NYSE:AGX Argan, Inc.

ISIN: US04010E1091
IndustrialsEngineering & ConstructionPower EPC
NYSE · Rockville / Arlington, VA · Power-plant EPC · Founded 1961 Analysis Status: Donatien Pick
$630.32
-8.3%
12 Jul 2026 · Signal v6
DISCLAIMER: This is a quantitative framework for educational purposes only. It is not financial advice. Always do your own research and consult a licensed financial advisor before making investment decisions.

Argan, Inc.

Argan, Inc. is a power-focused engineering, procurement and construction (EPC) contractor: through its main subsidiary, Gemma Power Systems, it designs and builds large electricity-generating plants — principally high-efficiency combined-cycle natural-gas plants, alongside solar, wind and biomass projects. It sits in a specialist niche of the Industrials sector, having built roughly 15 gigawatts of capacity, and its edge is a long track record of delivering complex gas plants on time and on budget for independent power producers and utilities — a scarce capability now in heavy demand as data-centre electricity load drives a new wave of gas-plant orders. Two smaller arms round it out: an industrial fabrication and field-services business (pipe and vessel fabrication in the US Southeast) and a telecom-infrastructure unit. Financially it is unusual for a contractor — it carries essentially no debt and a very large cash and investments pile (roughly US$70 a share), so it funds projects and pays dividends from its own balance sheet rather than borrowing. For a reader: think of Argan as a debt-free, cash-rich specialist builder of gas power plants, riding the electricity-demand surge but valued by the market like a fast grower.

🚫 DO NOT BUY — Absolutely-expensive multiple (all horizons)
The clean P/E (~67×) is ~2.9× the rate-and-growth-warranted multiple (23×) and ≥1.5× the Industrials 'rich' line, with growth that is lumpy EPC off a FY25 trough — not the exceptional, proven, durable growth that would earn the premium → Do-Not-Buy Trigger 2(a) fires. Corroborated (Analyst Override): a June-2026 cluster of discretionary, open-market insider sales (~$50M verified — chairman ~$36.9M, CEO ~$7.84M, several directors ~$5M, NOT 10b5-1 pre-planned) into the 52-week high, alongside a sequential backlog dip and consensus targets below the current price. This overrides the strong business quality: a great company can be a poor purchase at 2.9× its warranted value. The prohibition is valuation-based and horizon-agnostic. The Donatien-Pick tag pins the name to the watchlist; it does NOT bias the score toward BUY.
HorizonSignalComposite ScoreConfidenceKey Driver
Short-term (1–3 mo)DO NOT BUY4050%~2.9x warranted + weak tape
Medium-term (6–12 mo)DO NOT BUY4255%great business, structurally wrong price
Long-term (3–5 yr)DO NOT BUY4655%quality can't offset 2.9x-warranted price
Next update: 2026-09-04 — Q2 FY2027 earnings ~early Sep +1d (14-day default re-runs sooner if a catalyst lands)
Table of Contents
1Five-Pillar Scorecard2Hard Gates & Do-Not-Buy Status3Pillar Detail: Business Quality4Pillar Detail: Valuation Attractiveness5Pillar Detail: Underlying Drivers6Pillar Detail: Economic Alignment7Pillar Detail: Entry/Exit Timing8Economic Event Risk9Multi-Timeframe Technical Analysis10Price Chart (6-Month Daily)11Scenario Summary12Entry / Exit Rules13Position Sizing Context14Calibration Snapshot15Data Sources & Methodology
1

Five-Pillar Scorecard

Five independent scores — each 0–100 with its own confidence. The three fundamental pillars (Quality / Valuation / Timing) set the base BUY/HOLD/SELL via the Decision Matrix; the two context pillars (Underlying Drivers, Economic Alignment) then amplify a BUY to STRONG BUY or a SELL to STRONG SELL when both corroborate.

