Why is mathematics important in a bicycle wheel? A wheel looks like a circle held by many thin spokes, but it is really a tensioned structure. The rim must be round, centred laterally, correctly positioned between hub locknuts and supported by spokes whose tensions suit their side, material and component limits. Small adjustments interact, so measurement and geometry matter.
The mathematics includes circles, coordinates, vectors, trigonometry, averages, variation and tolerances. Yet a wheel is safety-critical. This article explains the model with synthetic examples and records; it does not tell a reader to ride, repair or approve a wheel. Physical inspection and service should follow the bicycle and component makers' instructions and a qualified mechanic's judgement.
Park Tool's current explanations distinguish lateral and radial truing and describe spoke-tension measurement. DT Swiss publishes wheelbuilding tools and manuals for its products. Those sources are useful precisely because they treat tension, trueness and tool use as related but distinct.
Choose a Reading Route
- Begin with the tensioned structure for the big idea.
- Follow the wheel geometry for coordinates and triangles.
- Compare kinds of trueness for measurement.
- Study tension statistics for variation and uncertainty.
- Use the student plan for safe, non-repair modelling.
Spokes Work Mainly in Tension
A common beginner picture imagines the hub hanging from the upper spokes. A more useful structural model starts with every properly tensioned spoke pulling between hub and rim. When a load is applied, tension changes around the wheel; the pre-tensioned system shares the change while the rim keeps the geometry coupled.
Tension is a force and is commonly expressed in newtons, sometimes through kilogram-force conversions in workshop charts. A tensiometer may display an instrument reading that must be converted using the spoke's material and dimensions. The scale reading is not automatically the tension.
Direction matters
Each spoke pulls along its own line. A vector can be resolved into components. If a spoke tension T makes angle θ with the wheel's centre plane, its lateral component is T sin θ and its component in the plane is T cos θ under the declared convention.
For T = 1000 N and θ = 5°, the lateral component is about 87 N because sin 5° ≈ 0.087. This does not mean one spoke independently supplies a useful 87 N correction. Opposing spokes and the rim form a coupled system. The calculation only reveals how bracing angle changes the direction of force.
Geometry Connects Hub, Rim and Spoke
Place the hub centre at the origin. A rim hole can be represented by its radius R and angle around the wheel. A hub-flange hole has its own radius r, angular offset φ and lateral distance d from the centre plane. The spoke length follows from the three-dimensional distance between those two points.
In a simplified model,
L² = R² + r² + d² − 2Rr cos φ.
The cosine term captures the tangential offset created by the lacing pattern. The lateral term d² captures the flange's position away from the rim centre plane. Real calculators also require precise definitions for effective rim diameter, flange diameter, flange spacing, hole count, crossing pattern and spoke-hole offsets.
Why measurements must match definitions
Nominal rim diameter is not the same as effective rim diameter used for spoke length. Measuring to the wrong surface can shift every result. A flange dimension may be given as diameter while the equation needs radius. One side's centre-to-flange distance may differ from the other.
Dimensional analysis catches some errors. Every term inside the length-squared equation must have units of mm² if the dimensions are in millimetres. It cannot catch a correctly labelled but incorrectly measured radius.
Crossing changes angle, not only appearance
In a radial pattern, a spoke travels more directly from flange to rim. In crossed patterns, it leaves the flange at a tangential angle and may cross other spokes. Changing the crossing count changes φ and therefore the required length.
The aesthetically neatest line on a sketch is not automatically compatible with the hub, rim, spoke count, brake loads or manufacturer requirements. Geometry narrows possibilities; component instructions decide what is permitted.
Dish Is a Symmetry Question With an Asymmetric Answer
A front rim-brake wheel may look approximately symmetric about its centre plane. A rear wheel with a cassette, or a disc-brake wheel, often has unequal flange positions. The rim must still be placed at the specified lateral position, but left and right spokes may have different bracing angles and average tensions.
Suppose the left side has a larger bracing angle than the right. To balance lateral force, fewer newtons may be needed on the better-braced side for the same opposing component. A simple equilibrium statement is
T_L sin θ_L ≈ T_R sin θ_R,
when aggregated consistently across corresponding spoke groups. If θ_L = 7° and θ_R = 4°, then T_R/T_L ≈ sin 7°/sin 4° ≈ 1.75. This is an illustrative ratio, not a target for a real wheel.
