WAB / Q3 2026

Westinghouse Air Brake Technologies Corporation

Reported: October 21, 2026

Published forecast

2026-10-19T23:15:34.987786+00:00

Opus 5.5 High in Shortcut

Shortcut

2.72

Shortcut normalized EPS

-0.8% vs Street

Off by $0.11 per share

Street

2.74

Street normalized EPS

Frozen October 18 snapshot

Off by $0.09 per share

Real

2.83

Issuer adjusted diluted EPS

Street win

Reported result · October 21 ↗ · Issuer adjusted diluted EPS of $2.83; GAAP EPS was $2.42. Original unrounded forecasts and frozen Street estimate retained.

USD per Wabtec diluted common share. Percentages use unrounded forecasts.

Model download

Explore the assumptions and calculations behind the forecast.

Open in Shortcut ↗Download XLSX ↓

Shortcut’s reasoning.

The segment model estimates total revenue of $3,160.465 million. Freight and Transit operating income, less the unallocated operating cost assumption, produce adjusted operating income of about $687.216 million.

After $76 million of net interest and other expense, adjusted pre-tax income is $611.216155 million. Applying the assumed 24.5% tax rate and $2 million of noncontrolling interests gives adjusted net income attributable to shareholders of $459.468197 million.

Dividing by 169.111568 million diluted shares gives adjusted diluted EPS of 2.716953089669583, versus the frozen Street estimate of $2.74. The complete public export preserves all 77 EPS dependency cells from the reviewed forecast.

The tax rate and growth assumptions remain judgments. Applying a full-year tax guide to this quarter does not establish the uniquely correct quarterly rate. The separate 60.9% beat probability blends the historical cohort estimate and a model-implied estimate tail; it does not multiply forecast EPS.

On the record.

Reporting basis: Issuer adjusted diluted EPS / CapIQ normalized EPS
Forecast completed: 2026-10-19T21:45:36.597961+00:00
Published: 2026-10-19T23:15:34.987786+00:00

Model assumptions and review notes

MEDIUM: The approximately 24.5% full-year tax guide is applied to Q3 as a judgment. The preserved guidance does not establish whether it includes first-half discrete items. First-half adjusted tax was about 22.9%; assuming Q4 pre-tax income equals Q3, a full-year catch-up interpretation would imply roughly 26.1% second-half tax and EPS near $2.659. This is a conditional alternative, not a compelled correction.
Growth components are approximate judgments. Transit assumptions of 14.5% and 18.5% differ from rough component arithmetic of about 13% and 17.5%; the Frauscher assumption of $43 million differs from a simple roughly $41.7 million annual-to-quarter estimate. The financial assumptions were preserved.
The opening share anchor is the July 17 cover count rather than a verified June 30 count. Early-July buybacks may overlap with the full-quarter buyback assumption. Small-overlap sensitivity examples are not bounds: an extreme assumption that all roughly 0.70 million buybacks occurred before July 17 would change EPS by about $0.011. Actual timing is unproved.
Historical repurchase counts are rounded. The 0.82 million Q2 input differs from the note citing 817,959 shares; Q1 uses a rounded 1.0 million. Derived historical repurchase prices are indicative. The forecast buyback price of $287 is a judgment, rather than the exact $287.89 mean quoted in its note.
Some assumption notes quote historical percentages approximately: the workbook arithmetic is about 20.63%, 25.42% and 31.90%, while the notes say 20.5%, 25.6% and 31.8%. These note differences do not alter the preserved financial inputs.
Historical GAAP amortization and cash-flow add-backs use different bases, while forecast amortization and the add-back are both assumed to be $92 million. Segment adjusted operating income and residual SG&A also involve judgment. Simplified balance-sheet, equity, liability and noncontrolling-interest classifications remain limitations; a balanced model is not independent validation.
The cash-flow measure is a model CFO-minus-capex convention. An exact issuer-defined free-cash-flow definition was not identified in the reviewed cached evidence. This additional metric is tracked, not scored.
MEDIUM: Some sources were retrieved after the information cutoff. Displayed pre-cutoff publication dates, including October 14 metadata for the preview with an October 21 event in its URL, do not prove immutable page content as of the cutoff. No precise capture timestamp is invented.
Current S&P 500 membership is not independently established by a 2019 inclusion release. Market capitalization alone qualifies for the frozen large-company prior. The 50/50 blend, 0.07 estimate standard deviation and $2.745 rounding threshold in the probability calculation are judgments, not independently established empirical calibration.
Presentation limits remain: raw CSV headers may clip at their default row height; wrapped notes can create tall rows; reference tabs retain a contextual period band above nonperiod tables. Some annual guidance statuses are fixed as pending and will not update automatically when future actuals are entered. The current snapshot correctly has no reported results.
Two recalculation-engine results were typed as early-1900 dates. Direct numerical arithmetic for Balance Sheet D40 and D44 matches their stored values; those ISO strings are not exact Excel serial-date equivalents. No workbook formula or value was changed for the engine artifact.

Research sources