Retul or go(niometer) it alone?

Over the last year nothing in the world of tri bike fit has been more written about, and less understood, than the phenomenon of Retul. A triathlete wants, or reports on, or wonders about, a "Retul fit." When that happens I often ask for a description of a Retul fit?

During these moments of appearing the inquisitor to the "Retul fit" inquirer I might note that, boy, it's been awhile since I've had a really tasty "measuring cup" apple pie. This, because Retul is a measuring device, not a fit system. But it would be wrong to simply describe it as this. We sometimes, in our workshops, call Retul a "goniometer on steroids." But this also sells the device short, because there are qualitative benefits that exceed simply a level of precision that a goniometer may not give you.

A Retul will grant you the ability to track metrics that a goniometer does not track. But here's the real question, I think: Does a Retul, in the hands of a master tri bike fitter, result in a better end product—a better fit—than a goniometer? Most of the time, as of this writing, probably not.

But it can, and should, would, and will, deliver a superior fit as these fitters become more accustomed to the relevant features Retul offers. What I can tell you is that a Retul allows me, as a fitter, a chance to observe metrics that I can't observe in any other way.

What I might be talking about, really, is a motion capture system in general, not a Retul system in particular. But I don't have experience with any other motion capture system, so I don't know if Dartfish, for example, produces what I think are a Retul's most relevant and important metrics. So, I'll limit my comments to Retul, noting that Retul may not be the only method to generate specific metrics; rather that Retul bests my goniometer in certain ways I'll note here.

We'll get to those metrics in a minute.

Certainly, the most obvious benefits to a motion capture system, specifically Retul, are, first, the ability to detect metrics generated during a dynamic fit protocol using a dynamic measuring device. What do I mean by that?

I can determine your seat height simply by measuring your inseam. This is a static system for finding saddle height. "Static" because you don't have to pedal a bike, you can just stand there while I measure you. The problem with this system: It requires, or assumes, that you pedal your bike in a specific way. For example, two people who have the same inseam will have different saddle heights if one is a "toe pointer" and one is a "heel dropper." This, because what moves a crank are your hip and lower leg extenders, and your hip angle at top-dead-center, and your knee angle at bottom-dead-center, are really what matter most. These angles describe or inform the advantage your levers will have over the crank.

This argues for a dynamic measuring system—one that tracks knee angle (and maybe hip angle) through the pedal stroke. But if I use a goniometer (a big protractor that measures the angles of the body), then I have to stop you during the execution of your "dynamic" fit protocol to measure you using a "static" tool.

That makes a goniometer a less precise measuring tool than a Retul. Still, in the hands of a good practitioner, a goniometer can be almost as precise as a Retul. This feature alone is not enough to say that a Retul generates a better fit than a goniometer.

That established, Retul does a very nice job of tracking the knee angle around the pedal circle, and it generates a metric that is harder to measure with a goniometer: knee angle flexion. This is the knee angle at top-dead-center, and this is going to become a more important metric in fitting as practitioners wrestle to get their arms around a truth that attends TT and tri bike fits: crank lengths used by riders are often too long. By looking at knee angles at both the pedal stroke's top and bottom you can assess both saddle height and crank length in tandem.

Then there is the ability of a Retul to capture and store fit data, and to generate norms and stats based on the quantity of data archived. Let's assume that, over time, data is batched by dozens or hundreds of Retul practitioners around the world to a single server. Any of these fitters can see whether the fits he's executing produce a set of metrics that follow or deviate from norms.

This is a slick feature, especially when a light bulb goes on in the brain of a fitter that a specific metric in which he hasn't previously evinced any interest all of a sudden becomes relevant to him. The Retul tracks all these metrics even though the fitter hasn't previously cared about half of them. When he suddenly discovers a metric's importance, he can check Retul's database and see whether all the accumulated data of fitters worldwide produce a result of statistical relevance. In other words, Retuls are now generating the raw materials for inquiries that few if any of us have, as of yet, conceived.

But this also does not mean that, as of today, a Retul produces a better fit in the hands of a good practitioner than a goniometer.

If there is a game changer, it is this: a Retul offers the best method of which I'm aware to measure angles relative to the horizon, which we might call "global" angles as opposed to "rider-centric" angles. What I mean is this: a goniometer measures body angles. And this is important, as far as it goes. But, let's take a case in point: seat angle. One problem encountered during fit sessions is the tendency of some riders to change their saddle positions as their saddles are moved fore and aft. So, if I'm trying to figure out whether I like you better at 79° of seat angle than 78° (and whether you like it better yourself), you may tend to nose-ride 78°, but move back a bit in the saddle when I position you at 79°. Unless I catch you doing this, what just happened is this: I executed two trials at two different seat angles that were, for practical purposes, the same trial—the saddle may have been in a slightly fore or aft position, but you "normalized" for this by sitting in a different place on the saddle.

