Cold River Canine Anatomy and Working Structure · Chapter 2
Rear Angulation in Boerboels: The Stifle, the Power Train, and the Mechanics of a Working Dog
Why the rear assembly deserves as much scrutiny as the head and front end.
By Cold River Boerboels · Updated October 20, 2026
A working dog needs a rear assembly that can produce force, control it, and apply it repeatedly. The stifle—the canine knee—is an important link in that system. Its relationship with the pelvis, femur, tibia, hock and foot influences how the entire limb functions. Correct structure means appropriate angulation and stability for the breed, rather than the most bend someone can put into a photograph. [1], [2]
At Cold River Boerboels, we evaluate that assembly through a working background. In weight pull, a broad head and an impressive chest cannot finish the job by themselves. We want a dog that can organize its body, establish traction and move against resistance with coordinated, repeatable effort. That is why rear structure receives deliberate attention in our breeding program.
The Cold River position: A head may catch your attention. The whole dog has to do the work.
From Chapter 1: follow the assembly from the croup to the ground
Chapter 1 examined the croup and the underlying pelvic foundation. Here we follow that foundation through the hip and femur to the stifle, then down the tibia and fibula to the hock, metatarsals and toes. Rear angulation is the combined arrangement of those segments. The visible bend at one joint cannot describe the complete assembly. [2]
The croup is the external rump region. Pelvic inclination concerns a defined bony reference. Stifle angulation concerns the relationship between femur and tibia. These are related parts of a connected system, but they are different measurements. A sloping rump does not, by itself, tell us the angle of the knee.
Figure 1. Original anatomical map of the connected rear limb. Segment axes and joint landmarks are simplified; this is not an X-ray or a scale drawing. The lateral foot points forward, toward the left. Anatomy background: references [3–5].
What is inside the canine stifle?
The stifle includes the femorotibial articulation and the patellofemoral articulation. The patella is the kneecap within the knee extensor mechanism; the patellar ligament connects it to the tibial tuberosity. The tibial plateau is the upper articular region of the tibia. The menisci lie between the femoral condyles and tibia. Imaging studies show why these structures must be considered together rather than treating the knee as two sticks joined by a pin. [4]
The cranial and caudal cruciate ligaments guide and restrain relative motion of the femur and tibia. In an anatomical and functional study of 50 canine stifles, Arnoczky and Marshall demonstrated that different ligament components tighten and loosen through the range of motion; cutting the cruciate ligaments produced marked instability. Their findings establish that stability depends on coordinated soft-tissue restraint as well as the bones. [3]
Collateral ligaments also help control joint motion. Their behavior changes with flexion, including their contribution to rotational restraint. The stifle therefore permits controlled rotation as well as flexion and extension; it is not a simple hinge. [5]
Yair and colleagues experimentally documented coupled tibial rotation and translation during passive flexion and extension. Their work involved cadaver limbs, not dogs pulling competitively, but it demonstrates the three-dimensional mechanics concealed beneath an apparently simple side profile. [6]
The power train: why geometry matters under load
Muscles generate force; bones act as levers; joints permit and guide movement; tendons transmit force; ligaments and other tissues help control unwanted motion. The limb then applies force against the ground. Calling the rear assembly a power train is useful shorthand, provided we remember that it works with the trunk and forelimbs.
Joint torque depends on force and its perpendicular moment arm. Joint power depends on torque and angular velocity. Consequently, a knee that looks more bent is not automatically a more powerful knee. Muscle size, architecture, activation, leverage, movement speed and stability all contribute. Gregersen and Carrier’s jumping-dog research showed that changing gearing at the knee can help the extensor muscles operate in a manner favorable to power production. That finding concerns movement, not a preferred standing angle for Boerboels. [7]
Colborne and colleagues compared hind-limb joint mechanics in six Labrador Retrievers and six Greyhounds. Their inverse-dynamics work found breed differences in how the joints handled mechanical work and power. The practical lesson is that power is distributed through the limb and depends on the animal and task. A single stifle angle is not a universal performance score. [8]
Their separate Greyhound study found that increasing trotting velocity increased local power bursts without changing the basic work patterns. Loading and speed matter when interpreting movement. A free trot, acceleration, a jump and a slow heavy pull are different mechanical tasks. [9]
Our working interpretation: In weight pull, we want useful force applied against the surface with controlled limb placement. Slipping, uncontrolled rotation or repeated loss of position interrupts the effort. The rear assembly is central to that task, but the harness, footing, technique, conditioning, body mass and forelimb contribution also matter. We have not identified a controlled Boerboel weight-pull trial establishing one stifle angle as optimal.