Equal tension is not the same as balance
On an asymmetric wheel, demanding identical left and right tension can pull the rim away from its required dish or create another conflict. A better comparison is usually within each side, while side-to-side target relationships come from the actual geometry and component guidance.
Trueness Has More Than One Axis
Lateral trueness describes side-to-side deviation as the wheel rotates. Radial trueness describes variation in radius—high and low spots sometimes called hop. Dish describes the rim's lateral position relative to the hub's reference faces. Tension balance describes the distribution of spoke forces.
A wheel can pass one check and fail another. It can be laterally straight yet radially uneven, or round yet off-centre. That is why a single phrase such as “the wheel is true” needs a defined measurement.
Record deviation as a function of angle
Let x(α) be signed lateral deviation at wheel angle α. A table sampled every 15° can be plotted around one rotation. Positive and negative signs show direction. The total indicated runout is max(x) − min(x), while the maximum absolute deviation is max|x|. Those values answer different questions.
For readings from −0.6 mm to +0.4 mm, total indicated runout is 1.0 mm, while maximum absolute deviation is 0.6 mm. Reporting only “1.0 mm out” hides the reference and direction.
Patterns reveal shape
One broad positive region opposite a broad negative region resembles an offset or one-cycle pattern. Two alternating lobes resemble a second harmonic. Local dents or measurement noise look different. Fourier language can describe such patterns, but a real rim may combine permanent damage, seating, spoke effects and setup error.
Pattern recognition is a diagnostic aid, not permission to adjust a safety-critical wheel without competence.
Average Tension Is Not Enough
Consider one wheel side with synthetic spoke tensions 980, 995, 1005, 1010 and 1010 N. The mean is 1000 N. Now consider 700, 900, 1000, 1100 and 1300 N. The mean is also 1000 N, but the distributions are very different.
Range, standard deviation and coefficient of variation reveal spread. Coefficient of variation is standard deviation divided by mean, often expressed as a percentage. It supports comparison when average levels differ, but only when values are positive and the measurement process is consistent.
Position matters
A list of tensions without spoke positions loses information. Alternating high and low readings around the rim can signal a different pattern from one local cluster. A circular plot indexed by spoke number preserves adjacency.
For an asymmetric wheel, label side as well. Combining both sides into one mean can manufacture “variation” that merely reflects two legitimate target levels.
Measurement has uncertainty
A tensiometer deflects a spoke and maps the reading through a conversion table. Repeatability, instrument condition, spoke dimensions, coatings, reading position and technique all affect the estimate. If repeated measurements vary by ±20 N and the conversion is coarse, reporting 1003.7 N is false precision.
Take repeated readings on a synthetic or approved training setup. Estimate within-operator spread and compare operators. The purpose is to learn how measurement behaves, not to certify a wheel.
Small Adjustments Form a Coupled System
Turning one nipple changes one spoke's effective length and tension, but the rim transmits the effect to neighbours and often to the opposite side. A local linear model might write
Δx = AΔu,
where Δu is a vector of small adjustments and A is a sensitivity matrix mapping them to lateral and radial changes. The matrix reminds us that one input can affect several outputs.
Real response can be nonlinear because of friction, spoke wind-up, seating and damage. That is why professional procedures use small adjustments, repeated measurement and stress-relief checks rather than one large calculated move.
Thread pitch provides a scale
If a fictional nipple advances 0.45 mm per full turn, one quarter-turn changes its axial position by about 0.1125 mm before elastic response and seating are considered. That number does not equal rim movement. It is only the screw displacement in the simplified geometry.
Torque also cannot serve as a universal substitute for tension. Friction at threads and contact surfaces changes the torque-to-tension relationship. The same torque can produce different tension under different lubrication and component conditions.
Tolerance Is a Set, Not One Number
A complete wheel assessment can involve maximum tension, minimum useful tension, within-side balance, lateral runout, radial runout, dish, spoke line, stress relief and component condition. Satisfying one target may disturb another.
Mathematically, the acceptable region is the intersection of several constraint sets. An adjustment that reduces lateral deviation but pushes one spoke outside its approved tension range is not a successful optimisation.
Damage changes the problem
A cracked rim, damaged eyelet, bent component, corroded spoke or uncertain history is not merely another measurement point. It may invalidate the model and require replacement or specialist inspection. No amount of statistical smoothing makes damaged hardware safe.
What the Mathematics Cannot Tell You Alone
The equations do not confirm that a rim is structurally sound, that a tyre is seated, that fasteners are secure or that a bicycle is safe to ride. Component-specific limits and procedures matter. Wheelbuilding also involves touch, sound, sequence and experience that a few formulas cannot reproduce.
