvious for small and large riders, if you fall in the middle of the bell curve, you usually ‘learn to live with them,’ since in a way you don’t know any better. The methods mentioned above require time and some understanding of functional anatomy. Experience cannot be taught so it takes some getting used to it and to be able to translate somebody fit dimensions into a frame geometry. There are ‘standard sizes’ out there, however, you need understand what changing each of the numbers would do to the bigger picture, most importantly why pick one number over the other.

Some random examples:

  • If you need 140mm stem to be in a functional position the frame is too short for you
  • If you need an upright stem and lots of spacers, the frame is too low for you.

Obviously bike components have quite some adjustment built in them, however below are some aspects that can be directly influenced by the builder and comprise geometry.

Three Riders on a Bike

Below you see how a bicycle frame grows. The images are to scale representing a 48, 55 and 60 cm frames (S-black, M-red, XL-green). It’s not a perfect representation, though it serves a purpose.

While it may not be obvious to you, when frames grow, it is mostly the front triangle and wheels do not change in size.

How Bicycle Sizes 'Grow'
How Bicycle Sizes ‘Grow’

Two points stick out right away. Large frames put more weight towards the rear axle and small frames, if built with standard 700c wheels cause massive toe overlap and/or too long reach relative for the size. I am repeating myself that compromises become blatantly obvious at the two ends of the scale while the middle range ‘learns to deal with them.’

What do I mean by compromises?

How Bicycles Steer and Stay Upright

Almost everyone can ride a bicycle, yet apparently no one knows how they do it. I believe that the apparent simplicity and ease of the trick conceals much unrecognized subtlety, and I have spent some time and effort trying to discover the reasons for the bicycle’s stability.

-David E. H. Jones

It is not just the wheels acting as gyroscopes. David HE Jones went on to construct bicycles that were unrideable and he found it more difficult that he expected. In the end, coupled with the super computer of the time he made calculations and came up with the Jones stability index. Basically anything between 1 and 3 was rideable with 2 being the ideal. I am not posting the formula (you can read it for yourself) for a reason. This is only one part of what influences stability (head tube angle and fork rake) and cannot be viewed in isolation. Stable bike = bike that holds a straight line and takes more input in order to deflect it (corner) as compared to a less stable one. And no, unstable bikes are not the same as agile since agility involves coordinated-like movement, rather than twitchiness.

Therefore below are the elements of frame geometry that can be directly influenced by the frame builder that influence stability.

Obviously things need to be balanced and tailored for the purpose of the bike – extremely low bottom bracket on a bike to be used on a banked velodrome would cause pedal strikes, etc, etc. As usual keep in mind the big picture/purpose of the bike frame.

Chainstay Length

Longer chainstays = greater stability

Sometimes referred to as rear center, the longer chainstays shift more of the rider/bike weight to the FRONT axle relative to the wheelbase of the bike. Too much weight over one axle is not a good thing since it leads to an almost binary grip or no grip with no slip in between (and in my personal experience why I used to corner like a scared puppy on ice…). Riders who can get into a more aggressive position on the bike would need shorter chain stays then more upright ones for example. The length has also technical implications on the angle of the chain. Shimano’s 135mm spaced 10 speed (and above) group sets (mountain and cyclocross) have a recommended minimum length of 425mm, and I believe road ones (130 mm rear spacing) have minimum of 415 mm, yet most bikes fall extremely short of the mark. Why?

This is the place for a historical nugget of information or what I call “we have always done things this way, so it can’t be wrong” In and old edition of cycling textbook (Cycle Sport by Peter Konopka) I remember a picture of a “racing bike’ that had a curved seattube and the caption that was done to make for a stiff rear triangle (shorter tube is stiffer), and that was in the era of 19mm tyres….. Back in the day steel and bicycle tubing was a far away cry from even the low end materials of the last 2-3 decades and long tubes can be made extremely stiff without a weight penalty. Yet the old wisdom remains that the shorter the better. I can personally attest that a longer rear triangle makes a bike that rides (corners) like on rails rather than being plain scary and close to unrideable in anything other than straight line at speed. Granted someone of a smaller stature might not need 430mm like I do, 410-415mm on the SMALL sizes (48-52) and 420-425mm on the MEDIUM (54-58) ones is a good starting point (based on 28in 700c wheels). How is being able to corner FASTER with more confidence not racy?!???!?

