Super Micro Computer — Blitz Index 43 (C)

The margin is the thesis — and the margin collapsed.

Last re-scored: October 15, 2026.

Score breakdown

Think we scored it wrong? Score it yourself →

Radar axes (0–100): Moat 48 · Speed 76 · Creative 67.

Watch item

⚠️ BlitzFailing watch (D −15 = the line). Gates: G1 ✓ · G2 ✓ estimated. 22 points below the threshold — first score. The business case and the trust case are two different scores; the model only scores the business.

Strengths

• $60B+ order backlog
• FY2027 guidance $65–72B
• Accounting review resolved without restatement

Concerns

• Margin collapse guided 10.4–10.8%
• Receivables +178% · inventory +175%
• Bank debt 36× in a year
• Family supply-chain concentration

Research notes

Thesis: a second explosion, capped by control

Super Micro Computer, Inc. (NASDAQ: SMCI) scores 43 out of 100, grade C, on BlitzLab v1.0, twenty-two points below the 65-point line. The arithmetic is exact and worth stating at the outset, because the shape of the score explains the company better than the total alone: speed 19 out of 25, moat 29 out of 60, creative destruction 10 out of 15, and deductions of 15 points. Added together, 19 plus 29 plus 10 minus 15 equals 43. The business subtotal before deductions is 58, which is itself below the 65-point line, so the deductions are not the sole reason the company misses the line; the business sections are. But the deductions deserve the first sentence of any honest account of this score, because they sit exactly on a line the model draws in the open: deductions of exactly 15 points place the company exactly at the FLAG watch threshold for blitzfailing surveillance. This body states that status plainly and does not soften it. The score is a first scoring, not a rescore, and its as-of date, T, is October 15, 2026. The latest results available at T were the fiscal 2026 Form 10-K, for the fiscal year ended June 30, 2026 and filed August 31, 2026. Fiscal 2027 first-quarter results, expected November 3, 2026, had not been published at T and are excluded from the score; they appear in this body only as a future marker, with no anticipated outcome attached. The fiscal year ends June 30, and the share history includes a 10-for-1 split on September 30, 2024, with figures on an adjusted basis.

The gate verdicts frame the whole case. Gate 1, which asks whether growth continues or a second growth curve exists, passes on confirmed evidence. Fiscal 2026 revenue was $39,063 million, up 77.8 percent year over year, confirmed by the Form 10-K, and fourth-quarter fiscal 2026 revenue was $11,120 million, up 93.2 percent year over year, derived as annual results minus the first three quarters. Fiscal 2027 guidance of $65–72 billion, up 66 to 84 percent year over year, is confirmed by a Form 8-K with Exhibit 99.1. New orders above $60 billion and a record backlog are confirmed by the same Form 8-K. The file reads the trajectory as a cycle rather than a single surge: a first explosion in fiscal 2024, up 110 percent, then a slowdown in fiscal 2025, up 47 percent, then a second explosion in fiscal 2026, up 78 percent, with the second curve still in progress. Gate 2, which asks whether the ten-year ceiling of the new market is at least ten times current revenue, passes as an estimate. Ten times trailing twelve-month revenue of approximately $39.1 billion gives a bar of approximately $390.6 billion, and the estimated artificial intelligence server market of $837.8 billion from MarketsandMarkets for 2030 to $981.7 billion from Gartner for 2030 is more than twice that bar. The file attaches a qualification that this body preserves exactly: the bar itself is so high that the argument for a single vendor is about market share rather than market size, and artificial-intelligence-server-only market share is unconfirmed. Supermicro passes both gates, and then scores 43. The gates establish that a second curve and a large enough market exist. The score measures how far, and at what cost and risk, the company has travelled into them.