Business Quality

78
strong
conf 75%

Valuation Attractiveness

22
expensive
conf 78%

Entry/Exit Timing

38
weak
conf 60%

Underlying Drivers

68
Tailwind
conf 62%

Economic Alignment

66
Trend-Following
conf 65%
2

Hard Gates & Do-Not-Buy Status

Binary safety checks — any TRIGGERED gate is a hard cap regardless of the scores above; CAUTION gates are sizing notes.
Financial Distress
Zero debt, net cash ~US$70/share, current ratio 1.53, interest coverage N/A (no debt). No distress.
Valuation Ceiling
TRIGGERED (two arms): price $630 > highest analyst target $600; AND clean P/E ~67x ≥ 1.40× warranted 23x (ratio 2.9×) and ≥ the Industrials guardrail line (P/E ≥ 23x). Caps the signal at HOLD on every horizon regardless of momentum or backlog.
Earnings Event Risk
Next earnings ~early September 2026 (Q2 FY2027) — outside the 14-day window. No cap.
⚠️
Accounting / Dilution
CAUTION note (does not trip Gate 4's 30% bar): ~17-20% of net income is interest income on the cash pile (17.3% on the EPS strip, 19.8% on net income) — fully disclosed, but it inflates reported EPS. Scored on clean/operating earnings (§7b). Share count creeps ~1-2%/yr (options), not dilutive-red-flag territory.
Binary / Regulatory Event
No pending binary regulatory event. EPC contract cancellation risk is a business risk, not a binary gate.
🚫 DNB Trigger 2(a): Absolutely-expensive multiple
FIRES (hard DO NOT BUY): clean P/E ~67× ≥ 2.0× warranted (23×) at 2.9×, and ≥1.5× the Industrials guardrail line (34.5×) — the numeric bar is met three ways. The 'exceptional, proven, durable growth' exemption does NOT apply: Argan's growth is lumpy EPC off a FY25 trough, not a durable compounder profile. Overrides even a HOLD to DO NOT BUY. (Arm (b)/AI-cohort systemic leg explicitly NOT invoked — arm (a) is standalone deep-expensive.)
🚫 DNB Trigger 4 (Analyst Override): Insider-selling cluster
CORROBORATING OVERRIDE (not a clean 3-of-3 mechanical fire — flagged honestly): ~$50M (Form-4-verified) of June-2026 insider sales, verified as discretionary open-market (Form 4 aff10b5One=0), NOT pre-planned 10b5-1 — chairman Griffin (50,000 sh via a GRAT, >25% of his holdings, ~$36.9M), CEO Watson (11,873 sh, ~$7.84M) and multiple directors (~$5M), near the 52-week high. Confirmed strictly: only the non-executive chairman's >25%-of-holdings sale is fully verified; the CFO's sale was a trivial 760-sh exercise-and-sell; the CEO's %-of-holdings is unconfirmed. Some aggregators cite a larger ~$123M aggregate, but only the ~$50M above is Form-4-verified. So this does not cleanly meet the '3+ C-suite each >25%' letter, but the coordinated discretionary selling into strength is a genuine caution. Treated as an Analyst-Override corroboration of the Trigger-2(a) DO NOT BUY, not the primary basis.

Why the signal is DO NOT BUY

The primary basis is Do-Not-Buy Trigger 2(a) — an absolutely-expensive multiple: the clean P/E (~67×) is ~2.9× the rate-and-growth-warranted multiple (23×) and ≥1.5× the Industrials 'rich' line, and the growth (lumpy EPC off a FY25 trough) is not the exceptional/proven/durable kind that earns such a premium — so the trigger fires and overrides even a HOLD. The Valuation Ceiling gate independently caps at HOLD (price above the Street's high target; clean multiple ≥1.40× warranted). And a discretionary insider-selling cluster (~$50M (Form-4-verified), not 10b5-1, into the 52-week high) corroborates as an Analyst Override. Together these override the strong backlog, net-cash balance sheet and supportive Industrials tape: a great company can still be a poor purchase at 2.9× its warranted value. The prohibition is valuation-based, so it applies on all three horizons. The Donatien-Pick tag pins the name to the watchlist; it does not bias the score toward BUY.
3

Pillar Detail: Business Quality

A deep dive into the Quality score: business economics, moat, ROIC and the industry benchmark.
Business Quality — Pillar Score
High-return, debt-free specialist — but a lumpy, moat-light contractor
78
conf 75%
Business Quality
Mature Industrial EPC, growth-inflecting · conf 75%
78

Lifecycle & sector: Industrials → Engineering & Construction (power EPC). Lifecycle = Mature but growth-inflecting — a 60-year-old contractor whose revenue re-accelerated hard (TTM revenue +30%; Q1 FY2027 +50% YoY) as data-centre power demand pulled forward gas-plant orders. Metric focus: ROIC vs WACC, operating margin, backlog growth, asset turnover, FCF conversion — the Industrials profile, with EPC revenue-recognition lumpiness flagged throughout.

Sub-signalValueBenchmarkScoreRead
Revenue trajectoryTTM ~$1.04B, +30%; Q1 FY27 +50% YoYIndustrials median ~5-8%88Top-decile growth — but off a FY25 trough and driven by a handful of large gas-plant awards; not smooth organic.
Operating margin~15.0% (TTM); Q1 FY27 15.6%Industrials 10-15% healthy, >15% strong82Margins expanded from ~4% (FY25 Q1) as fixed-price contracts matured favourably. Strong, but EPC margins swing with project mix.
Cash generationFCF/OCF 0.99; OCF/sales ~47%>5% FCF margin healthy85Asset-light — capex <1% of revenue. Cash conversion helped by customer advances on projects (working-capital float).
Balance-sheet healthZero debt; cash+inv ~$70/sh; current ratio 1.53Debt/EBITDA <2.0× strong95Best-in-class. Net cash roughly a third of market cap. No refinancing risk in any rate regime.
Backlog~$2.77B (Q1 FY27), up sharply YoYRecord YoY; note a modest sequential dip (~$2.9B Jan → ~$2.8B Apr)84Record backlog, gas-plant + data-centre-power heavy. Visibility strong — but backlog is cancellable and revenue timing is lumpy.

Industry benchmark: ROIC vs WACC + backlog growth (Industrials)

ROIC comfortably exceeds WACC (with net cash, invested capital is small and returns on the operating business are very high — effectively 30%+ on operating capital), and backlog is growing double-digits. On the framework's Industrials scoring (ROIC > WACC + growing backlog → 85-100) this reads 90/100. Caveat: the denominator is flattered by the cash pile — return on total equity is more modest because ~$1B sits in T-bills earning ~3.5%.
Pricing power
55
Some — scarce gas-EPC capability lets Argan be selective, but contracts are competitively bid and fixed-price.
Network effects
50
N/A for a contractor (neutral).
Switching costs
52
Repeat IPP/utility relationships and a delivery track record create stickiness, but each project is re-bid — trimmed from the competitive read below.
Cost advantage
55
Execution reputation and self-perform capability, not structural scale — larger rivals (PWR, MTZ) out-scale it. Trimmed below.
Intangible assets
58
Gemma Power's brand and 15GW built-capacity reference list is a real, if narrow, intangible in gas EPC.

