The 2026–2027 Federal Recompete Wall: $1.2 Trillion Coming Due and Who Actually Owns It

Size of the wall

The number that can be published is not “every federal contract that might recompete someday.” It is the complete ≥$100M pull: 1,577 contracts, $1,195.02B estimated value, with a period-of-performance end date inside the next 18 months from 2026-08-15. SCOUT estimated_value on the named definitive contracts in that set tracks USASpending base_exercised_options — money currently on contract, including exercised options, not obligated transaction flow and not ceiling.

The unfiltered 18-month query (find_recompete_contracts, within_months=18, ≥$100K floor) is a different list. It matched 140,488 rows and scored the 50,000 soonest PoP-end rows: 50,000 contracts, $338.54B estimated. That scored subset is not the wall. Large-value contracts with later end dates were dropped by the cap. Use the complete ≥$100M pull.

The wall is work with a dated PoP end inside a defined window. That is a different list from what is currently posted on SAM.gov.

Related published SCOUT snapshots — useful as a cross-check, not as a substitute for this pull:

• DoD-only wall: $353.55B / 213 contracts inside 18 months (data as of 2026-07-06).
• DOE nuclear M&O wall: $320.92B / 38 contracts inside 24 months (data as of 2026-07-06).

Those two slices already show why a government-wide ≥$100M number is large, and why it cannot be copied from a single-agency page. The $1,195.02B total is a SCOUT construct. USASpending has no PoP-end recompete query, so this headline cannot be reproduced as a single public-site total. Where a row failed the independent spot-check, we withheld the row rather than subtracting a guessed correction from the headline.

Agency breakdown

Same window, same ≥$100M floor:

Agency	Contracts	Est. value	Share
Department of Defense	1,014	$532.31B	45%
Department of Energy	28	$265.80B	22%
Department of Veterans Affairs	74	$166.38B	14%
NASA	82	$77.92B	7%
Department of State	59	$36.70B	3%
Department of Homeland Security	73	$26.71B	2%
General Services Administration	59	$26.20B	2%
Department of Health and Human Services	73	$20.39B	2%
USAID	19	$16.08B	1%
Department of Justice	14	$6.48B	1%

DoD, DOE, and VA hold 81% of the dollars. Count tells a different story than dollars: DoD’s 1,014 contracts are the volume. DOE’s 28 contracts carry 22% of the value — M&O-scale awards, not a mid-size task-order pile. NASA has more contracts than DOE (82 vs 28) and less than a third of the dollars.

Civilian agencies are easy to miss under DoD’s 45%. State ($36.70B / 59), DHS ($26.71B / 73), GSA ($26.20B / 59), and HHS ($20.39B / 73) are each a real capture lane. They do not move the government-wide total. They do move a firm that does not sell to DoD.

This agency mix is SCOUT-only. There is no public USASpending equivalent for “PoP end inside 18 months, ≥$100M.”

Incumbent concentration

A name-normalized rollup of the 50 highest-value rows is not publishable as a vendor ranking. Two of those rollups failed the USASpending spot-check (parent-IDV ceiling leakage on one task-order row; an exercised-options mismatch on another). Until those rows are dropped from SCOUT or explained, the public page names verified singles only.

Incumbent / PIID	Contracts	USAS exercised options	PoP end (USAS)
NTESS / Sandia (DENA0003525)	1	$55.36B	2027-04-30
CNS Y-12 / Pantex (DENA0001942)	1	$45.68B	2027-09-30
Battelle / PNNL (DEAC0576RL01830)	1	$34.75B	2027-09-30
ISS / Boeing (NAS1510000)	1	$22.44B	2026-09-30
SAIC (47QFCA21F0001)	1	$1.40B	2026-11-27

Those five awards are $159.63B of exercised-options value. Sandia, CNS, and Battelle alone are $135.79B — three DOE M&O sites, not a diversified vendor list.

A full-vehicle OASIS+ pull on the same day (44,041 task orders, $196.48B obligated) does not show a single firm above 50%. Booz Allen Hamilton leads that vehicle at $15.14B (7.7%). The published Intuitive Research 62% figure was a 97-task-order scoped pull, not the full vehicle. This refresh does not restate a >50% vehicle-concentration claim.