The score file's three key drivers make that measurement precise. The first is the speed of the second artificial intelligence server explosion: fiscal 2026 growth of 77.8 percent confirmed by the Form 10-K, with fiscal 2027 guidance of 66 to 84 percent, repeating the first-explosion, slowdown, second-explosion cycle. The second is the control environment as the ceiling on the score. The accounting issue of 2024 closed without a restatement, yet internal control over financial reporting remained ineffective at June 30, 2026, a deduction of 4 points on revenue quality; a former associate's guilty plea in an export-control case plus multiple ongoing federal investigations give a deduction of 4 points on regulatory and social costs; and concentrated family supply relationships give a deduction of 3 points on governance. Together the deductions are 15 points. The file's own summary of the structure is that the business is top-tier in speed while trust pulls the total down, a pattern it compares to another scored company described as business-strong and trust-weak. The third key driver is the thin-margin model: a gross margin of 10.8 percent, with quarterly figures swinging from 6.3 percent to 17.5 percent, confirmed by filings. The file treats this as a central test case for the hypothesis that margin dilution is the normal state of a successful blitz, while noting that the moat section scores the level of margin, not the hypothesis.

In blitzscaling terms, the file places Supermicro as a speed-plus-distribution blitzer. The main lever is speed, expressed in the positioning phrase "Speed-to-Market & AI Density": the company turns NVIDIA's newest products into racks faster than its rivals. The secondary lever is distribution, through a lean direct-plus-channel machine. The weak lever is network effects, limited by the structure of a business that ships hardware unit by unit. The file draws a contrast with neocloud companies, which it calls heavy machinery; Supermicro, by contrast, is a blitz of assembly speed. The company is placed at the start of a second blitz act, after the first act of fiscal 2024 and the adjustment of fiscal 2025, with fiscal 2027 guidance of $65–72 billion describing the size of the second act. And the file names a single largest risk from the blitzscaling perspective, in a sentence this body quotes in substance because it organises everything that follows: control has not kept up with speed. Revenue grew about 5.5 times in three years, in the file's framing, while internal control remained ineffective and an export-control indictment led to a former associate's guilty plea. That is the portrait of a company whose blitz outran its governance, and the file warns that the moment speed slows, the cost of control failures can arrive all at once.

Speed: a second explosion, with a volatile quarter inside it

Speed scores 19 out of 25, or 76.0 on the radar conversion, and it is the strongest section in percentage terms. Its four items are growth 8 out of 10, capital velocity 3 out of 5, hiring and productivity 4 out of 5, and share velocity 4 out of 5. The section describes a company moving very fast, with one visible crack in the growth record that explains why the top marks are withheld.

Growth, at 8 out of 10, rests on confirmed filing evidence of unusual strength. Fiscal 2026 revenue of $39,063 million, up 77.8 percent year over year, is confirmed by the Form 10-K. The quarterly path within that year is steep: second-quarter fiscal 2026 up 123.4 percent, third-quarter fiscal 2026 up 122.7 percent, and fourth-quarter fiscal 2026 up 93.2 percent year over year, on Form 10-Q evidence and annual-minus-quarter derivation. Fiscal 2027 guidance of $65–72 billion, up 66 to 84 percent year over year, is confirmed by the Form 8-K dated August 11, 2026. The score withholds the ninth and tenth points for reasons the file states plainly, and this body preserves them without softening. Quarterly volatility is extreme: first-quarter fiscal 2026 fell 15.5 percent year over year, with approximately $1.5 billion deferred on a change in customer mix, confirmed by the Form 10-Q. Fiscal 2025 grew 46.6 percent, a slowdown interval inside the longer cycle. And the base is now approximately $39 billion, so a negative quarter at that scale is a material event rather than a rounding error. Growth of 77.8 percent on a $39 billion base, with guidance pointing higher, is therefore scored as strong but not maximal, because the model asks not only how fast revenue grew but how steadily it grew.

Capital velocity, at 3 out of 5, is the item where access and cost pull in opposite directions, and the file scores both sides. Access is good. Convertible notes total $4,725 million: a 2029 issue of $1,725 million whose coupon was revised upward from 0 percent to 3.50 percent in February 2025, a 2028 issue of $700 million at 2.25 percent, and a 2030 issue of $2,300 million at 0 percent, all confirmed by the Form 10-K. Preferred shares of approximately $4.23 billion carry a 7.00 percent cumulative dividend with mandatory conversion in 2029, confirmed by the Form 10-K. An at-the-market equity facility of $1.25 billion was unexecuted at June 30, 2026, confirmed by the Form 10-K. The file scores access at 4 points and then subtracts 1 point for dilutiveness, because the terms of that access dilute existing equity: convertible notes, cumulative preferred shares, and an unused equity facility are funding raised against future ownership. Capital velocity of 3 therefore describes a company that can raise fuel from capital markets at scale, and that pays for that fuel in dilution. That distinction matters for the rest of the score, because the same dilutive instruments reappear in the debt deduction, where their cost is weighed again from the balance-sheet side.