Bottom Bracket Drop

Custom Bicycle Framebuilding Bottom Bracket Drop
75 mm Bottom Bracket Drop

Lower bottom bracket = greater stability

Back in the day cyclists used toe clips and even the first iterations of clipless pedals were quite bulky. As such pedal strike could be a real issue, especially in criteriums when pedaling in corners (whether that is the best way to do stuff is another question….). Simple physics points that the lower the center of gravity of bike+rider the more stable. Therefore the lowest bottom bracket drop you can get away WITHOUT hitting the pedals in a corner, is the ideal one for YOUR bike. In my opinion 70-75mm for road riding is ok. Smaller sizes that use shorter cranks can go lower. Bikes designed to be ridden with bigger tyres (touring, cyclocross/gravel) should have even lower bottom bracket drop – 80mm – since the tyres raise the overall center of gravty. Last, this is my personal experience, take it for what it’s worth – a more stable bike leans LESS at the same speed as compared a more unstable one, so the chance of pedal strike decreases even further. Speedplay users also have a distinct advantage here. Also bike companies have liabilities so avoiding tricky situations like wiping out 20 guys in a high speed corner is a good reason to avoid ‘bad reputation.’

Head Tube Angle, Fork Rake and Trail

Custom Bicycle Framebuilding Geometry Trail
72.5 degrees HT angle; 46mm rake; 58mm trail

The relationship goes as follows.

  • More fork rake = less trail with the SAME headtube angle.
  • Less (more slack) head tube angle = more trail with the SAME fork rake.

HIGH vs LOW Trail Debate and Wheel Flop

This is what David Jones used in his calculations. All of those three are interrelated. Headtube angle combined with the fork offset/rake create Mechanical trail.

The theory goes that 56mm of trail is ‘neutral,’ anything higher (60ish) makes the bike more stable and easier to ride in a straight line making for a touring bike and anything lower (52ish) is for ‘racing.’

And here is a recent update, that I experienced myself first hand. The great French monkey wrench (see what I did there) of the low trail cyclotouring randonneur bikes. Here trail figures range from 15mm!!! to 40mm at the HIGH end. Well if you are to believe the internet and even some framebuilders, these bikes will be unrideable. So how come all those cyclotouring French bikes that have done many thousands of kilometers in events such as the 1200km Paris-Brest-Paris?

Introducing Wheel Flop

As you increase the Mechanical trail, you also increase Wheel Flop. To spare you the math here, it basically means that as you turn the handlebars to one side, the front axle does not turn parallel to the ground but goes higher. Obviously you have all your weight on the bike so gravity, just makes handlebars and the bike as a consequence flop/drop to that side, FAST.

Why is that a problem?

I am glad you asked!

Well the saying goes that to stay upright on a bicycle you need to keep moving. While you probably don’t realise it though you are constantly ‘catching’ the bike from falling by weaving from left to right. With a lot of trail (and hence wheel flop) small movements get amplified into big deviations left and right. This gets worse as you get tired and/or you put a load (handlebar bag/front pannier/ aerobars – my emphasis) on the bike. It is a vicious circle – you get more tired your movements become less refined (read more pronouced) so you weave more, so you overcorrect, etc. Also just looking over your shoulder is enough of a weight shift to make you go 1m (3ft) into that direction – this is outright dangerous when riding/commuting in traffic or in a group. Obviously you learn to ‘anticipate’ this, though it all takes energy. As the miles and fatigue sets in, you correct less and less or even not all. Also crosswinds make a high flop bike quite a bear to keep in a straight line and this is coming from me who is (over)built like a Classics rider (2m/85kg).

So that brings me to:

Misconception#2: Low trail bikes NEED a front load or they are unirdeable/explode/etc.

Here is the catch:

To get the benefits low trail (low wheel flop) bikes, you need tires in the 28-42+mm size WITH supple casings that make the tyre conform/spread nicely to the road surface. This creates a decent sized contact patch, referred to as Pneumatic trail from the folks over at Bicycle Quarterly. This serves to stabilize the front end for a lack of better word. Using supple tyres on a high trail/flop bike you can actually feel the mechanical trail and penumatic trail fighting each other as the bike straightens out of a corner – it’s a weird feeling.

