Correct face/path weighting to measured 0.76; add TrackMan newsletter archive

[dieterolson]

The correction

The launch-direction weight was given as w_f ≈ 0.85–0.87 for a driver. The only peer-reviewed measurement of the horizontal ratio — Wood, Henrikson & Broadie 2018, 157 golfers, 1,575 shots at 720 fps, filtered to centred strikes — gives 0.76 ± 0.08 driver, 0.69 7-iron, 0.61 wedge, with a PING Man robot returning 0.63 for the 7-iron.

Why this matters for OpenFlight specifically. Inverting the launch-weight relation to recover face angle amplifies any bias in measured launch direction by 1/w_f. At 0.76 that is 1.32, not the 1.15 implied by 0.85 — roughly twice the error sensitivity, on the single most important derived club parameter.

On provenance

The 85/15 figure comes from TrackMan's Ten Fundamentals, which asserts 85% in both planes. Against Wood's data the vertical claim holds (0.83) and the horizontal one does not (0.76). This is a vendor-versus-measurement disagreement, not a units error — worth stating precisely, because it bears on how much to trust a coefficient taken from vendor documentation.

The mechanism is also contested: TrackMan attributes the driver's higher weight to a smoother titanium face, while PING shows the loft dependence persists at constant friction and that below ~20° of incidence lower friction moves launch closer to the path. Any implementation fitting w_f empirically should record which explanation its calibration assumes.

New appendix: the TrackMan newsletter archive

Every official TrackMan newsletter URL 404s and Wayback holds only redirect stubs — but a complete live third-party mirror of issues #1–#10 exists. It contains:

• The only published TrackMan accuracy specification located anywhere: face angle ±0.6°, club path ±1.0°, attack angle ±1.0° for TrackMan III/IIIe at 95% confidence.
• The canonical club-delivery definitions, all specifying maximum compression.
• Bulge quantified: 2° of face-angle change per ½ inch of heel-ward strike, "for all drivers on the market."

A caution box sets those manufacturer figures against independent testing, which found club path met ±1° on 45% of shots — and that full clubhead parameters were returned on only 62% of shots overall and 19% for a utility wedge. A published tolerance and a per-shot success rate are different specifications, and only the first is ever advertised. That distinction is directly relevant to how OpenFlight should report its own performance.

Also

Adds a warning tcolorbox to the preamble for flagging corrected or contested claims, so they are not silently absorbed on later edits.

Verified: document compiles clean with pdflatex (1.05 MB PDF, no errors, no undefined references).

🤖 Generated with Claude Code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 8990eeb0ac

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Propagate the corrected weight through implementation guidance

When an implementer follows the report's actionable guidance, the correction here is not applied: sections/appendix-b-implementation.tex:133-136 still explicitly prescribes weights of 0.87–0.75, while sections/03-impact-physics.tex:148-155, sections/04-radar-systems.tex:213-218, sections/09-accuracy.tex:49-52, and sections/10-openflight-implications.tex:56-61 continue to specify the obsolete 1.15× error amplification. This would cause the implementation to retain the rejected model and understate driver face-angle bias despite this paragraph establishing 1.32×, so all downstream coefficients and recommendations need to be updated with this correction.

Useful? React with 👍 / 👎.