HScreen · PWR

PWR · Quanta Services, Inc. 5d old

conviction2

coverage tierT3-PRIMARY-COMPLETE

horizon12 months

Return distribution probability-weighted scenarios, never collapsed to a point

-28%22%

+7%48%

+33%30%

EV +7.10%

why these probabilities: All legs built forward from USD 694.72 (2026-08-18) and FY2025 reported EPS of USD 6.87, priced 12 months out on FY2027E EPS, because by mid-2027 the market prices the following year. Consensus source for every growth rate cited: fmp:/stable/analyst-estimates?symbol=PWR&period=annual, as_of 2026-08-19, 36 covering analysts (grade EST). BASE = what the market expects, DELIVERED. Consensus revenue growth 34.0% in FY2026 and 15.2% in FY2027, EPS +46.4% then +16.9%, i.e. FY2026E EPS 10.06 and FY2027E EPS 11.76 off the 6.87 reported base. Multiple de-rates modestly from 59.1x FY2027E today to 63x on the then-forward year as the growth rate steps down from 46% to 17%: 11.76 x 63 = 741, +6.7%, rounded to +7. This lands within 0.2% of the street's 742.65 consensus target, which is a corroboration of the arithmetic, not the source of it. NOTE the FY2026 revenue dispersion of 0.1% across 14 analysts is implausibly tight and is recorded as a data-quality gap; FY2027 dispersion of 14.1% is the honest measure of disagreement and is why the spread below is wide. BULL = growth MATERIALLY FASTER than the market expects: FY2027 revenue growth of ~24% versus consensus 15.2%, driven by evidence the street has not modelled — CCGT/JV scaling beyond the single NiSource award (PWR/PWR_2025Q4.json:16), MEP growing faster than the group (PWR/PWR_2026Q2.json:45), underground margins inflecting toward th

not built — options-implied distribution on the same axis, with the KL number and where the disagreement sits
needs: desk-options implied bands (v1.4 Part 4a) — activates at 60% coverage; currently 33%

Visibility how far ahead this name can be seen

not built — visibility score
needs: visibility_score.py

Trust how much weight this name's own record can carry (credibility Z)

not built — credibility
needs: credibility.py

Sizing survival-first size (half-Kelly) vs the enforced ladder

kelly not computable — no data

Falsifiers what would prove this wrong

Consensus FY2026 revenue growth of +34.0% is delivered, i.e. FY2026 revenue lands at or above ~USD 38bn
Margin progression toward the stated 10.5-11% segment framework continues rather than being consumed by headcount growth
The CCGT / joint-venture project class scales beyond the single NiSource award
Craft labour and supervision constraints do not force a guidance reduction
Growth is not predominantly bought — organic growth is a majority of reported growth
TAIL, held here rather than in a scenario leg: a return to the 52-week low of 363 (-48%)

Conflicts surfaced, never netted out (T4)

My expected return of +7.1% is within 0.2pp of the street's +6.9% implied to a USD 742.65 target (Buy, 36 analysts, 26 buy / 10 hold / 0 sell), so on LEVEL there is no disagreement to report — my base leg uses consensus	medium
The revision tape is UNAVAILABLE, not flat: estimates_query.py drift returns n_up 0, n_down 0, breadth 0.0, direction flat, median_move null and velocity 0.0 on both EPS and revenue over 365 days, with zero outliers — on	high
My p_bear of 22% is ~21pp above the pooled anchor of ~0.7% and my p_bull of 30% is ~34pp below the pooled anchor of ~64%. Both deviations are large, both are argued in deviation_defence, and both rest on ONE fact — the s	high
Internal inconsistency in the supplied data: EV/EBITDA 27.8x and EV/Sales 2.47x against FY2025 revenue of USD 28,351m both imply an enterprise value near USD 70bn, which is BELOW the USD 104.4bn market cap and would requ	medium
FY2026 revenue dispersion is reported as 0.1% across 14 analysts while FY2027 is 14.1% across 19 and FY2028 28.2%. A 0.1% spread on a year with 34% expected growth is not credible as genuine agreement and is most likely	medium
Three of the four cross-checking desks do not exist and PWR has no options chain entry, so my distribution cannot be tested against a market-priced one. The options_implied anchor is null on every row for exactly this re	high

The case against independent adversary, different model family (P6) broken

Version history from the research record

archived versions1

archived	hash	bytes
2026-08-19T08:32:01	3a3f6c870275	36,425

One version only — the archive was created today. The diff view fills in on the next re-run.

Rendered from canonical artifacts only (T1) — no terminal database exists. Every figure is read from the system's own files at render time; where an artifact is missing the panel names the dependency rather than showing an empty box. No execution path exists anywhere in this interface (T5), not even a link.