A lure simulator helps compare design choices and select prototypes. It does not replace weighing, sealing, float checks or a controlled water test of the finished lure. MakeMyLure uses a reduced-order physics model and a separate 2D flow lab; neither certifies a real lure’s diving depth, durability or fishing results.
What is being modeled?
The swim simulator combines the lure’s mass and buoyancy, simplified body and bill forces, wake-driven oscillation, line loading and modeled hooks or joints. For printed hard baits, generated internal structures and configured hardware contribute to the as-built mass model. That estimate still depends on the materials and parts entered, rather than measurements of your own print.
The flow lab computes a two-dimensional slice at low Reynolds numbers. It is useful for observing wake structure and informing frequency-related tuning. Its absolute drag, lift and stall behavior cannot be transferred directly to a full-size, three-dimensional lure at fishing conditions. A 2D flow picture is not a full 3D CFD validation.
What does “checked against published studies” mean?
The project benchmark rebuilds published test articles as closely as its parametric shapes allow, then compares matching conditions and observables. Some dimensions, hardware or attachment positions must be inferred when a source does not report them. Measured tank data, manufacturer ratings and builder experience carry different evidence grades; they are not pooled into one claim of proven accuracy.
These literature comparisons have also guided model development and parameter choices. They are development benchmarks, not a separate, blinded holdout study of independently manufactured lures. Passing a code regression check shows that a recorded behavior remains reproducible; it does not by itself show agreement with a new physical experiment.
Read the error definition before the score
The saved benchmark report generated on October 27, 2026 compares 743 value metrics: 495 fall within its chosen error threshold, 170 are marginal and 78 exceed it. These are related metric comparisons across reconstructed articles, not 743 independent physical experiments. The report is a dated development result, not a fresh benchmark run for this page.
Normalized error e = |simulated − measured| / max(0.25 × |measured|, 2 × source scatter)
The benchmark calls e ≤ 1 a pass and 1 < e ≤ 2 marginal. Its 25% reference scale is part of that scoring rule, not a guaranteed ±25% prediction interval. Averaging metrics, or dividing passing metrics by their count, does not establish the probability that your lure will behave as shown.
Known gaps that should affect your decisions
- Topwater walk-the-dog sweep remains smaller than the intended reference behavior. A convincing-looking animation is not evidence that surface retrieves have been quantitatively validated.
- Lipless-bait stability can change with integration resolution, and some reconstructed vibration-lure benchmarks still show large disagreement. Check sensitivity to settings rather than assuming every stable run predicts a stable real lure.
- Raising ballast does not consistently reproduce the expected increase in roll relative to yaw. Use the model to frame a comparison, and test the finished variants before relying on that trend.
Make a comparison that can be checked
Keep the same design source, mass basis, scenario, line, water properties, speed and retrieve pattern. Reset each run, use equal durations and analyze the same post-startup window. Report whether an angle is a half-amplitude or peak-to-peak, and whether a frequency comes from body motion or line tension. Tank tow depth and maximum cast-retrieve depth answer different questions.
What a physical validation record needs
Record the printed geometry and version, printer and slicer settings, actual finished mass, materials, hooks, rings and coating. Check sealing and rest attitude first. For action tests, record water conditions, tow or retrieve speed, line, rod or tow-point position, camera scale and repetitions. If speed or the environment cannot be controlled, label the result qualitative instead of reporting a precise error percentage.
Check static ballast against a real build
Use the two blank sheets to compare the predicted as-built mass with the same fully cured assembly, then record repeat still-water observations and mass before and after immersion. Keep the design version, printing settings, hardware, finish and raw photos with the record. Fill the files locally: blank values mean not provided, not zero; this pack contains no measured results.
A surface float check does not measure fully submerged displacement or validate swimming action. A mass change after immersion can indicate water uptake, but residual surface water and scale limits can affect it. Use the filling guide’s definitions and keep estimates separate from measurements; no universal pass threshold is supplied.
Inspect a documented example
The bill-angle case publishes numerical inputs, telemetry, a source fingerprint and read-only models. The ballast case reports static mass and displacement estimates. Both use public presets; neither currently reports a manufactured prototype or physical water-test result.
Research informing the model
These sources describe related physical models or measured lure experiments. They inform the project’s implementation and comparisons; citing them does not imply that their authors tested or endorsed MakeMyLure.