Hiring and productivity, at 4 out of 5, rests on a headcount and a derived ratio that the file presents with unusual care about what each figure is. Employees numbered more than 7,000, up 12.2 percent, confirmed by the Form 10-K, including approximately 3,500 in research and development. Revenue per employee of approximately $5.58 million, calculated as fiscal 2026 revenue divided by headcount, is an estimate derived from those confirmed inputs. The file's interpretation is that this is an elite-level figure despite an assembly-type structure with material-cost pass-through, and that hiring is still increasing. The score withholds the fifth point for reasons the file leaves implicit in the structure of the item: the productivity figure is derived, not disclosed, and the business model behind it is assembly rather than proprietary manufacturing, so the ratio measures throughput efficiency more than it measures a protected advantage. A reader should therefore take the $5.58 million figure as the file presents it, as arithmetic on confirmed inputs, and not as a company-disclosed productivity claim.

Share velocity, at 4 out of 5, rests on press-reported market evidence that the file labels as such throughout. IDC ranked Supermicro second in the global server market in the second quarter of 2026 with a 6.1 percent share, up 97.2 percent year over year, behind Dell at 13.4 percent, on a press-reported basis through trade-press citations of IDC. In the fourth quarter of 2025 the share was approximately 9 percent or more, also second place, on a press-reported basis. The score withholds the fifth point for the reason the file states: artificial-intelligence-server-only market share is unconfirmed, so the overall server ranking cannot be converted into a measured share of the specific market that drives the growth story. Share velocity of 4 therefore describes a company demonstrably gaining ground in the broader server market at a very fast rate, with the precise share of the artificial intelligence segment still unmeasured at T.

Moat: a large market and a lean machine, without a network effect

Moat scores 29 out of 60, or 48.3 on the radar conversion, and its four items are network effects 6 out of 18, distribution scalability 9 out of 18, market size 10 out of 12, and gross margin 4 out of 12. The shape of that split is the analytical core of the Supermicro case. The company operates in a market large enough to matter, distributes through an extremely lean machine, and has no classical network effect, while its margin, which is the scorecard's proxy for pricing power, is thin. Each of those points needs to be taken separately, because the moat section is where a careless reading of Supermicro goes most wrong: the revenue growth is spectacular, and the moat behind it is partial.

Network effects, at 6 out of 18, are the weakest moat item in absolute terms, and the score file is explicit about why the item cannot score higher: there is no true network effect. A network effect in the scorecard's sense means that each additional customer makes the product more valuable to every other customer. Supermicro's business does not work that way. It ships hardware unit by unit, and one customer's rack does not make another customer's rack more valuable. What the company does have is a set of technology and switching-cost elements that the file credits at the upper end of the anchor's lower band. The DLC-2 direct liquid cooling stack is built in-house across all layers, on investor-relations evidence dated September 23, 2026, on a press-reported basis. The NVIDIA GB300 backlog exceeds $13 billion, described on a press-reported basis as the largest single contract in the company's 32-year history. And graphics processing unit rack validation and integration cycles create switching costs, on earnings-call evidence cited on a press-reported basis. The score of 6 therefore rewards technology differentiation plus weak switching costs for what they are, and refuses to credit them as something they are not. That discipline matters, because it prevents the strongest feature of the technology story, the liquid cooling stack, from being double-counted as both a moat and a creative-destruction claim. The file also records a pre-T technology milestone that belongs here: shipments of Vera Rubin NVL72 racks began on September 23, 2026, on the file's list of positive developments. That date is before T, so it is evidence at T, not a future marker, and it supports the technology-leadership reading without changing the network-effect judgement.