Straight stifles, moderate angulation and excessive angulation
In conformation discussion, a straight or insufficiently angulated stifle has a relatively open femur–tibia relationship in a comparable stance. Excessive rear angulation describes too much bend or an exaggerated overall assembly for the breed. Moderate, balanced angulation lies between those extremes. A dog can also have different combinations of hip, stifle and hock relationships; not every rear fault belongs in the same category. [1], [2]
Figure 2. Original geometry comparison using the same internal-angle convention. Larger numbers indicate less bend. These are deliberately simplified segment models, not three diagnosed dogs. The middle panel is not a scientifically validated ideal, and the numbers are not Boerboel judging thresholds. A whole dog must be assessed in a comparable stance and in motion.
When the rear is too straight
The working-structure review by Zink and Schlehr identifies limited rear angulation as a potential source of increased axial torque and stress on the stifle and tarsal ligaments. This is a biomechanical concern supported by expert structural interpretation, not a quantified prediction that every straight-stifled dog will become injured. Limited angulation may accompany restricted-looking rear movement, but standing shape and actual joint excursion must be evaluated separately. [2]
From a practical perspective, watch how the dog brings the limb forward, supports the body and completes push-off. Does it move freely and symmetrically? Does it repeatedly compensate through the trunk or alter its tracking? Those observations deserve investigation. They do not establish the cause on their own. Pain, neurologic disease, conditioning, handling and surface can also change gait.
When the rear is over-angulated
More angulation is not an unlimited advantage. A rear that is exaggerated for the breed may require greater stabilizing effort and may show unwanted lateral or rotational movement. The review illustrates extreme examples in German Shepherd Dogs. Those examples teach an anatomical principle; they are not evidence that Boerboels should resemble German Shepherd Dogs or that the same degree of bend has the same effect in every breed. [2]
Figure 3. Published illustration and movement photographs from Zink and Schlehr (2020), original Figure 13. The German Shepherd examples illustrate excessive rear angulation and associated instability discussed by the authors. They do not establish a Boerboel injury rate or an optimal Boerboel stifle angle. Illustration by M. Schlehr; © 2020 Zink and Schlehr, reproduced unchanged under CC BY. Source: reference [2].
What does the Boerboel breed standard ask for?
The AKC Official Standard of the Boerboel calls for a sturdy, muscular hindquarter; sound, strong, moderately angulated stifles; and balance with the forequarters to support propulsion. Its movement description emphasizes powerful, purposeful, fluent and agile movement. It also prioritizes proportion and sound movement over size. Those requirements make the rear assembly part of breed type, rather than an optional sporting extra. [1]
Different registries can phrase their standards differently. Here we use the linked AKC standard as the verified primary standard. We do not assign an unsupported universal numerical knee target, or treat conformation terminology as interchangeable with a radiographic measurement. A Boerboel should be assessed against its applicable registry standard and its movement, rather than an angle borrowed from another breed.
Figure 4. Published moderate rear-assembly illustration, Zink and Schlehr (2020), original Figure 12. The vertical reference and colored ground markers are the authors’ conformation teaching method, not a radiographic test or a Boerboel-specific pass/fail rule. Illustration by M. Schlehr; © 2020 Zink and Schlehr, reproduced unchanged under CC BY. Source: reference [2].
Degrees need landmarks: a standing photograph is not an X-ray
Giansetto and colleagues measured standing dogs using radiographs and defined mechanical axes. After excluding three dogs for positioning, 21 remained. Femorotibial angles ranged from 129.5° to 156.6°, with a mean of about 145.3°. The sample was small and included several breeds. These are observed standing measurements, not proof that the mean maximizes athletic ability, and not a breed-standard limit. [10]
Figure 5. Actual research measurement diagram and standing radiograph from Giansetto, Picavet, Lefebvre and Balligand (2022), original Figure 1. Panel a shows positioning; panel b identifies the femoral axis, tibial axis and measured stifle angle. Reproduced unchanged under CC BY 4.0; © 2022 the authors, Veterinary Sciences 9:644. Source: reference [10].
