I know that the industry does little to discourage the broadly held notion that there is a blueprint of measurements that applies across the board (this horse is a wide — this horse is a medium), but in reality horse shapes and tree shapes are more complicated than that.  Added to which, the whole shape of the tree has to fit, not just the head.  But only the head measurement is even nominally considered in the M, MW, W designation.  This gives you none of the other information you need about the shape of the tree.   Is it medium and flat or medium and curvy?  How steep do the rails descend from the head?  How long are the tree points?

Ah, but there’s more!  For example, if you chopped off the ends of the tree points on a 36 cm tree so that the spread between them was 32 cm, without altering any angles on the head, it would fit considerably wider, even though it would have a narrower absolute distance between the points.  In other words, if the arms of the tree are longer, it will fit narrower, even if the absolute distance between the tree points is greater, which would mean it’s technically wider since it would be, say, a 36 cm tree with long arms as opposed to a 32 cm tree with short arms.  Some companies (generally European) use these numeric designations for trees, but they are only meaningful (which is being kind) if you are comparing trees that are otherwise identical.  Clearly one cannot encapsulate all relevant information about the fit of a tree in a single word, but that’s the blueprint the industry puts forward and it’s the standard that most consumers accept as credible.

This is a topic I feel very strongly about, and I think the best thing we can do for our horses is to drill down a bit deeper when it comes to understanding the parts of the saddle that are unseen.  In reality, it is most often the unseen bearing structure sandwiched between the visible padding that makes or breaks the fit.

Patrick Keane Dressage Saddle from Advanced Saddle Fit

It’s simply incredible to us Americans (or at least to me) how truly “cottage” much of the saddle industry is in the UK, where all of our saddles are made.  It’s extremely human-intensive in comparison to some of the big French, Italian, and other European saddle makers who turn out extraordinarily consistent saddles one after the next, because they are produced almost entirely by automation.

For example, a high-end French or Italian saddle typically comes out of very modern, high-tech, fully automated manufacturing plant.  Human hands do very little until the last finishing bits, when a person actually hammers in the enamel nail head.  The labor component of this sort of modern Continental saddle production is a fraction what it is in the traditional saddleries of England.  In these up-to-date European plants, even the wool is precision-blown into the panels by an air compressor (though some use foam panels exclusively).  There is little or no variation in these saddles, which are consistently lovely in their way.

There is, of course, a reason that we have not embraced this type of saddle up to this point, which is that horses are far from uniform.  Designing and producing saddles meant to suit “the mean” (whatever that would be in horses or humans) is a legitimate business objective, but whether this best serves the interests of individual horses, who definitely don’t conform to a mean in reality, is another matter.

Some days the surgical exactitude of this production method seems like my definition of heaven compared to the relative quaintness of the British saddle makers, some of whose premises could be stage sets for the very darkest side of Dickens – or Kafka.  On the other hand, we greatly appreciate being able to have a say in what trees, panels, and fit features we want in our saddles, based on what we have learned from experience over many years fitting saddles in the field to thousands of horses.

There are certain trees we have found to be very horse-friendly and versatile to fit.  We want to use these trees when and where appropriate, and we want the panels done the way we think works best to address a particular need.  Sometimes, I want to be able to say to the human who is making the saddle, “For this saddle, I would like you to put the stitch line up 1 cm, and cut the panel 4 cm deeper into the sweat flap.”  It would be difficult to do this anywhere but Walsall, because it requires the sort of individual customization in production that can only be achieved where handcrafting is not only feasible, it’s the norm.  Patrick Keane’s saddles, for example, have between 40 and 50 hours of hand work in each one, compared to the industrialized European producers whose saddle production methods involve only a small human component.

Even the mid-range British saddles that are closer to being mass-produced than Patrick’s saddles are still made on a bench by a skilled person using hand tools, though not necessarily from beginning to end by the same person.  With most of the larger Walsall manufacturers, all of the component parts of the saddle are largely standardized, and they arrive on the bench in a plastic bin for the bench saddler to assemble them by hand.  This does limit our capacity to get customization of trees and patterns, though we can usually choose from a short list of standard variations.

Not all Walsall saddle makers, even at the small, cottage-craft level, are willing to make saddles any way other than the way they are accustomed to doing them.  If they are willing, they still may not truly grasp why we ask for certain features to be done a specific way, and they sometimes seem to think that we are asking for pointless variations or exceptions that make no sense.  This is because — shocking revelation here — saddlery is an industrial sector that is not very closely connected to the real world of horses.

