Seeing Clearly: The Art and Science of Net Assessment

Post 2 of 12. The question: how do serious strategists actually evaluate military competition — and what does that methodology reveal that bean-counting hardware never can?

Mar 19, 2026

silhouette of soldiers — Photo by Duncan Kidd on Unsplash

Most defense analysis counts things. Warheads. Fighter squadrons. ships at sea. Satellites in orbit. The implicit theory is that the side with more of the right things wins, and so the central analytical task is to build and maintain an accurate tally. This is not wrong, exactly, but it is radically incomplete — and the incompleteness has consequences. A state that counts better than its adversary is better informed. A state that understands better — that grasps what the numbers mean, how they interact with doctrine and culture and organizational behavior, where the trends are heading across decades — holds a different and deeper advantage.

This distinction is what the Pentagon’s Office of Net Assessment was built to exploit. And its most important product was not a spreadsheet of Soviet capabilities. It was a theory of how the United States could out-think an adversary it could no longer reliably outspend.

Andrew Marshall and the Birth of ONA

Andrew Marshall arrived at the Pentagon in 1973 as the first Director of the newly created Office of Net Assessment, appointed by Defense Secretary James Schlesinger. He would hold that position for over four decades — longer than most analysts’ entire careers — becoming, in the words of Russian observers, the “Pentagon’s Gray Cardinal.” Brands captures the paradox neatly in The Eurasian Century: Marshall was one of the most influential strategists most people had never heard of. During the 1970s, he formulated a theory of how America could regain the strategic edge, which Reagan then turned into a theory of Western victory in the Cold War [1].

What Marshall found when he arrived was an intelligence community that had, in his view, a fundamental analytical problem. The CIA was producing technically competent assessments of Soviet capabilities but was doing so through a flawed model: the Soviets as a unified, rational actor pursuing a single, easily-stated strategy [2]. This model flattened the very asymmetries — in culture, organizational behavior, strategic culture, and resource allocation logic — that were most important for understanding the long-term competition. Marshall believed the Soviets were different in ways that the rational-actor model systematically obscured, and that understanding how they were different was more valuable than any weapons count.

The early ONA portfolio reflected this. Marshall and Schlesinger quickly settled on three core competition areas: the US-USSR strategic nuclear balance, the NATO-Warsaw Pact conventional balance in Europe, and the maritime balance. A fourth was soon added: a military-investment balance comparing defense economics, with particular emphasis on Soviet R&D spending as a leading indicator of future capabilities [3]. That fourth balance was telling. Marshall suspected — correctly, as it turned out — that the CIA was underestimating the Soviet defense burden. He worried the United States might be “pricing itself out of the military competition with the Soviets, or at least severely handicapping itself” [4].

His prescription was characteristically counter-intuitive. Since the U.S. could no longer outspend the Soviets substantially, Marshall argued, it would have to out-think them [1]. That sentence — quiet, almost offhand in its original context — became the intellectual foundation for the second offset strategy, for the precision-guided munitions revolution, and ultimately for the RMA debates that would consume American strategic thinking for the next three decades.

What Net Assessment Actually Is

The DoD’s formal definition is dry but precise: net assessment is the “comparative analysis of military, technological, political, economic, and other factors governing the relative military capability of nations” [5]. The key word is comparative. Net assessment is not an assessment of one side. It is, by definition, an assessment of a relationship — how two or more states stack up against each other, across multiple dimensions, as that relationship evolves through time.

Marshall’s 1972 conception was even more concrete. A net assessment is a careful comparison of weapon systems, forces, and policies — comprehensive, including operational doctrines, training regimes, logistics, known or estimated effectiveness, design practices, procurement patterns [6]. Its purpose is diagnostic: to highlight efficiencies and inefficiencies, comparative advantages and disadvantages. Crucially, it is not intended to produce recommendations on force levels or structure. That discipline — separating diagnosis from prescription — is part of what makes net assessment intellectually honest and strategically useful. It does not pretend to certainty it does not have.