Skinny tires 25mm or less with stiff (puncture resistant) sidewalls make even a low trail bike behave twitchy due to very small contact pach. For everyday riding even racing there is no reason to use anyting smaller than 25mm (Google Bicycle Quarterly and tyre tests), though traditions and old wives tales die hard in cycling.

So the etenral question:

Do you NEED a front bag on a low trail bike?

No. This is looked wrong – ass backwards. Yes low trail bikes usually had a front bag/paniers/etc. However, I can argue that this is a result of their high trail cousins being close to unrideable when (front) loaded, so low trial was a fix for the floppy problem. Front load has many advantages (it is easily accessible for one) and does not require a massively overbuilt frame and overall ridign characteristics suffer a lot less.

Is High trail/flop ALL Bad?

Well at high speeds having the front end ‘firm’ up is for sure nice and I sometimes forget that most people out there cannot just buy some tubes and weld a bike together like I can, so you are at the mercy of what is avaialble. Most big brands keep only one type of fork rake (usually 45mm) and that leads to compromises in trail as the frames change size. In the ferrous times of old (ie pre carbon), a custom frame was coupled together with a custom steel fork, and it wasn’t solely for aesthetics.

Having ridden bikes with ridiculously high trail at quite some high speeds and now having been using 35-40mm trail bikes daily and/or many hundres of kilometers, I see aboslutely no reason why the 56mm neutral handling should be considered standard. I am even going to go as far as claiming that this is one *siginifcant* (though not only) reason for the horrific crashes you see at big professional races such as the Tour de France. Overstiff bikes with very twitchy handling for sure add to it.

The current explosion in popularity of long multiday(gravel) racing and the use aerobars (more weight at the fornt axle) in my opinon would lead us to rediscover what French contructeurs had found out long ago – low trail.

Front Center/Toe Overlap

With the proper weight distribution between the axles as per the bike’s intended use, the wheels end up where they end up. While not directly influencing stability per se, too much front center causes a light front end which is too twitchy to control (at speed) with the tendency to wash-out under the rider on loose terrain. Toe overlap can be a byproduct of front center and  it might be annoying in stop and go traffic and/or in techical hairpin turns in cyclocross or on bikes with fenders.

Do you even front center, bro?

Toe overlap can be used, and unfortunately it is also abused by manufacturers. Big sizes (for riders with large shoes) suffer the most since increasing the headtube angle (less slack/higher number=LESS front center), in order to keep weight distribution somewhat balanced between front and rear axles (more weight towards the front of the bike; same principle as chain stay length, and not get a version of the green bike above). This seems all fine since at first glance, yes you can live with some toe overlap, though in order to keep inventory low, fork rakes come in limited sizes (or ONE size in some cases for ALL frame sizes) As you saw above more twitchy if you are already more unstable due to higher rider+bike weight (large sizes = taller and heavier riders), is NOT a good thing (coupled with too much weight on the rear axle).

Small sizes suffer from the opposite problem (slack head tube angle/lower number) in order to make toe overlap reasonable (i.e. not half your shoe), resulting in MASSIVE trail figures (62+mm), combined with the smaller overall rider weight, causing the front of the bike to be a bear to steer (or a version the green bike above). Hence why 26in/650c wheels are a viable option for small riders. As far as toe overlap, having ridden bikes both with or without, I lean towards one without. Hey, it’s custom so why not?

Stem Length and Height

Longer and/or lower stem = greater stability

This can be changed and is not directly influenced by the frame builder, however, it is a part of the whole bike stability thing (and extremely long/short stems point at frame that is the wrong size….).

Just like chain stays influence how much weight is on the rear axle you can load the front axle more by either having the stem/handlebars more forward or lower. This is important when climbing in the saddle so that the bike stays planted rather than trying to ‘wheelie.’ Also a light front end is not confidence inspiring at (high) speed cornering and causes upper body fatigue from having to constantly ‘point’ the bike forward as it veers due to road irregularities.