There are certainly exceptions, but contrary to what consumers reasonably assume, even first-rate saddle makers don’t commonly have extensive, in-depth experience fitting their saddles to horses.  Most saddle makers spend the vast majority of their time making saddles.  They may tear themselves away from their bench from time to time to see some horses under saddle, but it is not particularly common for a saddle maker to be an experienced horseman, nor is it common for a saddle maker to do a great deal of field research on the efficacy of his product for a broad population of horses.

Saddle fitters like us generally have much more real world experience of a particular saddle than the person who made it, though quite often we lack sufficient technical knowledge of the design and manufacturing process to be as effective as we should be in suggesting improvements.  This may not be the most apt analogy, but the highly skilled technician who makes medical devices is probably not as adept at the actual installation as an experienced surgeon who sinks his hands into human flesh day in and day out.  But I wouldn’t necessarily trust the surgeon to try his hand at making one of those precision devices either.

Some days we rejoice in our good fortune that we have found some saddlers in Walsall who are attuned to our real-world needs and willing to listen.  At other moments in time, this has felt like the worst of both worlds: in some instances, we might be denied the option to get variations like a lined panel or an altered stitch line or a particular tree that isn’t one of their standards.  But neither do we necessarily get the precision uniformity of automated European manufacture.  In our worst-case experience with some of the highly promoted brands with wide name recognition, we have suffered the vicissitudes of human-intensive production without enjoying the benefits of flexibility and responsiveness that this should facilitate.

Admittedly, we sometimes we try variations on a theme that don’t work as well as we think they are going to.  It takes time and experience to know how well a particular fit solution works over time in the real world of horses. It is hard to know for certain whether a particular saddle or some particular feature of a saddle will perform to our expectations until we have acquired some experience with it over time.  As in most of life, the pattern of progress is often two steps forward and one step back.

For me, the durability of a good fit solution is critical.  A durable solution is one that is tolerant of changes in the shape of the horse’s back in motion, and tolerant of some change in the horse’s condition over time.  A durable solution is one that many similar horses appear to thrive in over time.  Of course no one can guarantee that a saddle that fits well today will necessarily fit just as well next year (especially if the horse gets fat), but what I can say for sure is that the better the fit on day one, the greater the odds it will still be working well on day one thousand.

Clearly there are trade-offs in different approaches to saddle design and production.  I don’t think there is enough information in the public domain for consumers to make optimal decisions about what is the best value for their money.  I will close this with my recurring refrain that I believe the way forward is in educating ourselves about what we are getting for our money when we buy a saddle.  What do you think?

These are some of the primary “fact-based” problems we run into all the time:

1. Since riders have credit cards and horses don’t, rider appeal nearly always trumps horse fit in designing saddles commercially. (If you doubt this, take a look at how saddles are described in marketing material, which is overwhelmingly biased toward how well the saddle will suit the rider.)
2. Very few riders (and not all saddle fitters) have sufficient knowledge of saddle design and technology to evaluate comparative fit considerations accurately. In this, they get little or no help from manufacturers.
3. Not every saddle maker/designer works with hundreds of horses in the field every year as we do.  Many of them see only a small handful of horses, and rarely do they systematically follow the progress of individual horses to see how well particular saddle solutions work (or don’t work) for different types of horses over time.

The most deep-down, almost primal assumption among riders is that somewhere in the universe there is a saddle that the rider will absolutely love.  Moreover, this “one true saddle” will be just as brilliant a fit for her horse as it is for her. The corollary to this assumption is that saddles can be customized or custom-made to meet the most exact requirements of the rider without compromising on optimal fit for the horse.  In fact, this is exactly what a “custom saddle” is supposed to achieve (according to this assumption).

In reality, whether the craftsmanship of a saddle is exquisite or gross, the actual technology is similarly crude, about on par with furniture, which varies greatly in refinement, but not in basic concept.  In saddles there is a single, shared bearing structure (the tree), with a seat on top and panels underneath.  Since all the pattern pieces have to meld harmoniously to the shape of the tree above and beneath, the reality is that it isn’t always possible to achieve a particular rider feel without affecting horse fit.