Mahnken, who worked closely with Marshall’s legacy, identifies seven characteristics that distinguish net assessment from other forms of defense analysis [7]:

The first is an emphasis on competitive interaction — treating national security competition as a dynamic between organizations, not a static snapshot of capabilities. The second is attention to bureaucratic, organizational, and cultural factors that often produce sub-optimal behavior on both sides — the assumption that adversaries are not perfectly rational is not cynicism but realism. The third is comfort with uncertainty and imperfect information, using qualitative alongside quantitative data in ways that systems analysis or cost-benefit frameworks resist. The fourth, flowing from the others, is an emphasis on asymmetry — hunting the key differences between competitors, understanding their impact, and asking which ones can be exploited [8]. The fifth is thinking in time, characteristically over two to three decades, placing heavy emphasis on long-term trends rather than near-term capability snapshots [9]. The sixth is multidisciplinary analysis, combining scenario analysis, wargaming, economics, history, and organizational theory. And the seventh is being descriptive rather than prescriptive — illuminating two or three emerging problems or opportunities rather than dictating solutions [10].

That final characteristic deserves particular emphasis. Net assessment’s “limited aspirations,” as Mahnken puts it, are a feature, not a bug. It does not attempt to predict the outcome of a particular battle. It seeks to enumerate and evaluate the strengths and weaknesses of each side over time — and to put that diagnosis in front of decision-makers who can act on it.

The Ancient Roots: Sun Tzu Was Doing This First

Net assessment is a modern institutional practice, but its intellectual roots run to antiquity. The first chapter of Sun Tzu’s Art of War can be translated as “On Assessments” — and the concept lies at the heart of its most famous maxim: know the enemy and know yourself; in a hundred battles, you will never be in peril [11].

Two things stand out in Sun Tzu’s formulation that are worth making explicit. First, it is the balance between the two sides — not the absolute capabilities of either — that matters. A force that is objectively weaker can prevail over a stronger one if it understands the relationship correctly and the stronger side does not. Second, many of the dimensions of military power Sun Tzu identifies are qualitative and resist quantification [12]. Morale, cohesion, the quality of leadership, the adaptability of doctrine under pressure, the resilience of logistics under adversarial conditions — none of these appear cleanly in an order-of-battle count, and yet all of them have repeatedly decided contests that the hardware ledger predicted differently.

Mahnken and Marshall also reach back to the Peloponnesian War for a pre-modern example. Before the war began, both Pericles and Archidamus performed what were, in effect, net assessments — calculating the relative social, political, economic, and military strengths and weaknesses of Athens and Sparta, examining scenarios, and reaching public diagnoses [13]. They reached different conclusions from roughly similar analyses. That divergence was not a failure of the methodology. It was the methodology working as intended: surfacing the key judgments on which strategic decisions would turn.

The Intellectual Competition Argument

Marshall’s most important and most counterintuitive finding was not about Soviet hardware. It was about Soviet thinking. While the U.S. military was at the forefront of developing new technologies — precision strike, stealth, advanced surveillance — Soviet military leaders were, paradoxically, at the forefront of theorizing about the changing character of war [14]. Marshall believed the Soviet military theorists, rather than American ones, were intellectualizing the longer-term consequences of the technical changes the American military had initiated.

This is a striking inversion. The United States was building the tools of a military revolution it had not yet conceptualized. The Soviets were conceptualizing a revolution they lacked the resources to execute. The intellectual lead, in other words, was not where the material lead was — and Marshall understood that the intellectual lead was, in the long run, more consequential. It was Soviet theorists who coined the term “military-technical revolution” (MTR) and produced the foundational literature on what the West would later call the RMA. That story will be told in full in Post 3.

Mahnken’s Competitive Strategies for the 21st Century extends this logic into a practical framework. The competitive strategies approach is not about matching adversary capabilities symmetrically. It is about getting competitors to play your game — getting them to make the kinds of mistakes they are inclined to make, by exploiting cognitive asymmetries and structural constraints that are different on each side [15]. This requires a level of understanding of the adversary that goes well beyond capability counts: you need to know not just what they are doing but why — their institutional incentives, their doctrinal assumptions, their organizational blind spots [16].