A visual angle through palpable landmarks, a mechanical-axis angle on a radiograph, a tibial plateau angle and a flexion angle referenced to full extension are different quantities. For example, Yair and colleagues used 0° to denote full extension in their experimental system. It would be incorrect to compare that number directly with an internal femur–tibia angle drawn on a standing dog. [6], [10]
External croup slope: The surface contour of the rump. Do not confuse it with: a measured stifle angle.
Femur–tibia angle: The relationship of defined thigh and lower-leg axes. Do not confuse it with: tibial plateau inclination.
Tibial plateau angle: Orientation of the proximal tibial joint surface relative to a defined tibial reference. Do not confuse it with: the amount of visible bend at the knee.
Dynamic joint excursion: How far the joint moves during a defined activity. Do not confuse it with: a static photograph.
Does a steep croup cause straight stifles?
We cannot present that as an established scientific rule. The pelvis positions the hip sockets within the rear assembly, but the femur–tibia relationship also depends on segment proportions, joint geometry and posture. A steep-looking croup and an upright stifle can coexist. That observation does not show that one caused the other, or that they must occur together.
When evaluating a dog, examine the croup and underlying pelvic region, then independently assess the hip, stifle and hock, limb lengths, placement and movement. If a breeder repeatedly observes a combination in a particular family, record it as a program observation. Demonstrating a breed-wide correlation would require standardized measurements across a representative population. We did not identify such evidence for the proposed steep-croup/straight-stifle link in Boerboels.
Stifle angulation, injury and what the studies actually prove
Dennler and colleagues studied 16 limbs from nine cadaver dogs. They demonstrated that the relationship between the patellar ligament and joint-surface references changes with stifle flexion. Their analysis also illustrates that the reference chosen changes the calculated crossover of shear loading. This supports careful discussion of geometry and load; it does not supply a universal safe standing angle. [11]
A computer simulation study of intact and cruciate-deficient canine stifles found ligament loads in the intact model were particularly sensitive to tibial plateau angle. Again, tibial plateau angle is not the same as externally visible stifle bend. Surgical and simulation research helps explain joint mechanics, but cannot automatically rank breeding dogs from photographs. [12]
Kanno and colleagues provide a useful caution against simplistic injury claims. In six cadaveric Beagle stifles, the cruciate-deficient joints showed greater cranial displacement under one loading test at 135° and 90° than in hyperextension. The study did not show that a straighter live dog is safer; it showed that instability depends on angle, ligament status and test conditions. We should not cherry-pick a single number to support either extreme. [13]
The American College of Veterinary Surgeons describes cruciate disease as multifactorial, involving degeneration, body condition, genetics, breed, conformation and other factors. Correct structural selection is a sensible breeding objective, but no rear outline guarantees freedom from cruciate disease. Lameness, swelling, pain, repeated skipping or a new change in stance warrants a veterinary assessment rather than an online conformation diagnosis. [14]
Figure 6. Published rear-view alignment comparison from Zink and Schlehr (2020), original Figure 14. The authors describe the left example in a Labrador context, mild external rotation in herding breeds centrally, and excessive rotation on the right. Breed and degree matter. This is a rear-view comparison, not a side-view stifle-angle diagram. Illustration by M. Schlehr; © 2020 Zink and Schlehr, reproduced unchanged under CC BY. Source: reference [2].
The high-protein myth: food does not supply a stifle-angle dial
A claim we encounter in the Boerboel community is that high dietary protein makes a puppy straight in the stifle. Before accepting that explanation, ask for a controlled study demonstrating that protein itself changed femoral proportions or caused loss of rear angulation. We did not identify a study establishing that claim. That is a statement about the evidence located, rather than a claim to have proven a universal negative.
Nap and colleagues tested Great Dane puppies on isoenergetic diets containing 31.6%, 23.1% or 14.6% protein on a dry-matter basis over 18 weeks. Differences in protein intake did not produce demonstrable effects on calcium metabolism or skeletal development under those conditions. This is direct experimental evidence against blaming the tested higher protein intake for skeletal developmental harm. [15]
The study did not measure a Boerboel-specific stifle-angle outcome, test every commercial food or establish that unlimited protein is beneficial. It nevertheless gives no support to the assertion that feeding the higher tested protein level removes correct angulation. Dry-matter percentages also should not be compared casually with an as-fed label without accounting for moisture.
