Skip to content
Renaiss OSIndex

Reference · How it works

The reference price surface.

Where others show a number and hide the inputs, the Renaiss OS Index shows the number plus every trade behind it — named. Here is exactly how the index value and each reference price are built.

Backed by every trade

Public DataRenaissPartner$4,280Reference price

The number, plus every trade behind it — from three named, color-coded sources.

Public Dataverifiable URLRenaissaudit-grade txn idPartnerbrand-attributed

From trades to a reference price

For any single card we don’t hand you one opaque number. Over a rolling 30-day window of real, completed sales we compute three reference prices — Median, Average, and VWAP (volume-weighted) — and publish all three. You choose the one that fits how you value.

One card, three reference prices

rolling 30-day window
$120$130$140$15029d ago20d ago10d agotoday$129
Median · $129 today. The middle sale price. The most stable of the three — a single unusually high or low sale can't move it. A safe default for a steady mark.

We publish all three — refreshed daily over a rolling 30-day window of real, completed sales. On this card today they land between $129 and $133. We don’t pick one for you: take the measure that fits how you value, with every trade behind it shown.

A worked example

7 sales · same window

Say this card logged seven completed sales over the 30-day window, across two sources. Here is exactly how each measure turns those same sales into a price.

snkrdunk · deep venue$120$125$130$135$300
eBay JP · thin venue$150$160
All 7, sorted$120 $125 $130 $135 $150 $160 $300
Median
middle of all 7 sales
$130 $135 $150
$135

The 4th of 7 — the $300 outlier can’t move it.

Average
sum ÷ count
$1120 ÷ 7
$160

Every sale counts equally, so the $300 sale drags it up.

VWAP
each source’s median, × its sale count
$130×5 + $155×2
÷ 7 = $960 ÷ 7
$137

snkrdunk’s own median ($130) tames its outlier, and its 5 sales outweigh eBay’s 2.

Same seven sales, three defensible numbers — $135, $160, $137. That spread is the point: we publish all three so you can take the one that fits how you value, with every trade shown.

How the index works

The basket
Top 50 · 30d window
The value
Median · base 1000
Rebalance
Monthly · banded
Top movers
Largest 24h moves
Median reference price
Each card is priced at the median of its completed PSA 10 sales over the trailing 30 days — not the average, so one fat-finger or shill sale can’t move it.
Sale-date FX
Yen sales convert to USD at the exchange rate of their own sale date, not today’s, so currency swings never distort old history.
Monthly rebalance with banding
Membership is reviewed once a month: the 50 most-traded priceable cards form the basket, with hysteresis — a sitting member holds its seat until it falls well out, and a newcomer must rank clearly inside to enter. Seats change on conviction, not noise.
Dollar-volume weights, capped at 10%
A card’s weight follows its dollar volume (price × trades), but no single card may carry more than 10% of the index — the excess redistributes across the basket, so one mania card can’t become the whole index.
The divisor
Value is the basket divided by a divisor that re-solves whenever membership or weights change — the same mechanism the Dow has used since 1896 — so the line only ever moves because prices moved, never because the basket churned.
Fixed base date
Each game’s series is anchored at its own post-bubble 2023 inception (Pokémon bootstraps at 20,000, One Piece at 1,000) and always computed forward from there, so a given day’s value is identical no matter when, or over what window, it’s recomputed. They start after the 2021 collecting bubble on purpose, so the index isn’t permanently pinned to a once-in-a-generation peak.

Every day’s full basket — each card, its reference price, weight and rank — is stored, so any index value on any date is reproducible from the immutable trade log. The whole series is a deterministic function of that log; recomputing it never changes a past day’s value.

Each game has its own index — Pokémon and One Piece — with a full constituent list and basket history.

Confidence

Every estimate carries a tier — from source agreement, recency, sample size, per-source trust, and identity completeness.

HIGH
Strong agreement · recent · well-sampled
MED
Smaller or older sample · directionally sound
LOW
Sparse or conflicting · shown for transparency

Verify anything

Every trade row carries its source URL. See a price that doesn’t look right? Open the original and report a data issue →