So, here’s the nub of it: do you, dear reader, want a saddle that is designed from the rider down to the horse, or from the horse up to the rider? That should be a rhetorical question, but sadly it isn’t. There are any number of saddles on the market that are immensely popular with riders precisely because they are designed so well for human appeal.  Apparently it’s not too hard to convince people that the saddles they love best for themselves will also do justice to their horses.

Needless to say, the geometry of trees and the physics of load-bearing structures were never specifically addressed in my training as a saddle fitter, nor have they featured in any significant way in any book or magazine article I have ever read on the subject.

Without a doubt, good saddle fitters have valuable insights about fit based on empirical knowledge and experience, and I have been fortunate to have had mentors who are arguably the best saddle fitters in the world.  But here is a source of persistent frustration.  Over the years, I have asked the great ones: how do you actually know that a particular saddle fits the horse?  When it works, how do you know exactly why does it work?  And how do you pinpoint the problem (and the solution) with accuracy when a saddle isn’t right?

In essence, the answer always boils down to: “I’ve done this for decades and I have a good eye.”  Decades of experience and a good eye are invaluable assets to any professional, of course, but I would argue that reliance on conventional wisdom is not a methodology for either research or teaching.

It isn’t much of a leap to accept that behind the conventional wisdom about saddles and how to fit them lies the actual physics of what happens statically and dynamically between the bearing structure of the horse and the saddle.  Yet no one, to my knowledge, is breaking down (or elevating, really) saddle fitting to the level of a systematic, science-based methodology, or even claiming with much conviction that it is important to do this.

Instead, saddle makers and saddle fitters (even acclaimed “authorities” who write books on saddle fitting) seem pretty convinced of how saddles should fit strictly on the basis of the handing down of conventional wisdom, with almost nothing in the way of provable, research-based evidence.  Why?  Because conventional wisdom has been the accepted foundation of saddle fitting for years and years, and shifts in paradigm are never cheap or easy.

Several years ago, I began to get an inkling of how exciting it would be if we, as saddle fitters, could gain greater insight into the engineering aspect of saddle design and fit.  This slim shaft of illumination came not from anyone involved directly with fitting saddles, but from an engineer in an allied industry.  The long and the short of it is that not all conventional wisdom about saddles - even when it is pure and untainted by commercialism – has a solid basis in science.

Saddles are not in any way exempt from the laws of physics and geometry, though on the whole saddle fitters don’t tend to work by that methodology.  Certainly the Pliance research is a major step in that direction, but from what we are able to glean from the trickle of information seeping out, it is still a long way from providing a durable framework for research based in scientific method, let alone the development of a science-based protocol for saddle fitting.

For now it remains commercially convenient for most saddle marketers not to reveal too much specific information about the trees in their saddles, especially since it is hard for any of us to assess the exact shape of a tree inside a saddle when it is essentially invisible.  Instead people tend to fixate on what can be seen on the surface – like the width of the channel, for example, which can sometimes be a very, very misleading indicator of the shape or angles of the actual bearing structure.  Moreover, to a great extent, consumers have shown themselves quite willing  to suspend disbelief and accept at face value some rather questionable claims about how saddles fit and perform.

What if there really is some treasure to be reaped from the application of physics and geometry to the black art of saddle fitting?  What if there is a saddle equivalent of parabolic skis, which is a perfect example of a simple but brilliant application of geometry and physics to a thousand-year old technology.  As any skier will attest, adding side cut to skis – giving a parabolic shape to the edge of the ski as opposed to a straight edge – produces a channeled kinetic energy that creates power through a turn, making it easy for a skier to execute turns that would require a great deal more brute strength on straight-edged skis.

Imagine how much better horses might be able to carry dynamic weight – and never forget the first and great rule of saddle fitting: that horses never evolved to carry weight in the first place – if we could find a way to design trees that can dampen or disperse brutalizing kinetic energy far better than even the best trees can do now.  Maybe it’s a pipe dream anyway, but for sure it will never happen if no one is motivated even to try.

These photos are clear examples of conflicting philosophies of saddle design:

I find it extraordinary that there can be such divergence of opinion about some of the most fundamental aspects of correct saddle fitting.  On several occasions, I have had the opportunity to hear presentations by well-known luminaries in the field who have developed their own brands based on particular design concepts.  I have tried to be as objective as possible in evaluating what I understand their approach to be, and to be open-minded to new concepts that might help me to be a better saddle fitter.