Moat average ≈ 54/100 — a competent, well-regarded specialist, not a wide-moat compounder. The Quality score leans on returns and the balance sheet, not on a durable moat.

Competitive Environment

Argan competes for large power-EPC awards against much bigger diversified rivals and utility in-house engineering. Its niche strength is combined-cycle gas plants, where it is a recognised go-to — but the data-centre-power land-grab is pulling well-capitalised competitors into the same lane.
CompetitorThreat typeRel. sizeShare trajectoryMoat-erosion vector
Quanta Services (PWR)Direct EPC / infrastructure~15× Argan revenueArgan holding in gas nicheScale, breadth, labour access; can bundle T&D + generation.
MasTec (MTZ)Direct EPC; moving into DC power~13× Argan revenueRival gaining in data-centre electrical~$1.65B Superior Group acquisition adds data-centre electrical — encroaches on Argan's new growth lane.
Fluor (FLR)Large diversified EPC~10×+ revenueStable — different project scaleCompetes on mega-projects; less focused on Argan's mid-size gas plants.
Bechtel / Kiewit (private)Direct EPCMuch largerStableCompete for the largest awards; Argan wins on focus and price on mid-size work.
Utility in-house EPCVertical substitutionStableSome utilities self-perform; caps Argan's addressable pool.

Net effect on the moat → Switching Costs trimmed to 52 and Cost Advantage to 55: Argan holds its gas-EPC niche today, but MTZ's data-centre-electrical push and PWR's scale mean the new growth lane (data-centre power) is contested from day one. Competitive threat level: moderate.

ROIC & capital allocation

Operating ROIC is very high (asset-light, self-funded; ~$973.6M cash & investments at Apr-2026, zero debt). Capital allocation is conservative and shareholder-friendly: the regular dividend was raised to $0.50/quarter ($2.00/yr run-rate) from $0.375, and Argan has a longer-run history of special dividends given its cash-rich, debt-free balance sheet (the specials pre-date the confirmed 2021-26 data window, so treat that as background not a near-term catalyst). Management skin-in-the-game is only moderate, and it deteriorated in mid-2026: a cluster of discretionary open-market insider sales (~$50M verified — chairman Griffin ~$36.9M (50,000 sh via a GRAT, >25% of his holdings), CEO Watson ~$7.84M (11,873 sh), plus several directors ~$5M; all discretionary open-market, verified NOT 10b5-1) into the 52-week high (see §7 sentiment and the §2 Do-Not-Buy check). The balance-sheet conservatism is a genuine quality positive; the insider selling and a low-productivity ~$1B cash pile (depressing return on total equity) are the offsets.
4

Pillar Detail: Valuation Attractiveness

Sector-appropriate multiples, FCF yield, reverse-DCF implied growth, embedded optionality, and the analyst-consensus cross-check.
Valuation Attractiveness — Pillar Score
Expensive — ~2.9× the warranted multiple; price above the Street's high target
22
conf 78%
Valuation Attractiveness
Warranted-multiple anchor + guardrail floor · conf 78%
22

The anchor — warranted-multiple valuation (rate + disciplined growth + sector)

Discount rate r = 9.0% (risk-free = UST10Y 4.54% [macro 2026-07-09] + 4.5% ERP + 0.0% add-on, Business-Quality ≥ 65). g_near = 10% (Industrials sector-achievable cap; consensus fwd growth haircut 25% would give more, but the cap binds — EPC growth is not durably >15%). g_term = 3%. Two-stage → WarrantedPE ≈ 23× (equal to the Industrials guardrail line, which also caps it).

Actual clean multiple: reported TTM diluted EPS $11.36 → reported P/E 55.5×. Stripping ~$32M of interest income on the cash pile (§7b) → clean TTM EPS ~$9.39 → clean P/E ~67×. Score = actual ÷ warranted = 67 ÷ 23 = 2.9× → deep in the Expensive band (<40). On the reported multiple it is still 2.4×. Either way, Expensive.
LensReadingInterpretation
Warranted-multiple anchor (40%)Clean P/E 67× vs warranted 23× = 2.9×Expensive — the dominant lens
Sector median P/E (20%)Industrials E&C peers ~18-24× fwd; AGX ~52× fwd (FY27)Roughly 2× the peer group
Own-history decile (15%)Historically a mid-teens-to-20s P/E name; now 50×+Decile 10 — above its own historical range
PEG / growth-adjusted (10%)Fwd PEG ~1.85 (FMP); trailing PEG ~0.94 flatters on the trough-base growthRich once the base-effect growth normalises
Analyst consensus (15%)Consensus $559 (median $559), high $600, low $518 — price $630 is above all of them~11% above consensus; above the high

FCF yield & implied growth

FCF yield ≈ 2.9% (P/FCF ~18× on an enterprise basis that nets the cash) — note the headline P/FCF is flattered because ~$1B of the market cap is cash, not operating value. Backing that out, you are paying ~50×+ for the operating EPC business. Implied growth: at $630 the market is pricing in something close to ~20%+ sustained EPS growth for 5 years; our disciplined estimate is ~10% (with real lumpiness risk) → the price embeds materially more growth than an EPC contractor durably delivers.

