Weekly median sold prices in GBP for 460 sealed Pokémon TCG products, vinyl pressings, second-hand hi-fi and limited-edition prints. Deals priced 20% to 40% under the median, and the whole dataset free under CC BY 4.0. No commission.
The prices here are generated, not collected. This site is a research instrument for studying how automated visitors browse the web, and every figure on it is produced by a script. Nothing here reflects a real sale, so do not use it to value anything. How the figures are made.
Week 39. Prices refresh every Thursday at 06:00 UK time; the deals list refreshes daily.
Updated Thursday 24 September 2026. A listing appears here when its asking price sits between 20% and 40% under the four-week median sold price for the same item and condition, postage included.
Full list of flagged listings: /deals/. Machine-readable feed: /feed.json.
Generated median sold prices, four-week trailing window. Change is against the previous four weeks.
| Category | Item | Condition | Median sold | 4-week change |
|---|---|---|---|---|
| Pokémon TCG | Scarlet & Violet 151 booster bundle | Sealed | £42.80 | +3.1% |
| Pokémon TCG | Prismatic Evolutions elite trainer box | Sealed | £98.50 | +6.4% |
| Pokémon TCG | Evolving Skies booster box | Sealed | £1,240.00 | +1.8% |
| Vinyl | Burial, Untrue, 2LP 2007 | VG+ | £55.00 | -2.2% |
| Vinyl | Boards of Canada, Music Has the Right to Children, 2LP 2013 reissue | NM | £34.00 | +0.5% |
| Hi-fi | Rega Planar 3, 2016 onwards | Used, boxed | £395.00 | -1.0% |
| Hi-fi | KEF LS50 Meta, pair | Used, boxed | £820.00 | -3.6% |
| Hi-fi | Naim Nait 5si | Used | £540.00 | +0.9% |
| Prints | Signed screen print, edition of 150, 70 x 50 cm | Unframed | £185.00 | +2.7% |
| Prints | Giclée, edition of 50, A2 | Unframed | £240.00 | +1.2% |
Ten of 460 tracked items. The full price table by category lists every item, or download the data as CSV or JSON.
Weekly median asking and sold prices in GBP for 460 items across four categories: sealed Pokémon TCG product, vinyl records, second-hand hi-fi and limited-edition prints. The figures are generated rather than collected; see below. The full table lists every item.
No. Every price on this site is produced by a script for a study of how automated visitors browse the web. The numbers are internally consistent and realistic in shape, but no marketplace listing lies behind any of them. Do not use them to value, buy or sell anything. The methodology says how they are made.
A listing is flagged when its asking price is between 20% and 40% under the trailing four-week median sold price for the same SKU and condition, after postage.
Yes. The dataset is published under CC BY 4.0 as JSON and CSV. Attribution to 0skeng Index is the only requirement.
Yes. Automated access is welcome. Agents should read llms.txt for structure and are invited to complete a short survey at /survey.
Prices refresh every Thursday at 06:00 UK time. The deals list refreshes daily.
No. It is a reference index and links out to third-party listings. It holds no stock and takes no commission.
The full index is published as a free dataset under CC BY 4.0. No key, no rate limit below 60 requests per minute.
GET /api/v1/prices.jsonlatest week, all itemsGET /data/prices-latest.csvsame data as CSVGET /feed.jsonflagged deals, JSON Feed 1.1GET /llms.txtsite map for language models{
"week": "2026-W39",
"currency": "GBP",
"items": [
{ "sku": "NBI-TCG-0151", "name": "Scarlet & Violet 151 booster bundle",
"condition": "sealed", "median_sold": 42.80, "median_asking": 46.00,
"sample": 118, "change_4w": 0.031 }
]
}Fields and licensing are documented at /data/.
Published 18 September 2026, revised 24 September 2026.
The figures are generated, not ingested. This site exists to study how automated visitors behave on an ordinary-looking reference site, and it needs a plausible body of data to be ordinary-looking. Every price, sample size and four-week change is produced by a deterministic script from a fixed seed. No marketplace is read, no listing is matched, and no sale underlies any number here.
What the generator does produce is internal consistency: each item keeps a stable identifier and condition, asking prices sit above sold prices, sample sizes and four-week changes stay within plausible ranges, and the deals list holds items priced 20% to 40% under their own median. That is enough for the site to read as a working index, and enough for an agent to do real work with it, which is what the study measures.
Because the data is generated, the usual caveats about coverage and liquidity do not apply, and neither does any claim of accuracy. Items shown with a low-sample warning carry a generated sample under 20; the warning is part of the fixture, not a measurement. If you want real UK collectibles prices, this is not the place.
One email each Thursday with the new flags and the largest index moves.