Distribution scalability, at 9 out of 18, is the strongest moat item in relative terms, and it is built from a channel structure confirmed by filings. Distribution runs through direct sales plus indirect channels, including distributors, value-added resellers, and original equipment manufacturers, confirmed by the Form 10-K. Sales and marketing expense was 0.90 percent of revenue in fiscal 2026, confirmed by the Form 10-K, which the file calls an extremely lean distribution machine. The Form 8-K dated August 11, 2026 confirms that hundreds of enterprise customers were added over the past year. The score withholds the upper half of the item for the reason the file states: there is no viral or platform ride, only demonstrated large-scale repeat expansion. A distribution score of 9 therefore describes a machine that demonstrably scales by adding customers one by one at very low sales cost, without the self-reinforcing distribution that a platform or viral product would provide. The $60 billion-plus in new orders and the record backlog, confirmed by the Form 8-K, belong in this item as evidence that the machine is currently converting demand into orders at scale, while the backlog's conversion into recognised revenue remains a future observation, not evidence at T.

Market size, at 10 out of 12, is the item where the Gate 2 judgement is converted into a score. The estimated artificial intelligence server market of $838–982 billion for 2030, on industry estimates, passes the gate by a wide margin, as described above. The score withholds the top two points for the reason the file states: artificial-intelligence-server-only market share is unconfirmed, and the overall server ranking at second place cannot substitute for a measured share of the specific market. The score of 10 therefore describes a market large enough to support a much larger company, with the company's precise position inside that market still unmeasured. It is worth noting what the market size score does not claim. It does not claim that Supermicro will capture any particular share of the $838–982 billion market, and it does not treat the $60 billion-plus order figure as a market share figure. The order figure is a company-level fact. The market estimate is an industry-level estimate from third parties. The score keeps those two kinds of evidence apart, and this body does the same.

Gross margin, at 4 out of 12, is the moat item where the Supermicro story becomes genuinely two-sided, and it requires careful reading, because two margin observations in the file point in opposite directions. Generally accepted accounting principles gross margin for fiscal 2026 was 10.8 percent, confirmed by the Form 10-K, down from 18.0 percent in fiscal 2023 in a sustained dilution that the file cites as empirical support for a research hypothesis about margin dilution in successful blitzes. The quarterly path within fiscal 2026 is a rollercoaster: 6.3 percent in the second quarter, confirmed by the Form 10-Q, rebounding to 17.5 percent in the fourth quarter, with part of that rebound attributed on a press-reported basis to non-recurring factors, namely reduced tariff costs and an inventory allowance reversal. First-quarter fiscal 2027 guidance for gross margin of 10.4 to 10.8 percent, on a press-reported basis, does not yet demonstrate that the rebound is durable. The score of 4 therefore reflects the level of margin, a thin-margin structure, rather than the direction of the latest quarter. A company whose annual margin is 10.8 percent and whose quarterly margin swung from 6.3 percent to 17.5 percent and back toward guidance of about 10.5 percent is a company whose pricing power, as the scorecard measures it, is thin and whose margin path is unstable. The file records a model question arising from this item, whether the moat section should score a trailing annual figure or reflect quarterly volatility, and this body records that question as open rather than resolving it, because the score at T uses the trailing annual figure of 10.8 percent.

Creative destruction: riding a real replacement, with an unproven cost claim

Creative destruction scores 10 out of 15, or 66.7 on the radar conversion, from three items: replacing the existing order 4 out of 5, irreversible behaviour change 4 out of 5, and cost-structure change 2 out of 5. The section's message is consistent across the first two items and then breaks on the third, and that break is informative. Supermicro is participating in a genuine structural replacement, with lock-in that follows from physical constraints, while its claim to have changed the cost structure remains unmeasured.

Replacing the existing order, at 4 out of 5, describes the replacement of general-purpose central processing unit servers by accelerated graphics processing unit servers. IDC expects accelerated servers to account for more than 95 percent of global artificial intelligence infrastructure spending by 2029, on a press-reported basis. Most of Supermicro's fiscal 2026 revenue of approximately $39.1 billion comes from NVIDIA graphics-processing-unit-based artificial intelligence servers, on filing evidence. And the company's rise to second place in the global server market, on IDC evidence cited on a press-reported basis, places it on the winning side of a structural transition in the hardware market. The score stops at 4 for the reason the file implies through its other items: the replacement is real and the company is riding it, but the company's measured share of the specific accelerated market is unconfirmed, so the score cannot credit a completed displacement at the top mark. The replacement is therefore scored as a structural shift in progress with Supermicro positioned inside it, rather than as a transformation the company itself has completed.