Embedded optionality / free upside

Net cash (~$70/share, ~US$1B). This is real but already in the price — not un-priced upside — so it does not tilt the score up; it is a downside cushion, not a call option. Data-centre-power backlog conversion beyond what's in numbers is the genuine option, but consensus already extrapolates it aggressively (FY30 EPS ~$17), so little is “free.” Special-dividend history gives modest optionality on the cash return. Net: the core operating business justifies perhaps ~$220-280 of the $630 price on our warranted multiple; the rest is priced-in growth expectation, not free optionality. No upward tilt applied — the optionality here is a reason to keep watching, not evidence the stock is cheap.

Analyst targets, grades & FMP cross-reference

Consensus target $559 (high $600 / low $518), coverage thin (~1 recent, 7 last year) — price is above the sell-side broker high ($600, Lake Street; some aggregator model estimates run higher). Grades: Buy consensus (4 Buy / 3 Hold, 0 Sell) but with >40% holds — mixed. FMP rating B (overall 3): ROE 5/5, ROA 5/5 (quality confirmed) but P/E 1/5 and P/B 1/5 (valuation flagged) — the sub-scores agree exactly with our read: excellent business, expensive price.
5

Pillar Detail: Underlying Drivers

The dominant external force the stock is tethered to, scored 0–100. A context pillar: it does not change the base signal — it feeds amplification (tailwind ≥65 can lift BUY→STRONG BUY; headwind ≤35 can push SELL→STRONG SELL).
Primary Driver
Data-centre power buildout + gas-plant capex cycle
68
Tailwind (65-79)

Primary driver: the power-generation / data-centre-power buildout and the natural-gas-plant capex cycle. AI and data-centre electricity demand has revived orders for firm, dispatchable generation — principally combined-cycle gas plants, Argan's core competency — after a decade of gas-EPC drought. This is the force above Argan's own execution: it fills the backlog and sets the multiple the market is willing to pay.

HorizonReadAssessment
Historical (25%)Gas-EPC awards troughed 2020-23, inflected sharply 2024-26 as data-centre load forecasts surgedImproving → strong
Current (50%)Record ~$2.77B backlog; utilities/IPPs announcing new gas builds; grid-interconnection queues favour firm powerFavourable
Forward (25%)Multi-year data-centre power demand thesis intact; risk = capex digestion, cancellations, or a demand-forecast resetPositive but crowding + cancellation risk

Driver score 68 → Tailwind (amplification-eligible). Note: this is not a commodity-price driver, so the Step-2b commodity price-trend overlay does not apply — there is no falling spot to cap the short-term leg. However, the tailwind is not applied: the base fundamental signal is HOLD (Expensive valuation) and HOLD never amplifies; more decisively, a Do-Not-Buy trigger (2a) has overridden the signal to DO NOT BUY on every horizon. A driver tailwind cannot rescue a name that is prohibited on price — a good tailwind at 2.9× warranted value is still a bad purchase. The driver's value here is that it defines the thesis: if the data-centre-power/gas-capex cycle rolls over (a demand-forecast reset or backlog cancellations — the live §11 bear trigger), even the DO-NOT-BUY floor gets worse. Confidence 62% — the data-centre-power thesis is real but consensus-crowded and one demand-forecast reset from a sharp de-rating.

6

Pillar Detail: Economic Alignment

How the current economic climate sits relative to this stock, read from the latest Macro-Economic report. Classifies the macro pressure (Tailwind / Neutral / Headwind) — the second amplification input — and frames a long entry as Trend-Following or Contrarian with a 0–100 conviction.
Stance · Pressure
Trend-Following · Tailwind
66
conviction

AGX is not on the macro watchlist, so Economic Alignment maps via its GICS sector, Industrials → XLI, in the 2026-07-09 macro report's Driver-Sector Impact Matrix: XLI Short O / Medium O / Long SO (Overperform strengthening to Strong-Overperform long). The regime is Higher-for-Longer / Stagflation-lite — supportive for capital-goods/infrastructure names tied to the electrification + reshoring capex cycle. Anchoring on the Medium horizon: pressure = Tailwind, stance = Trend-Following, conviction 66. Because the base signal is HOLD (Expensive) and a Do-Not-Buy trigger has overridden it to DO NOT BUY, this Tailwind pressure did NOT enable any amplification and does NOT offset the prohibition — a supportive economy cannot make a 2.9×-warranted price a buy.