Use mathematics to ask clearer questions: Which axis is being measured? Which side? Which reference? Which conversion chart? How variable are the readings? Then let qualified mechanical judgement govern physical work.
Common Misconceptions
All spokes should have the same tension
Uniformity is normally evaluated within an appropriate side or group. Asymmetric geometry can require different side averages.
A straight-looking wheel has good tension
Visual trueness cannot reveal every tension imbalance or structural issue.
Tensiometer reading equals newtons
Many tools require a spoke-specific conversion. The raw scale value and force are different quantities.
One nipple turn moves the rim by the thread advance
Thread advance changes effective spoke length. Rim response is coupled through the whole structure.
Lower runout is always a safe improvement
Not if tension, dish, damage or component limits worsen. All constraints must be checked.
A Student Learning Plan
Stage 1: Draw the geometry
Model a rim circle, hub flange and spoke triangle with fictional dimensions. Label radii, lateral offset and angular offset.
Stage 2: Resolve vectors
Calculate lateral components at several bracing angles. Explain why a smaller angle requires a different tension relationship.
Stage 3: Plot runout
Use supplied measurements around a circle. Compare total indicated runout, maximum absolute deviation and pattern shape.
Stage 4: Analyse variation
Compare two synthetic tension sets with the same mean. Add range, standard deviation, position and side labels.
Stage 5: Build a constraint map
List trueness, dish, tension and condition constraints. Explain why no single score replaces them.
For Parents and Teachers
Use diagrams, photographs, safe models and synthetic data. A hoop with strings can illustrate angles, but it is not a rideable wheel. This boundary lets students explore vectors and measurement without turning the lesson into unsupervised repair.
Ask learners to state the reference for every number. “0.5 mm” is incomplete until they say lateral or radial, relative to which datum, at what angle and with which instrument.
Did You Know?
- A small bracing angle makes the lateral component much smaller than the spoke's total tension.
- Two tension sets can share the same mean while having completely different balance.
- Lateral runout, radial runout and dish describe different geometric errors.
- A nipple's thread advance is not equal to the rim's movement.
Frequently Asked Questions
Why are spokes tensioned before the wheel is loaded?
Pre-tension creates a stable coupled structure in which load changes are distributed through the rim and spokes.
What does truing mean?
It means improving geometric alignment, commonly including lateral and radial trueness, while respecting dish, tension and component condition.
Why can left and right tension differ?
Hub asymmetry creates different bracing angles, so lateral force balance may require different average tensions.
Is torque a tension measurement?
Not universally. Friction strongly affects the relationship, so direct calibrated tension measurement is preferred in the cited procedures.
Can a student true a bicycle wheel from this article?
No. The article supports mathematical learning only. Safety-critical inspection and repair require appropriate tools, manuals and competence.
Which school topics appear here?
Circles, coordinates, Pythagoras, cosine, vectors, means, spread, graphs, matrices and tolerances all have natural roles.
Useful Next Reading
- Compare measured performance responsibly in Why Mathematics? | Sports Statistics, Speed and Performance.
- Connect wheel rotation to drivetrain ratios in Why Mathematics? | Bicycle Gears, Cadence and Mechanical Advantage.
- Build broader foundations through the Mathematics Learning Hub.
A bicycle wheel is a beautiful lesson in connected reasoning. Geometry sets directions, tension supplies forces, measurement reveals deviation and constraints prevent a locally attractive answer from becoming a bad whole-system decision.
A Practical Mathematics Studio
Use synthetic or openly released teaching data. These investigations expose the mathematics and its limits; they do not authorise bicycle-safety, repair or component-selection decisions.
Investigation 1: Spoke polygon
Plot hub and rim hole coordinates for one wheel side. Use manufacturer dimensions rather than guessing. Use synthetic or openly released teaching data. Begin by naming the response variable, input variables, units and prediction before calculating. Preserve raw values and unrounded intermediates in a table, then make one graph whose axes communicate the model clearly. Add an independent hand estimate and a dimensional or limiting-case check. Deliberately introduce one unit error and describe the numerical signature rather than merely correcting it. Label every quantity observed, assumed, fitted or derived. Record the model domain and one condition that would require a richer model. Ask a peer to reproduce the result from the written procedure without seeing the answer. If the reproduction differs, locate whether the ambiguity arose in definitions, units, rounding or software defaults. End by explaining the result to a younger student in three sentences. This remains a classroom calculation, not professional validation or a safety, production or security decision.
