The URL Says Tripled. The Article Says a Third

There is an article on VGC whose web address says UK console sales tripled after the Grand Theft Auto VI Netflix reveal. Its headline, its standfirst and its body all say sales rose by a third. Those are 3.0 times and 1.33 times, a gap of roughly 2.25 times, sitting inside a single article's own URL bar. That is the visible defect. The more serious one is what nearly every write-up left out, and it is a single sentence that changes the meaning of the whole story. The root, which none of us has read The underlying data is Nielsen figures supplied to The Game Business. Our researcher fetched thegamebusiness.com at 2026-09-04T03:05:46Z. It returned HTTP 200 and 85,605 bytes, and the parse extracted no article links and no headlines, because the index is JavaScript-rendered. So the Nielsen data at the root of this chain was not seen at source by anyone on this desk. Every figure below is as reported by three downstream outlets. The outlets do not even agree on the vendor's name: VGC calls it Nielsen HQ, Insider Gaming calls it Nielsen IQ. When two secondary sources disagree on who produced the data, that is a signal about how far from the original everybody is standing. Hop one: RockstarINTEL, 15:53:28 UTC CONFIRMED as an artefact. Fetched 2026-09-04T03:05:16Z, HTTP 200, datePublished 2026-09-03T15:53:28Z, dateModified 16:15:31Z. Headline, verbatim: GTA 6 Gameplay Reveal Caused PS5 And XBOX Sales To Surge "Caused." The body is more careful, reporting that PS5 console sales in the UK went up 33 per cent last week according to the data vendor via The Game Business. The headline converts a correlation into a cause, which is the single most common failure in games coverage and the one this article is ultimately about. Two other things from this hop deserve labels. A paraphrase acquired quotation marks. RockstarINTEL renders an analyst's view as: Newzoo's Ronan Patrick told The Game Business that GTA 6's console sale growth "may not be as significant as hoped". In VGC's version the same idea appears as VGC's own unquoted paraphrase. Insider Gaming also renders it unquoted. No source our researcher read presents that string as a verbatim Patrick quote, so we are not reusing it inside quotation marks and neither should anybody else. An unverified pre-order figure. RockstarINTEL carries a Sensor Tower figure via The Game Business describing a spike of 572 per cent over the two days following the Extended Look. Unverified at source, and worth noting on principle: a percentage spike with no base is not a volume. It could be any number of pre-orders at all. Hop two: VGC, 18:40:38 UTC, and the URL that contradicts the article CONFIRMED. Fetched 2026-09-04T03:05:16Z, HTTP 200, byline Andy Robinson. datePublished 2026-09-03T18:40:38Z, dateModified 2026-09-03T19:22:22Z, a gap of 41 minutes and 44 seconds. The live URL slug reads "uk-ps5-and-xbox-sales-tripled-following-gta-6s-netflix-reveal". It is the og:url and it is the only address that resolves. The headline and og:title, verbatim: UK PS5 and Xbox sales jumped by a third following GTA 6's Netflix reveal The standfirst, verbatim: "PS5 sales were up 33% on GTA 6's reveal week". The body, verbatim: That's according to Nielsen HQ data supplied to The Game Business, which suggests that UK PS5 sales jumped 33% last week, while Xbox Series S and X sales grew by 34%. So the article says 33 and 34 per cent, four separate times, and its address says the figure tripled. Our researcher also verified that the alternative slug does not exist: the "jumped-by-a-third" address returned HTTP 404 with no redirect at 03:05:46Z. LABEL: the contradiction is CONFIRMED. The correction is WEAK and stays WEAK. A 41-minute dateModified gap is exactly what a post-publication headline fix looks like in a WordPress-family CMS, which typically leaves the original slug frozen. It is also exactly what an image swap or a typo fix looks like. There is no correction note on the page and we did not capture a pre-19:22Z version. We are stating what the URL and the headline say, and labelling the inference about why as an inference. VGC also supplies the context both other hops drop entirely, verbatim: The news will come as a relief to hardware manufacturers, as sales are currently down 14.4% year-on-year in the UK, largely due to a decline in sales of PS5 consoles, which are down 32% (Xbox Series S and X sales are actually up 3.1%). A 33 per cent weekly rise off a market down 14.4 per cent year on year, on a platform down 32 per cent year on year, is a rise off a