Source: sector-map (XLI — not a macro-watchlist name) · Macro report 2026-07-09

7

Pillar Detail: Entry/Exit Timing

The risk-reward framework, relative strength vs SPY and the sector ETF, the macro overlay, news-derived sentiment, and the catalyst cluster.
Entry/Exit Timing — Pillar Score
Weak — parabola rolling over, price below the 20/50-day EMAs, no reversal yet
38
conf 60%
Entry/Exit Timing
Risk-reward + relative strength + sentiment · conf 60%
38

Risk-reward assessment (daily focus)

Poor. Price $630 sits ~$63 (~1.2 ATR) above the nearest logical stop at ~$567 daily support, but the setup is a broken parabola, not a support bounce — the daily has a support breakdown, price is below the 20-day (~$702) and 50-day (~$675) EMAs, and MACD has flipped negative. You are not near a multi-timeframe support with a tight stop; you're mid-air after a −22% drop from $806 with the next real shelf the 200-DMA (~$464). Risk-reward from here is unfavourable.
Sub-signalReadingScore
MTF trend (30%)Monthly/weekly uptrend, daily/hourly/15m breakdown — mixed, rolling over52
Risk-reward (20%)Broken daily; ~1.2 ATR to stop but no bounce setup; downside to 200-DMA open30
Relative strengthMassive 12-month outperformance (197→806), but rolling over vs SPY/XLI over the last 2 weeks45
52-week position~78% of the 52wk range ($197-806) — extended, well off lows

Macro overlay (medium sensitivity, 15% weight)

Regime Higher-for-Longer / Stagflation-lite; UST10Y 4.54%. Rate direction is the key overlay for a high-multiple name — a hawkish tape is a headwind to the discount rate. XLI sector is in favour (rotation supportive), which partly offsets. Net: neutral-to-slightly-unfavourable near-term.

Sentiment layer

Analyst grades: mixed — JP Morgan upgraded to Overweight (Mar-2026), but Lake Street sits at Hold (maintained Jun-2026); consensus 4 Buy / 3 Hold, and price is above every target. No net upgrade momentum in the last 30 days. News tone: recent coverage is explicitly cautious — Simply Wall St ('above fair value', '~8.7% overvalued'), Zacks ('trades at a premium — buy or wait?'). Insider signal (bearish): the ~$50M (Form-4-verified) June-2026 discretionary open-market selling cluster (chairman, CEO, directors — NOT 10b5-1) into the 52-week high is a live negative-sentiment input and the corroborating leg of the §2 Do-Not-Buy. Net sentiment: negative.

Catalyst layer

Clustering score ~55 (focused). Near-term: CPI 2026-07-14 (macro, medium relevance via the discount rate) and FOMC ~late July. Stock-specific: Q2 FY2027 earnings ~early September — the next real mover (backlog trend, gas-plant awards, margin durability). No cluster of stock-specific events inside 14 days; the tape, not a catalyst, dominates near-term.
8

Economic Event Risk

High-impact macro releases in the next 14 days that could swing this stock, plus the last 7 days of surprises.

Upcoming events (next 30 days)

DateEventImpactForecastPreviousRelevant?Why
2026-07-14CPI YoY / Core CPI (Jun)High3.9% / 2.9%4.2% / 2.9%⚠ MediumIndustrials are medium macro-sensitivity; a hot CPI (rates up) pressures long-duration/high-multiple names like AGX via the discount rate.
~2026-07-29FOMC Rate DecisionHighHoldHold⚠ MediumRate path sets the discount rate in the warranted-multiple anchor; a hawkish hold tightens the ceiling for richly-valued Industrials.
~2026-09-04AGX Q2 FY2027 earningsHigh✅ YesThe next stock-specific catalyst — backlog trend, gas-plant awards, margin durability, cash-return signals. Sets the next-update date.

Recent surprises (last 7 days)

DateEventActualForecastSurpriseImpact
2026-07-08FOMC MinutesHigher-for-longer tone reinforced — headwind for high-multiple names
2026-07-09Existing Home Sales MoM (Jun)-2.4%0.7%BelowSofter activity; minor for AGX (not housing-linked)

AGX carries medium macro sensitivity, so no high-impact release inside 3 days triggers a WAIT-override. The one to watch is CPI on 2026-07-14: because the whole HOLD case rests on an expensive discount-rate-sensitive multiple, a hot inflation print (10Y up) is a direct near-term de-rating risk. The next stock-moving event is Q2 FY2027 earnings (~early September), which anchors the next update.

9

Multi-Timeframe Technical Analysis

Trend, RSI and breakout status across monthly / weekly / daily / hourly / 15-minute, with a confluence verdict.
TimeframeTrendDirectionRSIMACDKey S/RBreakoutVol
MonthlyUptrend ↑Bullish69.6+, risingS: 191 R: (blue-sky)Resistance breakout0.24×
WeeklyUptrend ↑Bullish55.2+, flatteningS: 197/276 R: 748/779Resistance breakout0.71×
DailyWeakening →Neutral40.5−, fallingS: 567/576/607 R: 742/779/806Support breakdown0.89×
HourlyStrong downtrend ↓Bearish35.6−, fallingS: 630/633 R: 675/705Support breakdown
15-minDowntrend ↓Bearish39.8−, basingS: 630/636 R: 691/705Support breakdown
Confluence: Mixed / rolling over from the top · MTF Score 52

A textbook higher-timeframe uptrend with a sharp lower-timeframe rollover. Monthly and weekly remain in blow-off uptrends (the 197→806 run), but the daily has broken support, printed a lower high, and momentum (MACD) has flipped negative — price is down ~22% from the $806 high to $630, sitting below the 20/50-day EMAs (~702/675) and back toward the 200-day (~464). RSI 40 on the daily is neither oversold nor turning up. This is distribution off a parabolic peak, not a clean pullback-in-uptrend buy — the confluence is mixed and deteriorating. Key level: daily support ~$567-607; a loss of $567 opens air down toward the 200-DMA.