The second offset strategy was, in practice, a competitive strategy of exactly this kind. By investing in precision guidance, stealth, and advanced ISR rather than matching Soviet conventional mass, the U.S. forced the Warsaw Pact to respond to American strengths on American terms. The result, as Jones notes in The American Edge, was that by the early 1980s Soviet defense minister Dmitri Ustinov was telling Warsaw Pact colleagues that the military balance was “not in our favor” [17].

Net Assessment as the Series’ Methodology

This post is not simply historical background. It is an introduction to the analytical method the rest of this series will apply.

Every post from here forward is, in structure, a net assessment of a specific military competition. Post 4 assesses the interwar competition between blitzkrieg innovators and their opponents. Post 5 assesses the emerging space power balance. Posts 8 and 9 assess the drone and missile warfare competitions in Ukraine and the Middle East. Posts 10 through 12 assess the U.S.-China competition across space, AI, and autonomous systems.

In each case, the analytical habits introduced here apply: look at both sides simultaneously, not just the American or Western perspective. Hunt for asymmetries — the structural differences that create opportunity or vulnerability for one side. Think in decades, not budget cycles or news cycles. Resist the purely quantitative — doctrine, culture, organizational adaptability, and conceptual innovation are often the decisive variables. And stay descriptive rather than prescriptive long enough to understand the problem clearly before reaching for solutions.

Krepinevich, who spent years inside the ONA orbit, captures the stakes clearly in The Origins of Victory: ONA examines the balance in key regions where the United States has vital interests and functional balances such as the competition in space, in the undersea, and in strategic forces [18]. The space functional balance — the subject of Posts 5 and 6 — is perhaps the most consequential and least understood competition currently underway. Net assessment is the tool designed to see it clearly.

The Most Dangerous Form of Strategic Ignorance

Marshall’s critique of the CIA’s rational-actor model of Soviet behavior points toward something that transcends the Cold War context. Every era’s strategic community has an equivalent blind spot — a simplifying assumption about adversary behavior that is comfortable, analytically tractable, and wrong in the ways that matter most.

The American strategic community has, at various points, assumed that adversaries share American risk tolerance, American time horizons, American definitions of victory and defeat, and American sensitivity to costs. These assumptions have not fared well against the evidence. The Vietnamese, the Iraqis, the Afghans, the Russians in Ukraine, and the Chinese in their long-horizon space and cyber investments have all behaved in ways that confounded American predictions built on these models.

Net assessment was designed to disrupt that pattern — to force analysts to sit inside the adversary’s decision-making logic, understand their institutional incentives, and map their blind spots as carefully as their own. Marshall’s lesson, still underinternalized in Washington, is that the most dangerous form of strategic ignorance is not knowing what your adversary can do. It is misunderstanding why they do what they do.

The series that follows is an attempt to apply that lesson — to the history of military revolutions, to the current wars, and to the competition now unfolding in the domain that may decide the next century.

Referenced Highlights

[1] “His second career, as director of the Pentagon’s secretive Office of Net Assessment, made him one of the most influential strategists most people had never heard of. Russian observers would later call Marshall the ‘Pentagon’s Gray Cardinal’ — and for good reason. During the 1970s, Marshall had formulated a theory of how America could regain the strategic edge, which Reagan then turned into a theory of Western victory in the Cold War. ‘The United States will have to out-think the Soviets,’ wrote Marshall in one paper, since it could no longer ‘outspend them substantially.’”

The Eurasian Century — Hal Brands. Open in Readwise

[2] “As far as Marshall could tell, the Central Intelligence Agency was not withholding information from the White House, and the analyses of hard data and the factual reporting on Soviet forces appeared to be good. On the other hand, he found CIA’s assumptions about Soviet behavior to be poor, including reliance on a ‘model of the Soviet government as a single unified actor pursuing an easily stated strategy.’”