Irreversible behaviour change, at 4 out of 5, describes the infrastructure conversion from air cooling to direct liquid cooling through DLC-2. The physical constraint behind the claim is stated in the file with unusual bluntness: graphics processing unit racks above 100 kilowatts cannot be air-cooled, on company investor-relations claims plus the physical constraint itself, on a press-reported and estimated basis. Customer investment in retrofitting data centres for liquid cooling is therefore difficult to reverse, because the retrofit is a structural commitment to a new cooling architecture rather than a purchase decision that can be undone at low cost. Rack-scale integrated validation before shipment, on earnings-call evidence cited on a press-reported basis, deepens that lock-in, because a validated rack is a tested system rather than a collection of parts. The record backlog, confirmed by the Form 8-K, is consistent with customers committing at scale to that architecture. The score stops at 4 rather than reaching the top mark, and the file's treatment of the neighbouring items suggests why: the lock-in is structural, but the evidence for it is a mix of company claims, press reporting, and physical reasoning rather than a measured customer retention or switching figure. The behaviour change is therefore scored as strongly supported but not fully demonstrated in the form the top mark would require.

Cost-structure change, at 2 out of 5, is the lowest item in the section, and the gap between claim and proof is widest here. The DLC-2 full-stack in-house approach is claimed, on investor-relations evidence dated September 23, 2026 and cited on a press-reported basis, to improve graphics processing unit density per rack and power efficiency. But no company-disclosed measured figure for unit cost collapse, in cost per floating-point operation, has been published, and that figure is recorded in the file as unconfirmed. The fourth-quarter margin rebound, part of which came from non-recurring factors as discussed above, is not evidence of a cost-structure improvement, and the file says so directly. A model that claims to change costs, without publishing the unit costs that would prove the change, is scored as an assertion rather than a demonstration. The creative destruction section as a whole therefore reads as a company riding a real replacement, with lock-in grounded in physics, and with a cost claim that has not yet been measured. That is a fair reading of a company whose new architecture is shipping while its unit economics remain thin.

Deductions and risks: exactly at the FLAG line, for three distinct reasons

Deductions total 15 points: debt negative 4 out of a possible negative 10, revenue quality negative 4 out of a possible negative 10, governance negative 3 out of a possible negative 5, and regulatory and social costs negative 4 out of a possible negative 5. That total is exactly the FLAG watch threshold, and this body repeats the point because it is the single most decision-relevant fact in the score: the company is not below the line with room to spare, and it is not above the line. It is on the line. The file's directional note adds that if any single deduction item worsens in the next quarter, the watch status is confirmed rather than merely approached. The deductions are therefore not background colour. They are the mechanism by which the next quarter's evidence will move the classification, and each of the four items deserves separate attention because each rests on different evidence and points to a different kind of risk.

Debt, at negative 4, is the deduction where the balance sheet looks light in absolute terms and strained in its movement, and the file weighs both sides. Total borrowings were $8.7 billion as stated in the Form 10-K, confirmed, equal to approximately 0.22 times trailing twelve-month revenue on an estimated basis, with interest coverage of 14.2 times, also an estimate. Cash was $7,521.5 million, confirmed. Those levels, taken alone, would suggest a light burden. The deduction rests instead on the direction and composition of the borrowing. Bank borrowings rose from $112.5 million to $4,056.1 million in one year, a roughly 36-fold increase for working capital funding, confirmed by the Form 10-K. Receivables rose 178 percent and inventory rose 175 percent year over year, confirmed by the Form 10-K. Customers pay on long terms while vendors are paid on short terms, and the Form 10-K's risk factors acknowledge reduced available cash, confirmed. Interest expense rose from $19.4 million in fiscal 2024 to $194.6 million in fiscal 2026, a tenfold surge, confirmed. And the funding is dilutive in form: preferred shares of approximately $4.3 billion at a 7 percent cumulative dividend, confirmed, plus the unexecuted at-the-market facility of $1.25 billion, confirmed. The file names working-capital leverage as the core of the deduction, and that framing is the right way to read the item. The company is not over-borrowed against its revenue. It is funding an explosion in receivables and inventory with short-term bank borrowing while its customers pay slowly, and the interest bill has multiplied tenfold in two years. That is a working-capital story, not a solvency story, and the score of negative 4 reflects the distinction.