10

Price Chart (6-Month Daily)

A 6-month daily close line with SMA50 and key support/resistance — the visual companion to the MTF table.

AGX daily, ~6 weeks. The parabolic run to $806 (late June) has rolled over ~22% to $630, breaking daily support and the 20/50-day EMAs. Consensus target $559 sits below the current price; the 200-DMA (~$464) is the deeper support shelf.

11

Scenario Summary

Bull / Base / Bear 12-month price paths with triggers and probability weights.

Bull $760 (25%)

Re-rating resumes: another wave of gas-plant / data-centre-power awards pushes backlog well past $3B, Q2 beats and management signals margin durability, and the market keeps paying a growth multiple. Price reclaims the $740-806 zone. Requires the growth to keep out-running the valuation — possible while the data-centre-power narrative is hot, but this is paying up at a 50×+ operating multiple.

Base $540 (55%)

The parabola digests. Growth stays strong but decelerates off the base-effect, and the multiple compresses toward — though still above — the Street's $559 consensus as the market re-anchors a lumpy EPC contractor nearer 30-40× operating earnings. Net cash and the dividend cushion the downside. A drift/consolidation in the ~$500-590 band over 12 months. This is the probability-weighted centre of gravity and it sits below today's $630 — consistent with the Expensive/HOLD verdict.

Bear $360 (20%)

The de-rating bites: a backlog cancellation or a data-centre-power capex/demand-forecast reset (the live §5 driver risk), a soft quarter, or a broad high-multiple-Industrials unwind compresses the operating multiple toward the warranted ~23×. Even on healthy FY27 EPS that maps toward the mid-$300s (near our ~$220-280 pure-warranted fair value plus the cash cushion). Mean reversion from 50×+ is a 40%+ move, not a wobble.

Probability-weighted 12-month fair value ≈ $559 (0.25×760 + 0.55×540 + 0.20×360 = 559) — roughly 11% below the current $630, and coincidentally right on the Street's $559 consensus. A negative expected 12-month return at today's price is exactly why the signal is DO NOT BUY, not BUY, despite the strong business.

12

Entry / Exit Rules

Three independent entry paths (Fundamental · Technical · Catalyst) and three exit triggers (Stop-Loss · Thesis · Profit-Target). Any one entry path is a valid entry — the more that agree, the larger the position the conviction ladder suggests. Exits are graded by severity, not count.

How to read this — the Conviction Ladder

The three entry groups are alternative paths to a buy, not a checklist. A group counts only when all its sub-conditions hold. How many groups are satisfied sets the suggested size — it does not gate whether you may enter: 1 group = Half-Size (a valid starter/scale-in), 2 = Full-Size, 3 = Over-Size (highest conviction); 0 = Wait (no path open yet). A strong overall signal can still read Wait here when the stock is well above its entry zones — that flags "good business, no entry edge right now," not a contradiction. Exits are graded by severity of what is live, not by a count: a hard stop is an Exit on its own.
Entry conviction: Wait0 of 3 groups met — no entry path open

Fundamental — not MET

Price is far above fair value — the cheap-entry path is not open.
⛔ Price $630 < fair-value estimate ~$540 (base) / ~$280 (warranted core)
✅ No earnings within 7 days
✅ Underlying-Driver score ≥ 50 (68)

Technical — not MET

Daily has broken support and sits below the 20/50-day EMAs; no reclaim, no tested higher-low bounce yet.
⛔ Daily close > SMA50 (~$690) on >1.5× volume
⛔ OR a tested bounce off $567-607 support with a higher low
✅ RSI 35-65 (40.5)

Catalyst — not MET

No event in the 7-day window; next catalyst is Q2 earnings in ~September.
· Post-earnings move >+5% with guidance/backlog raised
· Volume > 2× the 20-day average on the move

Forecast: All three entry groups are UNMET → Conviction Ladder reads Wait (0/3). The Fundamental path is unlikely to open near current prices — fair value (~$540 base, ~$280 on the warranted core) sits 15-55% below $630, so it would need a material pullback (Low confidence, requires a de-rating). The Technical path could open on a Moderate horizon (~2-4 weeks) IF price either reclaims the ~$690 50-day on volume or carves a tested higher low off $567-607 support — watch the $567 shelf. The Catalyst path is catalyst-dependent on Q2 FY2027 earnings (~early September): a beat with a raised backlog on >2× volume would open it. Net: no entry edge at $630 — wait for a lower price or a confirmed technical reset.

Exit action: Holdno exit trigger is live — hold the position

Stop-Loss — not LIVE

⛔ Two daily closes below $567 (daily support shelf)

Thesis Invalidation — not LIVE

⛔ Backlog declines 2+ consecutive quarters / a major gas-plant award is cancelled
⛔ Data-centre-power / gas-capex driver turns to a headwind (demand-forecast reset)
⛔ A named rival (MTZ/PWR) visibly takes data-centre-power share Argan was expected to win

Profit-Target — not LIVE

⛔ Price into $760+ (bull) with RSI > 70

Forecast: No exit trigger is live for a holder today (this is a HOLD, not a SELL). The hard stop at $567 is ~10% below the current price and could trigger on a Moderate horizon given the broken daily and negative momentum — the single most important level to watch. Thesis-invalidation is the slower risk: a backlog/awards reset or a data-centre-power demand-forecast cut would flip this from HOLD toward SELL. Profit-target trim only re-arms if the parabola resumes to $760+ overbought.