Bike geometry is probably the most important aspect when it comes to a “made-to-measure” bicycle frame. There is much more than just tube lengths. The frame geometry is a result of the rider straddlign and pedaling it and both should work in harmony to achieve good handling. It starts with body dimensions/proportions and continues with the rider’s functionality and finishes with the intended use of the bicycle. All you have to do in the end is connect the dots.

Theoretical Framebuilding Series.

I welcome comments, however, before asking a question please visit t

Overkill Bike: Source


8 responses to “Theoretical Framebuilding Part 1- Geometry”

  1. You’ve got your rake-angle-trail relations completely upside-down. More rake means less trail. Less angle means more trail. Big bikes typically have more angle which means less trail and more toe overlap (bit is necessary in order to get reasonable weight distribution while keeping a reasonable wheelbase for the handling that most racers prefer).

    1. Nikola Banishki Avatar
      Nikola Banishki

      Hi Pero,

      Thanks for pointing that out, I have corrected it (I was actually really careful not to reverse it…). Though, like anything I always like to consider and advise others to look at the big picture and indeed large sizes have LESS trail (for reasons you point out as well), which for riders that are taller and heavier than average is not a good thing. Also (and I am open to reasonable and justified answers) through personal experience, I am not convinced that short wheelbase and/or twitchy handling = racing. Longer wheelbase, besides high speed stability, makes harder braking possible (more difficult to raise the rear wheel) and proper trail figure (i.e. not less than 56 mm for tall riders to throw a number out there) makes a bike that holds a line in a confidence inspiring way. The latter would need less fork offset/rake causing even more toe overlap and that coupled with the fact that stock frames don’t come with anything longer than 600mm top tubes, means that the ‘status quo’ cannot be reversed without going custom. Also i suspect, since ‘conventional wisdom’ has labeled racing as a teeth-gritting/white knuckle experience that you ‘have to suffer through, who needs comfort?’ as such terms like slack and relaxed when it comes to frame angles, are not something you want in a race bike apparently, even if they render a perfect riding bike for a particular rider or most riders, PROs included, in my opinion. I am quoting Steve Hogg that Comfort + Efficiency = Performance, therefore as I mentioned, it is the whole geometry that anybody should be concerned with, as such if to get proper fit, handling and weight distribution, whatever the angles come to be, that is the perfect geometry for you. Hence why back in the day frame and fork were built as a set together, which sadly is rarely the case today.


  2. The optimal wheelbase is a matter of preference though it does have some general trends; this is exactly why I wrote “handling that most racers prefer”. What feels twitchy to you can feel agile for someone else. Luckily, most people can test this within what is available off the shelf; being of average height, I don’t have to design a custom bike to see what longer chainstays, more trail or a longer wheelbase will feel like.

    There is definitely an element of doing things because that’s how it was always done as well as reasoning that if X is good, more X is more better, and it is good to question the norms and try out things outside them. However, it is important not to get seduced by being the outsider and start believing that there is no reason and thought behind the convention. Especially an extreme outlier like yourself needs to recognise the difference between bad design and a design that simply doesn’t match your personal parameters.

    Lots of frame builders have experimented with various geometries, and a lot of them still do. It is not by accident that most end up within a similar set of parameters.