A related 1993 study assessed growth and GH/IGF-I patterns in 17 Great Dane puppies on the same three protein concentrations. The low-protein diet produced small effects on growth velocity and IGF-I and was considered marginal during part of early growth. This is why “protein does not affect bone growth” is too broad: adequate protein matters for normal growth. It does not follow that adequate or higher protein causes straight stifles. [16]
Cold River’s answer: If someone says the food straightened a dog’s stifles, ask them to show where protein was isolated as the cause. A repeated explanation is not a controlled study.
What nutrition can change—and why that distinction matters
The protein question must be separated from total food intake and mineral balance. In a controlled Labrador feeding study, limited-fed dogs developed less radiographic hip dysplasia than their pair-mates. That is evidence about food allowance and hip development, not proof of a protein-driven stifle-angle defect. It supports keeping growing dogs appropriately lean rather than maximizing rapid weight gain. [17]
Great Dane experiments also examined increased calcium and phosphorus intake using radiography, histology and biochemical assessment. They found skeletal developmental disturbances associated with mineral excess and imbalance. The nutritional concern is the complete diet, including absolute mineral intake and energy supply; calcium-to-phosphorus ratio alone cannot describe its suitability. [18]
For puppy owners, the practical objective is a complete diet appropriate for large-breed growth, an appropriate body condition and consistent monitoring. Protein is one part of that diet. Adding calories, minerals or supplements indiscriminately is not a method for improving structural angulation.
An adult or growing dog’s visible stance can change with muscle development, fatigue, pain, weight, handling and foot placement. A change in appearance is not proof that the femur was reshaped by protein. If a dog appears to lose angulation, document comparable photographs and movement, review the complete feeding and health history, and investigate the actual cause.
How Cold River evaluates the rear assembly
Our concern is that rear structure is too often overshadowed by the head, chest and overall mass. In our own experience reviewing dogs and having breeding conversations, insufficient rear angulation deserves more attention than it receives. That is our observation, not a prevalence study of the Boerboel population.
Within our program, the goal is moderate, breed-appropriate angulation with a strong, coordinated rear and a balanced front. We look at a dog standing naturally, in a consistent evaluation stance and moving from the side, toward us and away from us. We consider development over time rather than treating one stacked photograph as a final verdict.
Map the assembly. Locate the pelvic region, hip, femur, stifle, tibia, hock and foot. Check the complete relationship.
Standardize the comparison. Use level footing, similar camera height and a genuine side view. Avoid judging through a stretched or crouched pose.
Observe movement. Look for comfortable, symmetrical limb use, coordinated rear drive and stable tracking at an appropriate pace.
Consider the individual. Age, body condition, soundness and training history affect what you see.
Use records. Health screening, veterinary findings, family history and repeated observations add information a photograph cannot provide.
Our sport gives us repeated opportunities to observe how dogs coordinate their bodies under resistance. It does not make us exempt from evidence, and a competition result is not an orthopedic examination. We use performance observations alongside structural assessment and health information.
Cold River’s breeding objective is a complete Boerboel: breed character, stable temperament, sound structure and the physical capacity to work. The rear assembly is a central part of that objective. If the front end gets the advertisement, the rear still has to help deliver the performance.
Questions owners and breeders ask
Is a straight-stifled dog automatically lame?
No. An upright outline is not a diagnosis. Soundness must be evaluated clinically and in motion. The concern is structural suitability and possible mechanical disadvantage, not a guarantee of pain or injury.
Does more rear angulation mean more pulling power?
No. Angulation must work with stability, muscle, leverage, traction, conditioning and technique. No controlled study located here identifies one standing stifle angle that maximizes Boerboel weight-pull performance.
Can high protein turn correct stifles into straight stifles?
The studies cited here do not support that claim. The controlled Great Dane trial found no demonstrable adverse skeletal-development effect from the tested protein differences. It was not a direct Boerboel stifle-angle experiment.
Does protein have no role in femur growth?
Protein is required for normal growth. The unsupported claim is that higher protein in an otherwise appropriate diet specifically changes femoral proportions or removes stifle angulation. Those are different propositions.
Can you determine a knee angle from the croup?
No. Evaluate the croup and pelvis, then the limb and joints independently. A proposed association needs measured evidence; a surface slope cannot substitute for a joint measurement.
Continue the series
Next in the series: Chapter 3: Boerboel topline and rear-high structure.
For the other half of the balance between front and rear, read Chapter 5: Boerboel front assembly and shoulder angulation.
Start with Chapter 1: dog croup anatomy and rear angulation. For the work this structure has to do, read our dog weight pulling guide, the documented 22,039-pound world-record pull and our Boerboel strength and performance questions. For growing puppies, visit our Boerboel nutrition page. New chapters are posted in the Performance Dog Journal.
