Suffrago UK Data Hub
RankingsVerificationMethods

Independent service run by Suffrago – not an official government website. Every figure comes from an official UK publisher under its open licence and is checked against that source each time the data is refreshed.

Government defeats in the House of Lords by session

Government defeats in whipped divisions of the House of Lords, with the divisions held, in each session of the UK Parliament since 1999–2000, the current session to the day before retrieval.

Government defeats in the 2026–27 session to date

15

29 Oct 2026 · in 35 whipped divisions

Verified against source, 30 October 2026, 11:49 UTC
Government defeats in the House of Lords by session
  • Government defeats
  • Divisions held
0501001502002501999–20002003–042007–082012–132016–172021–222026–27 to date

Source: UK Parliament. Contains Parliamentary information licensed under the Open Parliament Licence v3.0.

Data table

25 rows

CategoryGovernment defeatsDivisions held
1999–200036192
2000–01240
2001–0256172
2002–0388226
2003–0465176
2004–053767
2005–0662192
2006–0745104
2007–0829125
2008–092589
2009–101443
2010–1248241
2012–132782
2013–141489
2014–151160
2015–1660114
2016–173877
2017–1969138
201911
2019–21114173
2021–22128210
2022–23126186
2023–244363
2024–26208409
2026–27 to date1536

Download CSVDownload JSON with full provenance

Source and licence

Publisher
UK Parliament
Check it yourself
https://votes.parliament.uk/votes/lords
Licence
Open Parliament Licence v3.0 – Contains Parliamentary information licensed under the Open Parliament Licence v3.0.
Retrieved
30 October 2026, 11:41 UTC
Raw files fetched

Verification

Verified against source, 30 October 2026, 11:49 UTC Run at 30 October 2026, 11:49 UTC; 50 of 50 figures matched the live source.

ResultCheckDetail
PassProvenance recordedUK Parliament; Open Parliament Licence v3.0; retrieved 2026-10-30T11:41:28Z; 35 raw file(s)
PassRaw files match recorded SHA-25635 of 35 files intact
PassRe-parsing raw files reproduces figures50 of 50 figures and the headline reproduced
PassFigures are finite numbers50 figures, 0 invalid
PassLive source matches every figure50 of 50 figures match exactly after downloading 35 source file(s) again (1 byte-identical); the headline matches
PassDefeat flags agree with the counts3105 of 3105 whipped divisions' government-win flags agree with the content and not-content counts and the government's side; 22 tied, left out
PassHansard's division search agrees2448 of 2449 divisions in Hansard's search (https://hansard-api.parliament.uk/search/divisions.json, house=Lords) are in the votes API, by sitting day and number or else by their counts, 2445 on the same day and 2400 with the same counts; in earlier sessions Hansard lists 1 the API lacks: 2016–17 No. 1 on 2016-11-09; Hansard's search lacks 857 of the API's 3305, the first 1 on 1999-11-24 and the last 2 on 2005-03-15; in earlier sessions 1 differ on which side won: 2005–06 No. 1 on 2006-06-12: Hansard 148–147, the votes API 147–147
PassParliament's record is currentthe newest division is dated 2026-10-28, 2 days before retrieval (limit 80)
Not runSource page answershttps://votes.parliament.uk/votes/lords refuses automated requests (HTTP 403); open it in a browser to compare

Notes

  • A defeat is a whipped division the government lost, as the Lords Votes API flags it; free votes are left out. An inquorate division has no result and is not counted as a defeat. Division numbers restart each sitting day, and sessions are as Parliament's What's On service lists them. The Lords Votes API's records start on the 24th of November 1999, a week into the 1999–2000 session.
  • There is no second route to the defeat count itself: parliament.uk's own table of defeats refuses scripts. Each division Hansard's search lists is matched instead, and each defeat flag with the vote counts. Hansard's search does not list the Lords' earlier divisions.
  • The stored raw files are the hub's copy of the Lords Votes API's records with peers' names and ids removed: the lists of peers who voted and of the tellers, the sponsoring peer, the titles and the free-text notes, all of which can name peers, are dropped before the files are written. Every run fetches the API in full again to check the figures.