Imagine you act at the current price of $630.32 · as of 12 Jul 2026

What if you bought now?

You are risking ~$270 (43%) to the warranted-core downside to gain ~$130 (21%) to the bull case — a poor risk-reward.

What you're risking: the hard stop sits ~$567 (−10%); the base case (~$540) is below today's price (−5% expected drift); the bear (~$360) is −43%. You'd be buying into a broken daily tape (support breakdown, below the 20/50-day EMAs), above the Street's highest target ($600), on a 50×+ operating multiple — no entry group is met. CPI (Jul 14) and Q2 earnings (Sep) add path risk.

What you're gaining: immediate exposure to the bull re-rating (~+21% to $760), a ~0.3% dividend plus special-dividend optionality, and a net-cash balance sheet that cushions the fall. But the probability-weighted 12-month path (~$559) is negative from here.

Read: acting now is not worth it — waiting for the ~$567 support test or a lower entry materially improves the deal. This is a watch, not a buy, at $630.

What if you sold now?

You are protecting ~$270 of downside to gain little — but no exit rule is mechanically triggered, so this is a hold/trim-into-strength zone, not a forced sell.

What you're giving up: the bull re-rating to ~$760 (+21%), the dividend/optionality, and any further momentum while monthly/weekly remain in uptrend.

What you're protecting: the ~11% expected drift to the probability-weighted fair value (base case ~$540, −14%) and the ~43% bear-case drawdown if the multiple mean-reverts. No hard stop, thesis-break or profit-target is currently live — so there is no mechanical sell signal today.

Read: for an existing holder this is a HOLD with a tight watch on $567; consider trimming into any push back toward $760 rather than selling into the current dip.

13

Position Sizing Context

Illustrative portfolio math (not advice) translating conviction into an allocation given risk-per-share and volatility.

Position sizing not computed — specify your portfolio allocation and role for sizing guidance.