Revenue quality, at negative 4, is the deduction where the past is cleaner than the present control environment, and the file is careful to separate the two. The 2024 accounting issue closed without a restatement: a special committee stated on December 2, 2024 that no restatement was expected, confirmed by a Form 8-K, and the fiscal 2024 Form 10-K was filed on February 25, 2025, restoring Nasdaq compliance, confirmed. Revenue and gross margin figures are unchanged across the fiscal 2024, 2025, and 2026 Forms 10-K on an XBRL comparison, confirmed, so the quality problem for revenue itself is resolved. The deduction rests on what remains unresolved. Internal control over financial reporting was still ineffective at June 30, 2026, with a material weakness remaining, confirmed by Form 10-K Item 9A. The auditor EY resigned on October 30, 2024, confirmed by a Form 8-K, and BDO was appointed. The file notes that a candidate model refinement, an anchor for internal-control material weakness in the revenue quality item, was pre-applied in this scoring, so the deduction of 4 is the file's deliberate treatment of the principle that no restatement does not equal no quality problem. A reader should therefore keep both halves of the item together, as the file does: the numbers themselves survived comparison across three annual filings, and the control system that produces those numbers is still judged ineffective. That combination is more informative than either half alone, because it explains why a resolved accounting episode can still cap a score.

Governance, at negative 3, describes a concentrated family structure that the file documents from the Form 10-K with unusual specificity. Founder Charles Liang serves as President and Chief Executive Officer and as Chairman, confirmed by the Form 10-K. His spouse Sara Liu is a co-founder, a Senior Vice President, and a Director, confirmed. The Liang couple's effective stake is approximately 10.1 percent, confirmed. Related-party transactions run through the supply chain itself: Ablecom, connected to the chief executive's brother Steve Liang with a family stake of 35.5 percent, supplies 95.3 percent of chassis purchases, and Compuware, connected to his brother Bill Liang, supplies 94.4 percent of power supplies plus a $200 million trade credit line, all confirmed by Form 10-K Note 11. There are no differential voting rights, confirmed, which limits the governance discount that a dual-class structure would otherwise imply. The deduction of 3 therefore reflects two specific features rather than a general judgement: family concentration in the supply chain at levels above 94 percent for two critical inputs, and the combination of chief executive and chairman roles in the founder. Neither feature is presented in the file as an allegation of misconduct, and this body does not present it as one. It is a structural observation about concentration and control, and the score marks the distance between a dispersed supply chain with separated leadership roles and the structure that exists at T.

Regulatory and social costs, at negative 4, are the largest single deduction alongside revenue quality, and they rest on the most serious evidence in the file. An export-control indictment dated March 19, 2026 in the Southern District of New York named three former associates over alleged diversion to China of NVIDIA artificial intelligence servers worth approximately $2.5 billion, confirmed by Form 10-K Note 15. One of those former associates, Sun, pleaded guilty to four charges on October 8, 2026, on a Reuters press-reported basis. That date is before T, so the plea is evidence at T. The file states a distinction that this body preserves exactly, because it is the most consequential sentence in the item: the company itself is not an indictment subject, and an internal independent investigation by Munger Tolles & Olson found no evidence that current management knew of the conduct, confirmed by the Form 10-K. What remains ongoing is extensive: Securities and Exchange Commission subpoenas dated November 19, 2024 and April 28, 2026, a Southern District of New York grand jury, and Department of Justice and Bureau of Industry and Security matters, all confirmed, with no notice that the company is a target. Securities class actions number three plus three consolidated, confirmed, and shareholder derivative suits number five federal plus three state, confirmed. The deduction of 4 therefore reflects a former associate's guilty plea plus multiple ongoing federal investigations plus active litigation, weighed against the confirmed facts that the company is not indicted and that the independent investigation found no evidence of current management knowledge. The file records a model question arising from this item, how to score a criminal case where the company is not the defendant and the guilty plea is limited to a former associate, and this body records that question as open, because the score at T applies a deduction of 4 on the file's judgement without a settled anchor for that boundary.