References and evidence notes
Primary experiments are distinguished from a structural review and an official breed standard. Where only an abstract was accessible, the discussion is limited to the reported methods and conclusions. References checked October 20, 2026. The published images below retain their source attributions; their authors do not endorse Cold River.
American Kennel Club. Official Standard of the Boerboel. Primary breed standard; qualitative hindquarter, balance and gait requirements.
Zink C, Schlehr MR. Working Dog Structure: Evaluation and Relationship to Function. Front Vet Sci. 2020;7:559055. Open-access review combining research and expert observation; not a controlled Boerboel performance trial. Published Figures 12–14 reproduced under CC BY.
Arnoczky SP, Marshall JL. The cruciate ligaments of the canine stifle: an anatomical and functional analysis. Am J Vet Res. 1977;38:1807–1814. Primary anatomical/functional study of 50 stifles.
Computed tomographic arthrography of the normal canine stifle. Vet Radiol Ultrasound. 2004. Primary imaging study identifying normal intra-articular and supporting structures.
Collateral ligaments of the canine stifle joint: anatomic and functional analysis. Am J Vet Res. 1981. Primary cadaveric anatomy and function study.
Yair N, Yiapanis C, Ben-Amotz R, Milgram J. The effect of joint orientation on passive movement of the dog’s stifle. Front Vet Sci. 2023;10:1207164. Primary passive-motion cadaveric experiment; not loaded sporting performance.
Gregersen CS, Carrier DR. Gear ratios at the limb joints of jumping dogs. J Biomech. 2004. Primary jumping biomechanics; small experimental sample and task-specific findings.
Colborne GR et al. Distribution of power across the hind limb joints in Labrador Retrievers and Greyhounds. Am J Vet Res. 2005;66:1563–1571. Primary inverse-dynamics comparison, 12 dogs.
Colborne GR et al. Effect of trotting velocity on work patterns of the hind limbs of Greyhounds. Am J Vet Res. 2006;67:1293–1298. Primary velocity and joint-work study.
Giansetto T, Picavet PP, Lefebvre M, Balligand M. Determination of the Stifle Angle at Standing Position in Dogs. Vet Sci. 2022;9:644. Primary standing radiographic study; 24 recruited, 21 analyzed. Figure 1 reproduced under CC BY 4.0.
Dennler R et al. Inclination of the patellar ligament in relation to flexion angle in stifle joints of dogs without degenerative joint disease. Am J Vet Res. 2006;67:1849–1854. Primary radiographic cadaveric study, 16 limbs.
Evaluation of varying morphological parameters on the biomechanics of a cranial cruciate ligament-deficient or intact canine stifle joint with a computer simulation model. Am J Vet Res. 2014;75:26–33. Primary computer-model sensitivity study; results depend on model assumptions.
Kanno N et al. Tibial displacement with stifle joint flexion and cranial cruciate ligament transection in the dog. Vet Comp Orthop Traumatol. 2014;27:277–284. Primary robotic ex vivo study, six Beagle stifles.
American College of Veterinary Surgeons. Cranial Cruciate Ligament Disease. Specialist clinical guidance on multifactorial disease; not a primary experiment.
Nap RC et al. Growth and skeletal development in Great Dane pups fed different levels of protein intake. J Nutr. 1991;121(11 Suppl):S107–S113. Primary controlled dietary-protein experiment; not a Boerboel stifle-angle trial.
Nap RC et al. Age-related plasma concentrations of growth hormone (GH) and insulin-like growth factor I (IGF-I) in Great Dane pups fed different dietary levels of protein. 1993. Primary growth/endocrine study, 17 puppies; illustrates why protein adequacy still matters.
Kealy RD et al. Effects of limited food consumption on the incidence of hip dysplasia in growing dogs. JAVMA. 1992;201:857–863. Primary paired feeding study; food allowance and hips, not isolated protein or stifle conformation.
Schoenmakers I et al. Effects of diets with different calcium and phosphorus contents on the skeletal development and blood chemistry of growing great danes. Vet Rec. 2000;147:652–660. Primary mineral-feeding experiment; distinguish mineral supply from protein.
Image licensing: Creative Commons Attribution 4.0. Original Cold River diagrams are teaching schematics, not measured data or reproduced experimental findings.
World-Record Performance Boerboels: Bred to Work, Proven to Win.