Investigation 2: Spoke length
Apply a declared triangle model to estimate spoke length. Leave exact build selection to an approved calculator or builder. Create a data dictionary before entering a spreadsheet or script. State how each row was obtained, which values are exact by definition and which are measurements or simulated observations. Calculate once with full precision and once with premature rounding, then compare the final difference. Vary the principal input across at least five sensible values and look for linearity, curvature or instability instead of reporting only two endpoints. Keep signed residuals if a model is fitted, because absolute errors hide direction. Include a graph and a small results table that another person can audit. Write a one-paragraph uncertainty note distinguishing input uncertainty from model-form limitations. A peer should be able to reconstruct one row from the formula and raw data alone. State what evidence would falsify the interpretation. Do not present the exercise as a certified engineering, manufacturing, metrology or cryptographic assessment.
Investigation 3: Dish geometry
Compare left and right bracing angles on an asymmetric rear wheel. Explain why equal tensions are not always expected across sides. Treat the investigation as a comparison of hypotheses, not a hunt for an attractive number. Write a baseline model and at least one plausible alternative, then predict where their outputs should diverge. Hold all unrelated inputs fixed while varying the chosen factor. Preserve coordinate, sign, phase, size-weighting or bit conventions beside the data. Use a ratio, residual or normalised error that has a declared denominator. Repeat the calculation with that denominator changed and explain why the percentage changes. Check one extreme case where the answer should approach zero, unity or another known limit. Make the graph before writing the conclusion and note any outlier without deleting it. Separate repeatability of one prepared sample from coverage across positions, times or populations. The final claim should match the observed range and must not be extrapolated into real operational approval.
Investigation 4: Vector components
Resolve one spoke tension into radial and lateral components. Keep force direction and units visible. Plan the computation so that errors leave visible fingerprints. Save the raw input, transformed input, intermediate result and final result in separate columns or variables. Insert one deliberate sign reversal, one missing observation and one hidden reset, each in a separate copy, and document how the plots or checks respond. Compare an analytical shortcut with the more complete calculation over a range where the shortcut is expected to fail. Report the largest absolute difference and where it occurs. Include conservation, balance, boundedness or monotonicity checks appropriate to the topic. If a fitted curve is used, inspect residual clusters and changing spread instead of relying on one fit statistic. Have a peer review only the assumptions first, then the arithmetic. Keep the conclusion educational and conditional on the fictional data.
Investigation 5: Tension mean
Calculate the mean of synthetic readings from one side. A mean cannot reveal one dangerously low or high spoke. Build a sensitivity map rather than changing several settings at once. Choose a baseline, vary one input downward and upward, and express output change in both original units and a dimensionless ratio. Then select a second input and repeat. If interactions seem likely, test a small two-dimensional grid and identify where the one-factor interpretation breaks down. State why the chosen ranges are plausible for the teaching model. Preserve failed or surprising runs and annotate them. Compare computational resolution or sample length at three levels so numerical artefacts are not confused with physical behaviour. Provide a hand estimate for the order of magnitude. Finish with a decision table using cautious language—stable, sensitive or unresolved—rather than safe or unsafe. Classroom evidence cannot authorise a product, structure, process or security control.
Investigation 6: Tension spread
Compute range and coefficient of variation for the same readings. Report the conversion chart and spoke specification. Make uncertainty visible at every stage. Assign each input a source and a plausible interval, then identify which inputs are correlated rather than automatically treating them as independent. Propagate uncertainty with a simple high-and-low calculation or transparent simulation, and retain the seed or full synthetic samples for reproduction. Compare the spread caused by measurement variation with the shift caused by changing the model assumption. Plot both if possible. Explain why more decimal places do not narrow an interval supported by weak inputs. If a result lies near a fictional decision boundary, rewrite the conclusion to reflect the chance of crossing it. Ask a peer to challenge the largest assumed uncertainty and rerun the analysis. Report the conclusion as evidence about the model, never as professional certification.