I haven’t always been disappointed.  The WoW saddle, for example, which is based on design technology that is entirely different from what we use in conventional saddle fitting, is mind-blowing; it’s conceptually ingenious; and as some of you know, we really tried to make the WoW product work for us and for our clients, but alas, the stars did not align.  I have often described WoW as a different paradigm in saddle fitting, but that’s not really accurate.  It is in fact a radically different technological means to the same conceptual end.  The path we have gone down instead is to seek out a wider variety of conventional saddles with good fit features.

The incredible thing to me is that clearly not every one in the professional community can agree even broadly on what good fit means!  According to conventional wisdom (by which I mean the teachings of the UK’s Society of Master Saddlers and the schools they have influenced), it means to achieve the least pounds per square inch of pressure by using a tree/panel configuration that will distribute the rider’s dynamic weight as evenly as possible over the muscular bearing area of the horse’s back.  In other words, you pick a suitable shape of tree in a suitable width, then you cushion and balance it with a suitable panel.  Cushioned, even weight distribution, like a well-fitting pair of athletic shoes: this is how I think it’s meant to work.

Not everyone is in agreement, however, and I have often come away from these presentations convinced that if the presenter’s approach is the right one, then mine has to be wrong.  Of course I don’t think I am wrong, but I’m not positive.  What if they’re right and I’m wrong?  Why can’t we know this?

These photos prompt the question:  Which of these is a better way to do panels?

Thanks to Pliance, the technology is available to test these duelling design theories head-to-head.  But one thing’s for sure: it ain’t going to happen until consumers force the issue.  I don’t see many saddle companies lining up to put their product to the science-based test.  I understand why; it could be really scary.

Even if a more objective means of measuring our success in getting an optimal saddle fit were readily available, there would still be commercial constraints in the production and marketing of saddles that complicate this effort.  But I’ll save that for next time.

More about trees

Wood spring trees are made by bending laminated wood strips around a form that looks like a horse’s back.  The wood spring tree takes on the shape of the form it is built on.  The shape is then re-inforced by steel bands and a head plate and gullet plate.

There are many other factors beside nominal tree width that affect the fit. Even a tree of suitable shape cannot do the job alone. No matter how much you narrow this tree, something more will be needed to prevent injury to this horse.

Synthetic trees are injection molded, using a plastic-like, nylon composite material.  The material for the tree is poured into a mold and it comes out in one solid piece.  Then its width at the head is established by pushing the points of the tree outward or inward.

In the manufacture of synthetic trees — which dominate in the lower half of the saddle market and are becoming increasingly prevalent even in the rarefied top end — the initial investment in a mold to produce a tree is expensive, but the production cost per tree is low.  This is one of several potent disincentives to producing saddles on many different shapes of tree for many different shapes of horse.

Typically there is a great deal more variety of shapes commercially available in wood spring trees, but this doesn’t mean that any individual saddle manufacturer uses a wide variety of different tree shapes.    From a production standpoint, this is a lot more complicated than it seems on the surface.

For reasons that are quite valid financially if not always beneficial in practical terms, many saddle manufacturers are producing on a single type of tree, which is then adjusted to different nominal widths.   The fundamental point is that the overall shape of the tree is what it is –  variations in width do not necessarily correlate to variations in the whole shape of the tree as the arms (tree points) are widened or narrowed and the nominal tree widths increases or decreases.  Inconveniently, the geometry of trees is actually a lot more profound than that.

Whether or not a particular manufacture chooses to build saddles on a wide variety of tree shapes, the universe of choices in wood spring trees is greater than the choices available in synthetic trees.   Where synthetic trees are used, their selling point seems to be that the material they are made from is more malleable, and therefore it is easier to make serial adjustments to the “width” of the saddle – provided that one is content with defining fit as being all or mostly about the fit through the tree points and not the overall shape and fit of the whole tree.

Simply widening this tree will still not make it a suitable shape for this horse.

Unfortunately, analyzing the suitability of tree shape is not something that figures prominently in many discussions of saddle fitting.  Trees can be either too curvy or too flat for a particular horse’s back, or the rails can be set too steeply and too close together or too flat and too wide apart for a particular horse, all of which can have a profound effect on the horse’s comfort and the evenness of the weight distribution.