    1. Nikola Banishki Avatar
      Nikola Banishki

      Hi Pero,

      Excellent points, however, keep in mind that my experiences just illustrate the principles I describe, they are just that – examples, the principles are universal (it’s quantity not quality) – average proportions or outlier. Also I stand by my statement that a mass produced frame has compromises (a lot dictated by inventory and ease of production and possibly liability) they are just more blatant at the extreme ends of the scale. Obviously for someone of average build, there are in theory more widely available opportunities to test for example chainstay lengths, etc. though people rarely go out to really test different geometries, in a way they don’t know any better (I was in that camp, before doing quite a bit of research), even so most stock options are copies (at least geometry wise) of each other with different paint and brand, etc. and bunch of fancy so called innovations.
      Mind you I fit with no problem on a stock frame (contact points), however, the ride qualities are light years away from how a bike should ride; you won’t see 175cm (5’10”) riders on a size 61, so as much as I don’t like to point fingers, it is indeed a bad design, because it will be ‘outliers’ like me that will be riding it. Same as you won’t see 2m (6’7″) cyclist on a size 56 and why as you mention what I call twitchy might be agile for others. Me being an outlier allows for fewer compromises. Also proprietary forks, a plethora of headset standards makes testing this or that trail almost impossible. Short chainstays are ubiquitous for outdated reasons, not applicable to modern materials today and too short for ALL sizes. Richard Sachs puts it eloquently that a custom frame is not this is or that builders version of what is available at the mall ie it is not just paint, logo and fancy material. Pegoretti, Richard Sachs and Baum are the perfect examples of a custom process should be executed, also from what I have heard through the grapevine, Rock Lobster is up there as well and bike fitter Steve Hogg is also a huge proponent of longer chainstays, same goes for Jobst Brandt, though he was a tall and lanky fellow as well. Wheelbase is in my opinion the result of everything else – if the weight distribution and the axles are the correct places, this is the wheelbase whatever it might be.

      We are not discussing nuclear physics, therefore many years of custom framebuilding has, as you point out, put some pretty good reference points on what works. The thing with ‘what racers’ prefer is a model that the industry keeps pushing on everybody. (PRO) Racers ride what they are given and are not allowed to say a single bad word about it, a lot of them also have no idea about functional fitting, knee-over-pedal-spindle…really????? (it ain’t their job, though not knowing is not a good excuse); the majority of cyclists buy it because it is what the PROs ride. It’s how things work, it is a professional sport that is highly commercial after all, so no complaints there. There are rumors of many PROs who rode rebadged Pegorettis (Tom Boonen for example) and Cadel Evans says his BAUM is the closest he has been to flying. Rumors they might be, though you never hear that somebody rebadged a bike that was designed like the rest of the pack. Back in the day all PROs rode made to measure frames that got stripped and repainted with sponsor colors. The model has changed, there are positives and negatives, though avoiding proper design should not be one of the latter.


  3. We do look at bike geometry from a similar perspective, and I too feel that too much focus is being placed on fitting the contact points, but not how the rest of the bike fits under the rider. We really do agree on the basics, but there are some areas where to me it seems like you let your skewed perspective (unavoidable as an extreme outlier) and personal preference lead you too far in one direction when trying to formulate a general rule. Case in point, chainstay length and your definite claim that it is too short for all the sizes on conventional mass-produced bikes.

    You mention Pegoretti, yet his stock geometry has chainstay length range of 40.5-41 cm, perfectly in line with big name bikes. That said, I’ve seen a Pegoretti custom size 55 bike with 42.5 stays. Does this Pegoretti dude even know what he is doing? Or could it be that there’s more to the whole story? (Richard Sachs, by the way, makes his stays 42+ cm primarily for better chain alignment and less friction and wear – his words.)

    When I said that one person’s twitchy is another person’s agile, I didn’t talk about a 180 cm person vs a 200 cm person, both on a size 56 bike. What I’m saying is that a bike that you personally find twitchy might be considered agile by someone of similar height and proportions (as rare as that might be). Different people have different preferences, different bikes have different “styles” of handling, and different road conditions put different demands on both the rider and the bike.

    Take Gianni Motta and Eddy Merckx for example. They were contemporaries, relatively close in height, and they both rode bikes designed by Ernesto Colnago. A bike designed for Motta had a 62.8 cm front-centre and 43 cm chainstays. A bike designed for Merckx bike had a 595 cm front-centre and 41.5 chainstays. The differences are much bigger than just the different frame sizes would warrant – yet do you think either of them had a bike that was wrong for them? And we’re talking 1960s steel here, exactly the material that would supposedly benefit from shorter stays for stiffness purposes if that simplified narrative was all there was to it. So when Mr Colnago designed bikes with shorter chainstays than these two, was it because he suddenly forgot how to design a bike? Colnago bikes today have chainstay lengths ranging from 39.7 to 41.4 cm across the size range, does that imply no thought being put into the design?