Investigation 7: Lateral runout
Record indicator deviation by angle around a teaching wheel image. Separate signed displacement from total indicated runout. Audit representation choices. Recalculate after changing the coordinate origin, phase wrapping convention, histogram bins, mesh labels or binary mapping while preserving the underlying situation. A correct physical or probabilistic conclusion should change only where the definition genuinely changes. Show one example where a visual choice exaggerates or hides a pattern. Use identical axes for fair comparison and include sample count or resolution in captions. Test an invariant such as total mass, probability, force, energy sign or average ratio. Where values are averaged, identify whether the mean is number-, mass-, volume-, time- or state-weighted. Ask another student to explain the plot without reading the conclusion; revise labels if their interpretation differs. The display supports reasoning but cannot replace domain review.
Investigation 8: Radial runout
Graph radius deviation against wheel angle. Do not mix hop with side-to-side error. Separate verification from validation. First verify arithmetic or code against a case with a known answer, including units and at least one manually calculated row. Next ask whether the teaching model represents the intended phenomenon and list the omitted mechanisms. Refine numerical resolution or increase synthetic sample length and record whether the chosen quantity stabilises. A stable answer can still arise from an unrealistic model, so compare with an alternative assumption or openly documented benchmark. Record software version, solver or library settings and stopping rules. Avoid tuning settings after seeing the desired result; predeclare the comparison instead. Summarise numerical error, input uncertainty and model-form uncertainty separately. No classroom agreement with a benchmark should be described as product or system validation.
Investigation 9: Harmonic shape
Compare one-lobed and two-lobed synthetic runout curves. Describe pattern without treating it as a repair instruction. Design the investigation to reveal dependence over position, time or sequence order. Keep the original ordering, then compare with a deliberately shuffled copy and explain which statistics change. Examine at least two lags, locations or neighbourhood scales rather than only the overall average. Plot local values alongside the aggregate so compensating errors are visible. State whether successive observations are plausibly independent and why that matters to the calculation. If repeated measurements share one specimen, source or initial state, do not count them as independent coverage. Add a block or subgroup analysis and note any drift. Preserve the random seed for simulated data but use more than one seed before generalising. The conclusion should describe detected structure without claiming causation from the pattern alone.
Investigation 10: Local adjustment
Model how a small paired tension change affects lateral position. Real response is coupled and must be checked incrementally. Create an adversarial edge-case test. Identify the smallest, largest, most symmetric and most irregular inputs allowed by the classroom model, predict the expected behaviour, and then compute it. Include zero or a limiting value only when mathematically meaningful. Watch for division by a small number, logarithms of invalid values, wrapped angles, negative geometry, impossible probabilities or solver singularities. Replace silent software errors with explicit checks and explanatory messages. Compare the edge cases with the ordinary baseline on the same scale and describe why a method that works in the middle may fail near a boundary. Record the first assumption that breaks. The exercise is about model literacy; it is not permission to explore hazardous physical extremes.
Investigation 11: Round count
Convert a nipple turn into thread advance using a fictional pitch. Do not generalise the value across hardware. Prepare a compact reproducibility package: raw synthetic data, formulas or code, version information, one chart, a results table and a short read-me file. Give every file and column a meaningful name and avoid values copied manually between tools. Re-run from a clean state and confirm that outputs are regenerated rather than cached. Ask a peer to alter one declared input and predict the direction of change before execution. Compare the prediction with the result and investigate any disagreement. Add a limitations section naming data coverage, numerical resolution, model assumptions and the next useful measurement. Archive corrections instead of erasing them so the reasoning trail remains visible. Conclude with what the calculation demonstrates, what it does not demonstrate and which qualified professional or authoritative standard would govern a real application.
Investigation 12: Torque warning
Explain why nipple torque is not a direct universal tension measure. Friction and lubrication change the relationship. Use synthetic or openly released teaching data. Begin by naming the response variable, input variables, units and prediction before calculating. Preserve raw values and unrounded intermediates in a table, then make one graph whose axes communicate the model clearly. Add an independent hand estimate and a dimensional or limiting-case check. Deliberately introduce one unit error and describe the numerical signature rather than merely correcting it. Label every quantity observed, assumed, fitted or derived. Record the model domain and one condition that would require a richer model. Ask a peer to reproduce the result from the written procedure without seeing the answer. If the reproduction differs, locate whether the ambiguity arose in definitions, units, rounding or software defaults. End by explaining the result to a younger student in three sentences. This remains a classroom calculation, not professional validation or a safety, production or security decision.