Here's what a Retul does: A "pick-up" or "lead" on the metatarsal joint of the cycling shoe notes its most forward protrusion during each pedal stroke which, of course, happens when the crank (on the drive side) is at 3 o'clock. Because the foot is attached to the pedal, it can't move. The rider can move fore and aft, but, the pedal can't move. Retul's metatarsal lead can move a little, because the rider's plantar angle changes as he rides more forward (a steep seat angled rider might appear like he's a toe pointer when he really isn't, because his body has been rotated forward around the axis of the bottom bracket).

Nevertheless, that metatarsal lead isn't going to change its position in space very much, but, as the rider moves forward, his knee will certainly change its orientation in space. A Retul notes what is analogous to "knee over pedal spindle," except it's landmarks aren't quite the same as those used in KOPS.

The ability to note when the knee is behind, right over, or in front of the metatarsal is a "global" measure of rider position, instead of a "rider-centric" metric. I can position you at 76° of seat angle, or 80° of seat angle, and all your "rider-centric" angles—your hip angle, knee angle, shoulder angle—might remain the same. There would be no way for me to know, just by looking at these rider-centric angles, where you ride in orientation to the horizon, that is to say, what your seat angle is, or how flat your back is.

But Retul produces two metrics that are global: knee over metatarsal, and the angle of your back (trochanter to acromion) relative to the horizon. This allows a fitter to know, once and for all, whether a rider is actually riding in a steeper position: If I position a rider at 78° of seat angle, then position him again at 79°, only to have that rider self-correct or normalize back to 78° by sitting 1.5cm or so farther back on his saddle, I'll know if I'm using a Retul, because his knee-over-metatarsal distance will not have changed.

Likewise, Retul's readout of a rider's back relative to the horizon is an absolute—global—measure that (all things equal) impacts frontal area. A goniometer does not do this.

What a Retul does for the clever and intuitive fitter is generate metrics that no one has thought of using before. Hip angle at TDC is a great example. This angle is hard to measure with a goniometer, for a variety of reasons, not the least of which is a difficulty in palpating the greater trochanter at TDC. In any case, this angle is important because there certainly is a minimum usable angle if one wants to still produce optimal power. But fitters have been late to account for this metric. Hip angle flexion, and knee angle flexion, can more easily enter into the lexicon of bike fitting metrics commonly used, as can back angle, because a good motion capture system can "capture" these metrics better than any static measuring instrument.

These metrics are, often, either ignored by our F.I.S.T. Protocol, or, we measure them but use different landmarks. For example, we use the center of the femur/tibia joint line as our knee landmark, Retul uses the lateral condyle. A line from the greater trochanter through the lateral end of the clavicle (the acromion) is what Retul uses to determine the line made by a rider's back; we use trochanter to the convex protrusion of the clavicle.

These sorts of variances in landmarks force one who uses a F.I.S.T. protocol with a Retul system to adjust the rider-centric angles that he hopes to achieve during his fitting process. For example, an included knee angle of 144° using F.I.S.T. landmarks might be 142° using Retul's landmarks. This doesn't change the end product of the fit. It just means that a practitioner trained in F.I.S.T. principles starts looking for 142° instead of 144° if he positions the Retul lead on the lateral condyle. If he decides to keep using the femur-tibia joint line, then, he keeps looking for 144° because, though he's using a Retul, he places the leads according to F.I.S.T. landmarks. The only problem with this is if and when Retul's worldwide library of archived fits becomes relevant—it is only relevant if one set of landmarks is used for every fit, and this is something we, at F.I.S.T., and they at Retul ought to address in the near future.

Note that, for all the above discussion, the Retul is a tool. It's not a system. It's a very powerful tool, and, the use of this tool means that it's easier to look at more angles quickly; to look at angles in flexion more precisely; and, to consider global instead of simply rider-centric angles. And, the plethora of metrics it generates will impact fit systems, that is, F.I.S.T. or any other protocol may morph to take advantage of metrics that Retul generrates.

Nevertheless, Retul remains one of several important tools a fitter ought to have in his arsenal. In my mind, a fitter armed with a Retul, but without a good, adjustable fitting apparatus (a fit bike), is still operating at a huge deficit. It's like a car mechanic owning an expensive diagnostic machine in a service bay that lacks a hoist.

The description here was not designed to be reader-friendly, that is to say, it isn't written to appeal to end-users who lack extensive knowledge of bike fit. If this has all been above your head, not to worry: Below are links to articles that'll get you up to speed. Mostly, this has been designed to explain to practitioners what a Retul will give you that a goniometer won't. But for all readers, I hope for at least one take-away: It isn't a Retul fit versus a F.I.S.T. fit, versus a Serotta or any other type of fit. We demonstrate Retul to every F.I.S.T. Workshop attendee. To consider a Retul fit as distinct from a F.I.S.T. fit is to wonder whether you should be consider open heart surgery, or sphygmomanometer surgery.

A sphygmomanometer, of course, is that device with the cuff you use to measure blood pressure—it's not antithetical to open heart surgery but complementary to it. What bike fit practitioners lack, as of now, is the ability to know with confidence what a Retul system adds to the fit process. Some of the answers are explained above.

But there is that other amorphous Retul feature: the analysis of metrics that matter, but we don't yet know yet that they matter. For this reason Retul is, or will be, the gift that keeps on giving.