Presidents Ranked by Foreign Policy Outcomes, FDR to Biden

Outcomes, not doctrines

Foreign policy reputations are built on speeches and summits. This table ignores both. It counts four things that leave records: wars begun and ended, treaties ratified and alliances formed, American service deaths, and results still verifiable decades later. The casualty numbers come from the Congressional Research Service's standing report RL32492 and the Pentagon's Defense Casualty Analysis System. The treaty record comes from the Senate, which either ratified an agreement or did not.

Measured this way, the top of the table rewards builders and enders. Roosevelt and Truman rank first and second because the war they won and the institutions they built, the UN, Bretton Woods, NATO, the Marshall Plan, are still the operating system of world politics 80 years later. Eisenhower ranks third for a simpler reason: he ended a war in six months and then went eight years without starting one, a record no Cold War successor matched (National Archives, Korean armistice; CRS deployment records).

The subscores make the arithmetic visible. Roosevelt earns a near-perfect War Outcomes mark and a perfect Treaties and Alliances mark, but his Force Protection meter reads a 4 out of 20 because the bulk of 405,399 World War II deaths fell in his tenure. He still ranks first because three of the four components are at or near the ceiling and the fourth is a cost of the largest victory in the history of the office. The framework does not pretend the deaths did not happen. It weighs them against what they bought.

A word on how the composite is built, because the ranking's credibility rests on it. Each president receives four component marks, each on a 20-point scale, and the four are summed and rescaled to the 100-point score in the entry. The marks are not opinions dressed as numbers. War Outcomes rises with wars ended or won and falls with wars begun and left open, using the CRS deployment record to decide which is which. Force Protection is the ordinal inverse of service deaths in tenure, so Carter and Clinton, who lost almost no one, sit near the top of that column while Johnson and Roosevelt sit near the bottom. Treaties and Alliances counts ratified instruments and formal alliance gains from the Senate record. Durability asks one question with a July 2026 answer: is the result still operating. A reader who disagrees with a placement can find exactly which component drove it and check that component against the cited source. That transparency is the point. The ranking is not asking to be trusted; it is showing its work.

The price column

Service deaths are the hardest column and the least ambiguous. World War II cost 405,399 American military deaths, Korea 36,574, Vietnam 58,220, the Gulf War 383, Iraq 4,431, and Afghanistan roughly 2,461 (CRS RL32492; DoD). Attribution follows the calendar. Vietnam's toll falls mainly on Johnson, about 30,000 deaths during his tenure, and Nixon, about 21,000, which is why Johnson ranks fourteenth despite genuine treaty achievements and Nixon ranks fifth only because his exits and openings, China, SALT I, the ABM Treaty, the Paris Accords, weigh against his share of the war.

Johnson's Force Protection subscore of 6 and War Outcomes subscore of 5 are the two lowest marks on the board for those categories, and they explain his fourteenth-place finish more than any other numbers. He signed the Nuclear Non-Proliferation Treaty in 1968 and the Outer Space Treaty in 1967, both still in force, which lifts his Treaties and Alliances mark to a respectable 15. But a war that killed 30,000 Americans on his watch with no settlement is a debit no treaty offsets on this ledger.

The same ruler produces the table's most contested placement. George W. Bush ranks last because the countables are one-directional: two wars started, zero ended, roughly 5,800 American deaths in tenure, and a casus belli, Iraqi weapons stockpiles, that the administration's own Iraq Survey Group concluded did not exist (Duelfer Report, 2004). His War Outcomes subscore of 3 is the lowest on the board. The ledger credits his NATO expansion and PEPFAR's diplomatic weight, and it does not care about his intentions. If that placement produces discomfort, the discomfort belongs to the reader, not the data.

The recent terms, scored coldly

The three most recent completed records, Obama, Trump's first term, and Biden, land mid-table for mirrored reasons. Obama ended the Iraq deployment, signed New START, and ordered the bin Laden raid, but the 2011 exit was followed by ISIS seizing a third of Iraq by 2014, forcing re-intervention. That sequence is why his Durability subscore reads 10 out of 20: the withdrawal did not bind, and the caliphate that followed erased much of the strategic gain. Trump started no new wars, a countable fact reflected in his Force Protection subscore of 19, and the Abraham Accords added four normalizations that have held, but he also withdrew from three ratified or signed frameworks, the INF Treaty, the JCPOA, and Paris, which pulls his Durability mark down to 9.

Biden's entry is the study in split outcomes. Ending America's longest war is a countable positive; the manner of the ending, the Kabul collapse and 13 dead at Abbey Gate in August 2021, is a documented negative (DoD casualty releases). Adding Finland and Sweden to NATO, doubling the alliance's border with Russia and adding two first-rate militaries, is the largest alliance gain since 2004, ratified by the Senate 95-1 (NATO; Senate records). His Treaties and Alliances subscore of 15 captures that gain; his Durability mark of 11 reflects a record still too fresh to test. The table nets these at eleventh, and shows its arithmetic.

What holds up over decades

The strongest signal in this dataset is durability, and it is the one subscore where the top of the table pulls away most sharply. Carter's Egypt-Israel treaty has held for 47 years without a violation between the parties, which is why his Durability subscore is 18 and a presidency remembered for the hostage crisis still ranks ninth (history.state.gov). Truman's NATO now has 32 members, earning a perfect Durability mark. Nixon's China opening set the frame for a half-century, a Durability subscore of 18. By contrast, unratified or reversed frameworks, the Paris Agreement exits and re-entries, the JCPOA's collapse, the Doha timetable, score in the single digits precisely because they did not bind successors.