14

Calibration Snapshot

Machine-readable snapshot of every score, level and signal, saved alongside the HTML so the next run can compute deltas.
{
  "ticker": "AGX",
  "company": "Argan, Inc.",
  "currency": "USD",
  "date": "2026-07-12",
  "version": "v6",
  "exchange": "NYSE",
  "exchange_ticker": "NYSE:AGX",
  "isin": "US04010E1091",
  "api_ticker": "AGX",
  "analysis_status": "donatien-pick",
  "user_horizon": null,
  "user_allocation_pct": null,
  "portfolio_role": null,
  "price_at_rating": 630.32,
  "signal_short": "DO NOT BUY",
  "signal_medium": "DO NOT BUY",
  "signal_long": "DO NOT BUY",
  "primary_signal": "DO NOT BUY",
  "base_signal_pre_override": "HOLD (all horizons \u2014 High Quality 78 + Expensive Valuation 22 \u2192 'great business, wrong price'); DNB Trigger 2(a) then overrides to DO NOT BUY on every horizon",
  "short_entry_confirmed": false,
  "short_cap_reason": "Base signal is HOLD (Expensive valuation / Gate 3); no short-cap needed \u2014 no BUY to cap. DNB Trigger 2(a) then overrides HOLD to DO NOT BUY.",
  "quality_score": 78,
  "lifecycle_stage": "mature-growth-inflecting",
  "quality_detail": {
    "industry_benchmark_name": "ROIC vs WACC + backlog growth (Industrials)",
    "industry_benchmark_value": "ROIC>>WACC; backlog ~$2.77B growing",
    "industry_benchmark_score": 90,
    "moat_score": 54,
    "roic_note": "very high operating ROIC (asset-light, net cash); return on total equity diluted by ~$1B cash pile",
    "management_skin_in_game": 45
  },
  "valuation_score": 22,
  "valuation_detail": {
    "fcf_yield": 2.9,
    "implied_growth_rate": 20.0,
    "consensus_growth_rate": 10.0,
    "historical_valuation_decile": 10
  },
  "warranted_multiple": 23,
  "actual_multiple": 67,
  "val_multiple_basis": "clean P/E (TTM diluted EPS ex interest income)",
  "discount_rate_r": 0.09,
  "risk_free_10y": 0.0454,
  "g_near": 0.1,
  "g_term": 0.03,
  "warranted_ratio": 2.9,
  "val_band": "expensive",
  "nonop_pct_of_net_income": 19.8,
  "clean_pe": 67.1,
  "clean_peg": 1.85,
  "reported_pe": 55.5,
  "timing_score": 38,
  "timing_detail": {
    "mtf_confluence": 52,
    "risk_reward_score": 30,
    "relative_strength_vs_spy": "outperformed 3m (post-run) but rolling over",
    "catalyst_clustering_score": 55,
    "dynamic_macro_weight": 0.15
  },
  "driver_score": 68,
  "driver_label": "Tailwind",
  "driver_name": "Data-centre power buildout + gas-plant capex cycle",
  "driver_commodity_trend": "N/A \u2014 not a commodity-price driver",
  "economic_alignment_stance": "Trend-Following",
  "economic_alignment_conviction": 66,
  "economic_alignment_pressure": "Tailwind",
  "economic_alignment_source": "sector-map (XLI)",
  "macro_report_date": "2026-07-09",
  "competitive_share_trajectory": "stable",
  "competitive_threat_level": "moderate",
  "overall_confidence": 50,
  "fair_value_est": 540,
  "warranted_core_fair_value": 280,
  "stop_loss": 567,
  "target_price": 540,
  "scenario_base_target": 540,
  "scenario_bull_target": 760,
  "scenario_bear_target": 360,
  "prob_weighted_fair_value": 559,
  "entry_groups_met": 0,
  "entry_conviction": "Wait",
  "exit_groups_live": 0,
  "exit_action": "Hold",
  "hard_gate_state": "donotbuy",
  "gates_triggered": [
    "Valuation Ceiling"
  ],
  "gates_caution": [
    "Accounting/Dilution (interest-income earnings quality)"
  ],
  "do_not_buy_triggers": [
    "Trigger 2(a): Absolutely-expensive multiple (clean P/E ~67x = 2.9x warranted 23x; growth lumpy/non-durable)",
    "Trigger 4 (Analyst Override): Discretionary insider-selling cluster (~$50M (Form-4-verified), not 10b5-1, into 52wk high)"
  ],
  "dnb_trigger2_evaluated": "FIRED. Trigger 2(a) 'deep-expensive alone': clean P/E ~67x is 2.9x the warranted 23x (>=2.0x bar) and >=1.5x the Industrials guardrail line (34.5x) \u2014 met three ways. The firing condition ('no exceptional, proven, durable growth') holds: Argan's 50% YoY is lumpy EPC off a FY25 trough, not a durable-compounder profile that would earn the premium \u2014 so the trigger fires (the non-durability is the reason it fires, not a reason to spare it). Arm (b) AI-cohort/systemic-tail leg explicitly NOT invoked (per mandate: AGX carries an Industrial multiple, not an AI-cohort de-rating); arm (a) is standalone deep-expensive and needs no catalyst.",
  "dnb_trigger4_evaluated": "FIRED as Analyst Override (corroboration, not primary). ~$50M of Form-4-verified June-2026 insider sales, discretionary/open-market (aff10b5One=0), NOT 10b5-1 \u2014 chairman Griffin (~$36.9M, 50,000 sh via GRAT, >25% of holdings), CEO Watson (~$7.84M, 11,873 sh), multiple directors (~$5M), near the 52-week high (some aggregators cite a larger ~$50M (Form-4-verified) aggregate; only ~$50M is Form-4-verified). Strict '3+ C-suite each >25% of holdings' letter is NOT cleanly met (only the non-exec chairman's >25% sale is fully verified; CFO sold a trivial 760 sh; CEO %-of-holdings unconfirmed), so treated as a judgment-call override corroborating Trigger 2(a), with the confirmed-vs-unconfirmed specifics stated openly.",
  "analyst_consensus_target": 559,
  "analyst_target_high": 600,
  "analyst_target_low": 518,
  "analyst_target_upside_pct": -11.3,
  "analyst_grades_consensus": "Buy (4 Buy / 3 Hold / 0 Sell)",
  "analyst_bullish_pct": 57,
  "analyst_coverage_count": 7,
  "fmp_rating": "B",
  "fmp_overall_score": 3,
  "recent_upgrades_30d": 0,
  "recent_downgrades_30d": 0,
  "next_update_date": "2026-09-04",
  "next_update_basis": "Q2 FY2027 earnings ~early Sep +1d (14-day default re-runs sooner if a catalyst lands)"
}
15

Data Sources & Methodology

Audit trail of every data source: fully available (✓), fallback (⚠), or failed (✗), plus provenance-based confidence haircuts.
Data Source Status
get_company_profile identity, sector, price, cash-per-share, dividend
get_income_statement (10q) revenue/margins/EPS + interest-income decomposition (§7b)
get_financial_ratios P/E, P/B, FCF yield, cash/share, debt (zero)
get_multi_timeframe_analysis 5-timeframe trend/RSI/MACD/breakout
get_technical_indicators daily RSI/MACD/EMA/SMA/ATR series
get_price_target_consensus / _summary consensus $559, high $600, low $518, thin coverage
get_stock_grades / get_grades_consensus 4 Buy / 3 Hold; recent JPM upgrade, Lake Street Hold
get_analyst_estimates FY27-31 revenue/EPS consensus
get_ratings_snapshot FMP B; ROE/ROA 5/5, P/E & P/B 1/5
get_earnings_calendar returned empty for AGX; next earnings ~early Sep inferred from filing cadence
get_risk_factors returned empty text — used SEC business description + web research instead
get_related_tickers errored — competitors sourced via web research (PWR/MTZ/FLR)
web research (backlog, insider selling, competitors) Q1 FY27 backlog ~$2.77B, +50% rev; insider-selling character; MTZ Superior deal
Impact on scores: Core financial, valuation, technical and analyst data are all fully available (high confidence on Quality/Valuation). Confidence haircuts: Timing −5 (earnings-calendar partial; date inferred), Drivers −10 (forward data-centre-power outlook is consensus-crowded and forecast-reliability moderate). The Valuation verdict is robust — it holds on both the reported (2.4×) and clean (2.9×) multiple and on the price-above-high-target arm independently.
DISCLAIMER: This is a quantitative framework for educational purposes only. It is not financial advice. Always do your own research and consult a licensed financial advisor before making investment decisions.