    Here’s an anecdote from personal experience. I had a chance a few months ago to test ride several bikes back to back, including a Pinarello and a Merckx. Both of these were what I would consider the proper size for me. The Merckx felt steady and sure-footed. Keep the handlebar straight and it goes straight, turn the handlebar and it turns, always very predictable. The Pinarello had a very different character. If you try to steer it by turning the handlebar it’s easy to over steer, then overcompensate and get into trouble. But if you relax the hands and let them just keep the handlebar steady while you steer the bike by leaning it from the hips (like riding no-handed), that’s when you can stop thinking about the bike and just go where you want to go.

    On my test ride, the Pinarello felt more lively and agile than the Merckx when the road surface was good, but more twitchy and chattery when the road gets bad. They are both well-designed, purpose-built bikes. The Merckx is a true Belgian bike, made for roads of variable quality, with strong winds and rain that’s never far away. The Pinarello is a true Italian bike, made for smooth paved roads, long climbs and twisty descents with frequent sequential corners and curves. Which had the longer chainstays? The Merckx. Which one was better? For me personally, the answer is “It depends.” For anyone else, it’s up to them to decide.

    Sadly, most big manufacturer’s don’t put as much thought into the geometry of their bikes, especially at the ends of the size ranges. There are manufacturers of bikes that are extremely advanced and well-designed in some areas, but which show a lack of understanding of some very basic concepts when it comes to bicycle geometry. But, again, this very much doesn’t mean that nobody in the bike industry knows what they are doing other than a few choice custom builders, not that the basic design elements that most modern bikes share are misguided and wrong. You prefer relatively longer chainstays. That really does not mean that anything shorter than what you like is wrong.

    1. Nikola Banishki Avatar
      Nikola Banishki

      Hi Pero,

      Apologies for the late response, I was away from technology last weekend (as hard that it is nowadays).

      Indeed the modern so-called ‘compact’ geometries have in mind fitting as many people (ie contact points), while largely ignoring important aspects like weight distribution and high speed stability (when cornering). I mentioned Pegoretti since his bikes ‘grow’ together with the riders, while most stock stuff out there is more or less the same thing minus TT and ST length. As far as i know, Pegoretti is mostly a custom builder and the stock geometries are just a starting point, however, I am guessing. Though longer chainstays do make a bike that holds a line better in a corner (you can drive/carve the rear through the BB, the road feel is crazy!), they increase stability. Darren Baum does not hide his affinity of that type of geometry and even the stock sizes show that on his website and reviews of his bikes point towards my personal experience, of course confirmation bias can be involved. The Gaulzetti Corsa as well has similar design ideas.

      Unfortunately short of going custom, not a lot of bikes exist with longer chainstays so that you can ‘try before you buy’ and see whether that type of rear end stability is your thing or not. One can try CX bikes or the so called endurance bikes, though choices are limited. A little known fact is that Shimano MTB/CX (135mm spacing) drivetrains have a minimum recommended length 420mm chainstays and road groupsets (130mm spacing) have 415mm, therefore with 9+ cogs in the rear you, in theory, get less drivetrain friction (as Richard Sachs also points out) and better chain alignment. I have not noticed that (drivetrain efficiency) myself going from 410mm to 430mm, though I am more finicky about gear adjustment/chain length than most. Also even today short chainstays are considered ‘racy’ as for example in this article here about Iban Mayo’s Orbea.

      As far as Merckx and Motta and your personal experience you do bring an interesting point. Regarding those two particular riders, we really don’t know anything more specific about their functionality, body proportions, etc and what ideas they had as far as a bike. Though long time ago I remember finding this image of Eddy Merckx’s bike in 1970 which is actually very close to Motta’s geometry you bring up. Though as your own test ride shows, terrain might have played a role in rider’s preference and riding style (Italy vs Belgium), though most riders (for sure not Eddy Merckx) are pretty clueless about geometry and fit and the sad part is they often do not realize how much better things can be. If you are to look at any framebuilding manual, longer rear triangle is almost always equated with touring and described as ‘whippy’ aka not racy and the myth gets perpetuated ad nauseum.