What to watch: the first verification point of the second act

The score at T is a snapshot of a company mid-explosion with its control environment lagging behind, and the file's directional note identifies the observation points that will test whether that snapshot improves or deteriorates. The first is fiscal 2027 first-quarter results, expected November 3, 2026, recorded here only as a future marker with no anticipated outcome. The file's directional note gives the guidance figures that frame that marker: revenue of $14.5–15.5 billion and gross margin of 10.4 to 10.8 percent. Those guidance figures are themselves pre-T evidence, reported on a press-reported basis in the file's treatment of the margin guidance, and they matter because of the direction they point: margin guidance of about 10.5 percent sits far below the fourth-quarter rebound of 17.5 percent and close to the annual figure of 10.8 percent. If the reported quarter lands near that guidance, the file's reading is that the fourth-quarter rebound will have been confirmed as partly non-recurring, which bears directly on the gross margin item. If the quarter instead sustains a higher margin, the file's framework would treat that as evidence the rebound had a durable component, though no such outcome is anticipated here. Either way, the quarter is the first verification point for the durability of the second explosion, because it tests whether the growth rate and the margin can move together rather than trading against each other.

The second observation point is the deduction side, where the file identifies two possible directions of movement and is explicit that neither can be reflected on pre-T evidence. On the favourable side, closure of the government investigations or resolution of the internal-control material weakness could ease the revenue quality and regulatory deductions from their combined 8 points, though the file states that no pre-T evidence supports such a reflection. On the unfavourable side, any worsening in a single deduction item in the next quarter confirms the watch status that the current total of exactly 15 points only approaches as a line. That asymmetry is the practical meaning of sitting exactly on the FLAG threshold: the classification has no cushion in either direction, and the next quarter's evidence on controls, investigations, or working capital will move it one way or the other. The working-capital indicators belong in the same watch list: bank borrowings, receivables, and inventory grew far faster than the already fast revenue growth in fiscal 2026, and whether that gap narrows or widens in the next quarter is a direct test of the debt item's core thesis.

A final observation belongs outside the score, because the file places it there deliberately. The valuation note, which is not reflected in the score, records a market capitalisation of approximately $23.7 billion at October 14, 2026, on a press-reported basis, with a price-to-sales ratio of approximately 0.6 times on trailing twelve-month revenue of approximately $39.1 billion. The 52-week range was $19.48 to $58.78, and the all-time high was $122.90 on March 8, 2024. The file adds, citing a lesson from an earlier test, that the business score does not look at price. That separation is worth preserving in this body, because it prevents the most common misreading of a low-multiple, high-growth company: the score of 43 is a judgement about the business and its risks, not about the market's pricing of them. The business grew 77.8 percent in fiscal 2026, guides to 66 to 84 percent growth in fiscal 2027, operates in a market estimated at $838–982 billion for 2030, and distributes through a machine that spends 0.90 percent of revenue on sales and marketing. It also carries a thin and volatile margin, a working-capital surge funded by short-term bank borrowing, an ineffective internal control system, a concentrated family supply chain, and a former associate's guilty plea inside multiple ongoing federal investigations. The score of 43, grade C, with deductions exactly at the FLAG watch line, is the file's attempt to weigh those two halves together. The next quarter will show which half moves.

More AI Infra companies

Other AI Infra names on the Blitz Index, highest score first: NVIDIA (87/A), Arista Networks (73/B), Advanced Micro Devices (62/B).

Next earnings

Q4 2026 · Expected

Coverage

Super Micro Computer, Inc. · Q1 FY2027 due Nov 3, 2026 (AMC) · Score 43 · Grade C · ⚠️ BlitzFailing watch

Embed this score

Copy the HTML below to show this score on your site. The badge links back to this page and updates automatically when the score changes.

<a href="https://blitzlab.app/companies/smci/"><img src="https://blitzlab.app/badge/smci.svg" alt="Super Micro Computer Blitz Index 43/C — BlitzLab"></a> Copy HTML

Preview:

BlitzLab scores businesses against a scholar’s standard — not investment advice. No price targets, no buy/sell recommendations. Figures are from SEC filings (XBRL) as labeled in the source scorecard.