The pattern is not partisan. It is structural: outcomes ratified by the Senate and embedded in institutions outlive presidencies, and outcomes that live on a president's signature usually do not. Reagan's INF Treaty, ratified 93-5, gave him a Durability mark of 16 even though the treaty itself was later abandoned, because the framework governed for three decades. The George H. W. Bush administration's START I, ratified and implemented, anchors a Durability subscore of 15. The lesson the numbers teach is old and unglamorous: get it through the Senate, and it lasts.

Wars begun, wars ended

The cleanest way to read this table is to sort presidents into enders, builders, and starters. The enders sit at the top or recover from low positions on the strength of a single act. Eisenhower ended Korea. George H. W. Bush fought and ended the Gulf War in six weeks. Nixon signed the Paris Accords. Obama ended the Iraq deployment. Biden ended Afghanistan. Ending a war is the single most reliable predictor of a strong War Outcomes subscore, and the enders' marks range from 11 to 18.

The starters occupy the bottom. Johnson escalated Vietnam. George W. Bush began Afghanistan and Iraq. Both left office with the wars unresolved, and both carry War Outcomes subscores in the low single digits, 5 and 3 respectively. The pattern is stark enough to be a rule: on this ledger, starting a war and not finishing it is the surest path to the bottom of the table, and no volume of treaty signings offsets it. Ford is the instructive middle case. He started nothing, but he presided over the fall of Saigon, a war's end in defeat rather than settlement, which is why his War Outcomes subscore of 7 sits below every ender and above every starter.

The builders are the presidents who left an institution rather than a battlefield. Truman built NATO, the Marshall Plan, and the Berlin Airlift, and his Treaties and Alliances subscore of 20 is the reason he ranks second despite an unresolved Korean War. George H. W. Bush managed the Soviet collapse and German reunification without a shot fired between the superpowers, then signed START I and won a war he also ended, a rare double that lifts him to fourth. Reagan's INF Treaty eliminated an entire class of nuclear weapons by treaty, ratified 93-5, and his summit diplomacy closed the Cold War. The builders tend to rank high because their work is exactly what the Durability subscore is designed to reward: results still standing decades later. A war ended is a moment. An alliance built is a permanent fixture until someone dismantles it, and dismantling a ratified alliance is far harder than reversing an executive order.

The treaty gap and the modern presidency

One structural fact runs underneath the whole table: the ratified treaty has nearly vanished from the modern presidency, and the Treaties and Alliances subscores show it. The last four completed presidencies, Obama, Trump's first term, Biden, and the tail of Clinton's, produced exactly one alliance expansion at Cold War scale, Biden's Finland and Sweden accessions. The great instruments of the postwar order, the UN Charter, the NATO treaty, the nuclear arms-control regime, all cleared the Senate's two-thirds bar in an era when that bar was reachable. It is no longer reliably reachable, and presidents have adapted by governing through executive agreements, memoranda of understanding, and unratified frameworks that a successor can erase.

This is why the New START of 2010 (ratified 71-26) and the Finland-Sweden accessions of 2023 and 2024 (ratified 95-1) stand out so sharply in the recent data: they are the exceptions that prove the rule (Senate treaty records; NATO). The Abraham Accords, the JCPOA, the Paris Agreement, and the Doha agreement were all executive instruments, never submitted for ratification, and all four scored weakly on Durability precisely because they lived on a signature. The framework does not penalize a president for the collapse of Senate comity, which no one president caused. But it does count what bound and what did not, and the record is unambiguous: the ratified instrument outlives its author, and the executive framework usually does not survive the next inauguration.

The consequence for the ranking is that the top of the table is likely to stay frozen. To crack the top four, a modern president would need to end a major war and win a Senate-ratified alliance or arms-control treaty of Cold War magnitude in the same tenure, the combination Truman and George H. W. Bush achieved. Nothing in the recent record approaches it. The mid-table crowding of Obama, Trump's first term, and Biden, all within four points of one another, reflects a common ceiling: strong Force Protection marks, modest Durability marks, and no treaty gain large enough to break out. The scoreboard is not rewarding or punishing anyone for that. It is describing an era.

What the table cannot count

Every honest scoreboard names what it leaves off. This one omits three things. First, covert action and intelligence outcomes rarely leave a public ledger entry, so coups, cyber operations, and clandestine deterrence do not appear in the subscores even when historians agree they mattered. Second, economic statecraft, sanctions regimes, trade agreements, and development programs like PEPFAR, is credited only where it produced a ratified instrument or a verifiable strategic result; PEPFAR lifts George W. Bush's Durability mark but cannot rescue two unfinished wars. Third, the counterfactual, the war that did not happen because a president deterred it, is by definition unrecorded, and Eisenhower, Kennedy, and George H. W. Bush all benefit from crises that ended without shots the table can count.

These omissions cut in every direction, which is why the framework keeps them out rather than guessing. A ranking that credited unprovable deterrence or debited unprovable missed opportunities would become an opinion survey wearing a scoreboard's clothes. The four subscores stay inside the documented record: deployments in CRS deployment tables, deaths in RL32492 and the DoD casualty system, treaties in the Senate's ratification record, and durability in whether the result still operates in July 2026. Everything the reader sees on an entry traces to one of those four sources. That is the whole discipline of the exercise, and it is the reason the surprising placements, Roosevelt first despite 405,399 deaths, George W. Bush last despite genuine achievements, hold up when readers check the receipts.