      Here I bring a personal experience. My current bike (430mm chainstays, 75mm BB drop and 59mm trail) is super easy to ‘guide’ into a turn, I have tried getting my ass off the side moto GP style and the bike just holds a line in a very confidence inspiring way – point and shoot type of thing. Once I was going into a tight S turn, I had a good line so I could have gone almost straight at full speed, though I thought I might have been going a touch too fast and as I started correcting and coming wide in the first part, if i hadn’t braked forcefully (and slid a bit the rear) I was going to overrun the second turn and end up in the kerb. It took quite some effort to ‘deflect’ the bike, it was in a way a touch too stable, though it was me who put it in the ‘wrong’ line and I am still fighting old cornering habits from years of riding twitchy oversteering bikes. That happened only once, though it showed that “Don’t think, you and the bike know where to go.”

      It is indeed sad that big manufacturers don’t put as much thought into geometry and one reason, in my opinion, is since their geometry is derivative – ie the guys in the past have been doing it that way so no need to upset the apple cart and create too much fuss with ‘experts’ coming out of the woodwork how things should be done. My design preferences and experience points towards that rear triangles are touch too short across the board (it is one aspect, though a significant one imo); not wrong, just not optimal. Cornering stability is never a bad thing and CX/gravel bikes that see rough terrain follow that philosophy since in a way there is no choice. Of course personal preference and riding style come into play, though once you get a bike that example allows you to corner faster and brake later before turns, your style will change, it’s constant process of fine tuning and without trying new things, you will not know. As well as the human body is amazing at compensating and there are some very good riders out there that can make any bike go fast, the design goal should always be to make the riders even better/faster.


  4. Ramon B Avatar
    Ramon B

    Thank you very much for sharing so much knowledge. It’s been a pleasure learning from your articles here.
    Perhaps the followiing bike-motorbike analogy is simply wrong and I’m asking nonsense… but as I just don’t know, I ask. Hope you don’t mind me asking.
    As a former motorcyclist, I have understood that a lower center of gravity requires more leaning (for the same turn, same speed, that is), this being increased the wider the tire.
    For this reason, a sport bike motorcyclist will need to lean more (or shift more body weight inwards) than his trail bike fellow at the same speed, same turn, while the lowrider motorist, with low cg and wide tires, would need to lean more than the sport bike… if his bike allowed such leaning.
    This makes me wonder if the common assumption of higher bottom brackets requiring more leaning and the adjacent bigger tires require more bb drop, shall be reconsidered, specially as bicycles tires are getting wider… and more specially as I like to keep pedaling on turns… and pedal scraping is an unpleasant experience.
    Sorry for my English and thank you for your time,

    1. Nikola Banishki Avatar
      Nikola Banishki

      Hi Ramon,

      Thank you for the nice words and you english is fine, no worries. I have been thinking about your comment and trying to see if i notice some of the things you mention on some of the bikes i have made (which re lower than what is ‘common.’. The bike i ride almost every day has 26in wheels, 185mm cranks and only 65mm BB drop so something like 270ish mm BB height, however with 44mm tires it probably is more like 260mm. This bike is VERY low and I have scraped the pedal even going straight over some road irregularities. This particular bike is extremely stable and I would say that it takes more effort (I am using the term lightly here) to lean and I can lean it


      in a corner to the point my front paniers/bags have scraped the road on a couple of occasions. Of course this bike has a low trail geometry and very long wheelbase – those two in my opinion are much more significant in making it stable. Higher bottom bracket to allow pedalling through corners would not change the behaviour much. A stable bike is more confidence inspiring and you can ride it faster and lean harder into turns. I think the weights and average speeds involved in bikes are much lower than motorcycles so the effect you mention is probably less pronounced. Honestly, i have never considered/measured lean angle, though wider tires together with a more stable geometry gives you the confidence to lean more at MUCH higher speeds. Lower trail (40ish instead of 56+mm) and long (420-440mm chainstays), in my opinion and experience, make for a bike that is more stable. Bottom bracket height is a distant third and if like in your case one needs to pedal through corners, you can have a custom frame built accordingly. What I am trying to say is that weight distribution between the two wheels is much more important than center of gravity when it comes to how much you can lean a bike into a corner. I think people have tested that 10-15mm of BB drop does not make a significant difference all other things being equal. It’s all a combination really of your riding style and what you want the bike to do.

      I am always open to hear other people’s experiences of course.=)


Leave a Reply

Your email address will not be published. Required fields are marked *