Google News Top News rank_change climbers via get_top_trends

Google News rank_change on Trends MCP is get_top_trends with type Google News Top News, sort rank_change, and a window such as 1d. From the August 27, 2026 pull, the board stamped 2026-08-27T20:01:33 UTC. The 1d window used baseline 2026-08-27T02:01:32 UTC, a snapshot from the same calendar day, and returned count 0 with a held-or-fell note. Windows 3d, 7d, 14d, and 30d also returned count 0, each with a complete-turnover note. Default sort rank filled 38 headlines, then count 0 at offset 38. Rank 1 was Trump directs federal government to rename Lake Ontario to Lake America, NPR. Wikipedia Trending 1d on the same stamp returned count 42. Free tier access is 100 requests per month.

News 1d returned 0 climbers on a same-day 02:01 baseline

From the Trends MCP scan of August 27, 2026, Google News sort rank_change stamped 2026-08-27T20:01:33 UTC and filled 0 climber rows on window 1d against baseline 2026-08-27T02:01:32 UTC, so a 1d News pull can diff two scans on the same date and still return an empty list with a held-or-fell note.

Default sort rank on that stamp returned 38 headlines at limit 50, then count 0 at offset 38. Rank 1 was the Lake Ontario rename. Rank 2 was Nepal flash floods, The New York Times. Rank 3 was a Kennedy Center demolition threat. Rank 5 was Iran attacks another commercial ship in the Strait of Hormuz. Rank 7 was the Lindsay Clancy trial. Rank 11 was Ratko Mladic. Rank 15 was Nvidia and tech stocks. Rank 19 was GTA 6 Netflix. Rank 23 was Peter Cullen. Rank 24 was Yayoi Kusama. Rank 26 was Tim Curry. Rank 31 was a partial lunar eclipse. Rank 38 was a Frankfurt airport malaria outbreak. Length 38 is the durable cap from this scan. The rank 1 headline is not.

The Google News Top News API page still describes a July 30, 2026 scan with a Trump and Iran headline at rank 1. This pull closed at 38 with Lake Ontario at rank 1. Length 38 held. The leaderboard did not.

Same-stamp Wikipedia Trending 1d used that 02:01 baseline and returned count 42, led by 2026-27 UEFA Europa League at rank 78, prev_rank 179, rank_change 101. Google Trends 1d returned count 22, led by gta 6 netflix at rank 34, prev_rank 159, rank_change 125, then mica miller at rank 37, prev_rank 142, rank_change 105, then hurricane dolly at rank 72, prev_rank 156, rank_change 84. lake ontario sat on that Google Trends climber list at rank 14, prev_rank 78, rank_change 64, matching News rank 1 by topic, not by string. X Twitter Trending 1d returned count 1: Lucia at rank 10, prev_rank 99, rank_change 89. Reddit Hot Posts 1d returned count 0 with a held-or-fell note versus a 00:01 baseline, not the 02:01 News baseline. App Store Top Free 1d returned count 35, led by Netflix at rank 43, prev_rank 90, rank_change 47. ChatGPT sat at App Store Top Free rank 1 and was absent from News 38.

An empty 1d News list is not a missing feed. The current-rank board still has 38 rows. rank_change only lists headlines that climbed versus the stored snapshot.

Exact API calls for climber boards

sort rank_change plus window is a separate get_top_trends request from default sort rank, so a 1d climber pull and a current-rank pull on August 27, 2026 are two requests, and each extra window is another request because window is per call.