Investigation 13: Uncertainty budget
Combine tensiometer repeatability, conversion and reading-position effects. Round reported tension accordingly. Create a data dictionary before entering a spreadsheet or script. State how each row was obtained, which values are exact by definition and which are measurements or simulated observations. Calculate once with full precision and once with premature rounding, then compare the final difference. Vary the principal input across at least five sensible values and look for linearity, curvature or instability instead of reporting only two endpoints. Keep signed residuals if a model is fitted, because absolute errors hide direction. Include a graph and a small results table that another person can audit. Write a one-paragraph uncertainty note distinguishing input uncertainty from model-form limitations. A peer should be able to reconstruct one row from the formula and raw data alone. State what evidence would falsify the interpretation. Do not present the exercise as a certified engineering, manufacturing, metrology or cryptographic assessment.
Investigation 14: Calibration check
Compare a tool reading with a known reference curve. Do not invent a calibration outside the maker's procedure. Treat the investigation as a comparison of hypotheses, not a hunt for an attractive number. Write a baseline model and at least one plausible alternative, then predict where their outputs should diverge. Hold all unrelated inputs fixed while varying the chosen factor. Preserve coordinate, sign, phase, size-weighting or bit conventions beside the data. Use a ratio, residual or normalised error that has a declared denominator. Repeat the calculation with that denominator changed and explain why the percentage changes. Check one extreme case where the answer should approach zero, unity or another known limit. Make the graph before writing the conclusion and note any outlier without deleting it. Separate repeatability of one prepared sample from coverage across positions, times or populations. The final claim should match the observed range and must not be extrapolated into real operational approval.
Investigation 15: Before-after plot
Plot synthetic tension and runout before and after an adjustment sequence. Avoid claiming causation from one wheel. Plan the computation so that errors leave visible fingerprints. Save the raw input, transformed input, intermediate result and final result in separate columns or variables. Insert one deliberate sign reversal, one missing observation and one hidden reset, each in a separate copy, and document how the plots or checks respond. Compare an analytical shortcut with the more complete calculation over a range where the shortcut is expected to fail. Report the largest absolute difference and where it occurs. Include conservation, balance, boundedness or monotonicity checks appropriate to the topic. If a fitted curve is used, inspect residual clusters and changing spread instead of relying on one fit statistic. Have a peer review only the assumptions first, then the arithmetic. Keep the conclusion educational and conditional on the fictional data.
Investigation 16: Constraint map
List roundness, lateral truth, dish, tension and stress-relief constraints. Show how improving one can disturb another. Build a sensitivity map rather than changing several settings at once. Choose a baseline, vary one input downward and upward, and express output change in both original units and a dimensionless ratio. Then select a second input and repeat. If interactions seem likely, test a small two-dimensional grid and identify where the one-factor interpretation breaks down. State why the chosen ranges are plausible for the teaching model. Preserve failed or surprising runs and annotate them. Compare computational resolution or sample length at three levels so numerical artefacts are not confused with physical behaviour. Provide a hand estimate for the order of magnitude. Finish with a decision table using cautious language—stable, sensitive or unresolved—rather than safe or unsafe. Classroom evidence cannot authorise a product, structure, process or security control.
Investigation 17: Safety boundary
Identify damage, cracks and uncertainty that stop classroom analysis. Refer physical repair and inspection to a qualified mechanic. Make uncertainty visible at every stage. Assign each input a source and a plausible interval, then identify which inputs are correlated rather than automatically treating them as independent. Propagate uncertainty with a simple high-and-low calculation or transparent simulation, and retain the seed or full synthetic samples for reproduction. Compare the spread caused by measurement variation with the shift caused by changing the model assumption. Plot both if possible. Explain why more decimal places do not narrow an interval supported by weak inputs. If a result lies near a fictional decision boundary, rewrite the conclusion to reflect the chance of crossing it. Ask a peer to challenge the largest assumed uncertainty and rerun the analysis. Report the conclusion as evidence about the model, never as professional certification.
Investigation 18: Wheel record
Archive rim, hub, spokes, lacing, side, readings, tool and conditions. Follow the component maker's limits. Audit representation choices. Recalculate after changing the coordinate origin, phase wrapping convention, histogram bins, mesh labels or binary mapping while preserving the underlying situation. A correct physical or probabilistic conclusion should change only where the definition genuinely changes. Show one example where a visual choice exaggerates or hides a pattern. Use identical axes for fair comparison and include sample count or resolution in captions. Test an invariant such as total mass, probability, force, energy sign or average ratio. Where values are averaged, identify whether the mean is number-, mass-, volume-, time- or state-weighted. Ask another student to explain the plot without reading the conclusion; revise labels if their interpretation differs. The display supports reasoning but cannot replace domain review.