The point is ownership, not a leaderboard. On the awards we can show, a handful of M&O and ISS vehicles hold the dollars.

Notable contracts and clusters

Only rows that cleared the USASpending match, and only as upcoming work from 2026-08-15:

Why it is watchable	Incumbent	Agency	Exercised options	PoP end
Size; already inside two months	ISS / Boeing (NAS1510000)	NASA	$22.44B	2026-09-30
Size; inside four months	SAIC (47QFCA21F0001)	GSA	$1.40B	2026-11-27
Size; largest matched single	NTESS / Sandia (DENA0003525)	DOE	$55.36B	2027-04-30
Same-day cluster with Battelle	CNS Y-12 / Pantex (DENA0001942)	DOE	$45.68B	2027-09-30
Same-day cluster with CNS	Battelle / PNNL (DEAC0576RL01830)	DOE	$34.75B	2027-09-30

CNS and Battelle coincide on 2027-09-30 — $80.43B of matched exercised-options value on one calendar day. That is a cluster, not a claim that the sites are one procurement.

We are not restating the older “four DOE nuclear sites on 2026-09-30” line from the nuclear recompete wall here. One of those dates did not survive the spot-check on this pull. ISS / Boeing on 2026-09-30 is the matched NASA row, not a substitute for that DOE cluster.

A prior DoD piece cited a TRICARE West pair at ~$37.9B the same week. This pull’s TRICARE West match (HT940216C0002) agrees on exercised-options value within 3%, but the USASpending current PoP end is 2026-05-01 — already past as of the pull date. It is not listed above as upcoming work.

What “early” actually means

Waiting for the SAM.gov posting is late. By the time a recompete hits the feed, the incumbent has usually been positioning for a year.

Window	What a serious capture team is doing
18–24 months	Account map, teaming, past-performance gaps, customer conversations
12–18 months	Draft solution, pricing thesis, team locked
6–12 months	Color reviews, orals prep, incumbent protest watch
Posted on SAM	You are reading a document someone else helped shape

This pull is PoP end inside 18 months. Nothing in it sits in the 18–24 month zone — that work is not on this list yet. Of the matched singles, ISS / Boeing (2026-09-30) and SAIC (2026-11-27) are already inside four months. Sandia (2027-04-30) is inside nine. CNS and Battelle (2027-09-30) are the only matched large awards still in the 12–18 month band. If your thesis depends on “plenty of time,” this wall is the wrong exhibit.

How SCOUT surfaces this

SCOUT tracks period-of-performance end dates on awarded work, then ranks what is coming due. The public version of that list is the Recompete Radar. The same data is queryable over MCP (find_recompete_contracts, generate_recompete_priority_list) and inside a Pilot session.

This page is a dated snapshot. The radar is the live list.

Methodology

• Source: PrimeRFP SCOUT recompete intelligence, built from USASpending award records with a ≥$100K award-level floor unless a tighter floor is stated above.
• Publishable wall: find_recompete_contracts, within_months=18, min_value=100000000. Complete pull — 1,577 contracts, $1,195.02B estimated. Report r_b8itfh3DWJT0Xe5j.
• Do not use the unfiltered 18-month scored set (50,000 soonest-end rows, $338.54B) as the government-wide wall.
• Value: SCOUT estimated_value on this pull tracks USASpending base_exercised_options on named definitive contracts — not obligated spend. Do not mix the two in one total. The $1,195.02B wall cannot be reproduced as a single USASpending query.
• Independent check: USASpending API v2, 2026-08-15. Holds: Sandia, CNS, Battelle, ISS/Boeing NAS1510000, SAIC 47QFCA21F0001 (0–0.1% on exercised options). Rows that failed (parent-IDV leakage, exercised-options mismatch, SCOUT-vs-USAS end-date split) are withheld, not caveated into the body.
• Claims that cannot be re-run from the query log in the JSON file get cut.