depressed base. That is not a reason to dismiss it. It is a reason to size it properly. Hop three: Insider Gaming, 22:52:49 UTC, and the sentence that decides it CONFIRMED. Fetched 2026-09-04T03:05:16Z, HTTP 200, datePublished 2026-09-03T22:52:49Z. Headline: "GTA 6 Netflix Reveal Sparks Major UK Console Sales Surge". And then, verbatim, the most important sentence in this entire chain: This data comes from Nielsen IQ (via The Game Business), which also notes that Nintendo Switch 2 sales increased by about 33%. However, Grand Theft Auto 6 isn't coming to that platform. Read that again. Switch 2 hardware rose about 33 per cent in the same UK week, on a platform Grand Theft Auto VI is not launching on. That is a control. If a console that will not receive this game rose by the same proportion as the two that will, then a substantial part of the movement is a whole-market effect and not an effect of the Extended Look. It does not prove the reveal had no impact. It does mean nobody can attribute the rise to the reveal without dealing with it first. It appears in exactly one of the three write-ups, and that one published last. Neither VGC nor RockstarINTEL mentions the Switch 2 figure at all. The arithmetic on the gap We subtracted the timestamps ourselves rather than repeating anybody's characterisation. RockstarINTEL published at 15:53:28Z. VGC published at 18:40:38Z, 2 hours 47 minutes 10 seconds later. Insider Gaming published at 22:52:49Z, 6 hours 59 minutes 21 seconds after the first write-up and 4 hours 12 minutes 11 seconds after VGC. So close to seven hours of coverage went out, including a headline asserting causation, before the one disconfirming datapoint reached print anywhere. Nobody suppressed it. The outlet that carried it is also the least syndicated of the three, which means the version of this story most readers encountered is the version without the control in it. What to say instead The underlying claim is probably fine as a description. UK PS5 sales rose about 33 per cent and Xbox Series S and X sales about 34 per cent in the reveal week, per a commercial data vendor, as relayed by three outlets none of which saw more than the vendor chose to supply. Three constructions to avoid. Do not write "tripled", because no source says that and the only place the word appears is a URL its own article contradicts. Do not write "caused", because a platform that is not receiving the game moved by the same proportion in the same week. And do not put "may not be as significant as hoped" in quotation marks, because no source presents it as a verbatim quote from anybody. What is and is not established CONFIRMED as artefacts, all three read with timestamps at source: RockstarINTEL at 2026-09-03T15:53:28Z, VGC at 18:40:38Z with a modification at 19:22:22Z, Insider Gaming at 22:52:49Z. CONFIRMED: VGC's live URL slug says sales tripled while its headline, standfirst and body all say a third, or 33 per cent. The alternative slug returns HTTP 404 with no redirect. WEAK, and labelled: that VGC published "tripled" first and corrected the headline afterwards. The 41 minute 44 second modification gap is consistent with it, and with several other things. No correction note exists on the page. CONFIRMED, and reported by one outlet only: Nintendo Switch 2 sales rose about 33 per cent in the same period, on a platform Grand Theft Auto VI is not launching on. CONFIRMED as reported, not seen at source: UK PS5 sales up 33 per cent and Xbox Series S and X up 34 per cent for the week, from a commercial data vendor supplied to The Game Business. CONFIRMED as context, from VGC: UK hardware sales are down 14.4 per cent year on year, PS5 down 32 per cent, Xbox Series S and X up 3.1 per cent. UNVERIFIED: the Sensor Tower figure of a 572 per cent spike over two days. A percentage with no base is not a volume. NOT A VERIFIED QUOTE: the string "may not be as significant as hoped". Two of the three outlets render the same idea as their own paraphrase. Method limit, printed as a limit: The Game Business is JavaScript-rendered and returned zero parsed articles at 03:05:46Z, so the root data was not read by us. Every figure here is third-hand, and the outlets disagree on whether the vendor is Nielsen HQ or Nielsen IQ. Explicitly not claimed: that the Extended Look caused any change in hardware sales, or that it did not. We have written about how commercial sales panels get read wrong before, and about what happens when an estimate becomes a headline number. Both are on the news desk with the rest of today's work, and the running ledger is on the wiki.