{
  "mode": "get_top_trends",
  "type": "Google News Top News",
  "sort": "rank_change",
  "window": "1d",
  "limit": 50
}

{
  "mode": "get_top_trends",
  "type": "Google News Top News",
  "sort": "rank_change",
  "window": "7d",
  "limit": 50
}

{
  "mode": "get_top_trends",
  "type": "Google News Top News",
  "limit": 50,
  "offset": 0
}

MCP clients send the same fields on get_top_trends. The type string is exact: Google News Top News is correct; Google News fails. Window values are 1d, 3d, 7d, 14d, and 30d. Default sort is rank. Default window when sort is rank_change is 30d, so omit window only when that 30d lookback is the intent. On this stamp the 30d call returned count 0 with a complete-turnover note, not a climber list. Offset and limit behavior is on get_top_trends pagination. For the Google Trends sibling, see Google Trends rank_change. For a current-rank join against Wikipedia, see Google News vs Wikipedia Trending.

Official Google News RSS and publisher homepages do not emit prev_rank against a stored Trends MCP snapshot. Those products are not substitutes.

3d through 30d turned over while 1d held or fell

A longer rank_change window on Google News can return the same count 0 as 1d with a different note, because held-or-fell means every current headline was already on the baseline board at the same rank or lower, and complete turnover means no current headline appears in the older snapshot at all.

Every window on this stamp used the 02:01 cluster except the current as_of, which sat at 20:01. 1d compared 20:01 against 02:01 on August 27, about eighteen hours, not a full calendar day. 3d through 30d found zero shared headlines, so rank_change cannot be computed. Count 0 is not one error. Read the note before retrying a shorter window or switching to sort rank.

The sibling Google News vs Wikipedia Trending page recorded News 1d count 1 at 14:01 on the same calendar day, led by Ottumwa Pizza Ranch at +1. This 20:01 call returned count 0. A News 1d fill can empty within hours while length 38 on sort rank stays put. Lake Ontario at rank 1 never entered a climber object on this stamp. Rank 1 can sit outside the climber list when it held, fell, or was missing from the baseline snapshot.

Live News ranks and search indexes can disagree

google news on get_growth is a weekly 0-100 index for a keyword, not a live headline rank. Follow-ups use google news, google search, or youtube on a related keyword, each billed as its own request, and a high News rank can sit next to a flat or falling index whose recent_date is still last week.

On the 2026-08-22 google news points, lake ontario sat at 75.0, down 20.21% over 7 days from 94.0 and up 275.0% over 30 days from 20.0. That weekly 75.0 sat next to News rank 1 and X rank 1 on the 20:01 stamp. nepal sat at 25.0, down 30.56% over 7 days from 36.0 and up 108.33% over 30 days from 12.0, next to News rank 2. nvidia sat at 87.0, down 13.0% over 7 days from 100.0 and up 480.0% over 30 days from 15.0, next to News rank 15. lindsay clancy sat at 100.0, up 51.52% over 7 days from 66.0, with a low-quality warning; the 30D percent was omitted because baseline_value was 0.0. ratko mladic sat at 5.0 versus a 0.0 baseline on both windows, also low-quality; quote 5.0, not the dummy percent. peter cullen sat at 0.0 on google news for both windows, with the same low-quality warning, next to News rank 23. youtube for peter cullen sat at 3.0, up 50.0% over 7 days from 2.0 and down 25.0% over 30 days from 4.0. google search for peter cullen sat at 0.0 on both windows. gta 6 on google search sat at 37.0, up 164.29% over 7 days from 14.0. gta 6 netflix sat at 15.0, up 50.0% over 7 days from 10.0; the 30D percent was omitted because baseline_value was 0.0 and the series carried a low-quality warning. lucia, the only X 1d climber, sat at 48.0 on google search, down 4.0% over 7 days from 50.0. wikipedia source get_growth for 2026 Nepal floods returned no_data, even though that title sat at Wikipedia rank 38 beside News rank 2.

Live rank is not a get_growth guarantee. Quote both values. A News rank 1 headline about Lake Ontario is not the same story as a weekly google news index falling from 94.0 to 75.0.

Unofficial Google News scrapers still advertise a current headline page. Trends MCP rank_change is a snapshot diff with prev_rank, not a click-count field from that HTML. The August 27, 2026 20:01 list had zero climber rows on every window. Publish that zero. Do not pad to 25.