The clock nobody watches

And yet the price moves, sometimes sharply, in a month where nothing observable happened. The usual explanations are sentiment and speculation, which are labels rather than mechanisms. It is more useful to separate the processes that are actually running, because they run at very different speeds and respond to entirely different things.

Matheson's study of collectible prices around the deaths of Hall of Famers documents what he calls a nostalgia spike: heightened interest following the death and the coverage of it, larger for more famous players[1]. The effect is real, it is short-lived, and it is not something anyone can position for in advance without being ghoulish about it.

Hall of Fame induction is more interesting, because the evidence suggests it is already priced. Prices begin drifting up one to two years before the vote, and the induction itself adds little for a player whose election was expected[5]. That is a market doing exactly what an efficient one should: incorporating a foreseeable event before it occurs. It also means induction is not a trade, and treating it as one is buying the news after everyone else has.

This clock explains the shape of the 2020–21 boom better than any account of speculation does. Card sales on eBay rose 142% in the boom year, with basketball up 373% and soccer up 1,586%[4]. A market does not do that because a few people got excited; it does that when a large cohort arrives at once, with unusual amounts of time and, briefly, unusual amounts of disposable income.

The demand clock is the reason to be structurally interested in this category and, as the companion paper argues, it is measurable and slow. It is also the clock that eventually stops: the cohort that wanted antique furniture aged out, and the furniture did not recover.

A share count does not rise because the share price rose. A certified population does.

In an equity market, supply is fixed between corporate actions, and those actions are announced. In this market, supply is a flow that responds to price. When a card's value rises, grading a copy becomes worth the fee; raw copies come out of collections; and some months later the population report prints a larger number. The card is now competing with more of itself.

The crucial point is that no new cards need to be manufactured for certified supply to expand. Grading converts an ungraded card, which is nearly untradeable at the top of the market because condition is arguable, into a standardised, describable, tradeable unit. It is supply creation by reclassification. And it is running hard: 26.8 million cards were certified in 2025, up 32% on the year[2].

Two features make this clock hard to see. The response is lagged. Months pass between the price move and the census that records its consequence, and the census itself publishes annually, so by the time the number is visible the supply has already arrived. The wave is observable in the price record before it is observable in the population record.

Between roughly 1987 and 1994, the junk wax era, manufacturers responded to booming demand by tripling output. Topps alone is estimated to have printed over a billion cards in 1986[3]. Demand did not fall. It kept rising for years. The vast majority of what was printed in that window is now worth cents, and collectors who bought on the assumption that popularity implied value were left holding thousands of near-worthless cards.

The lesson is specific and it is not "bubbles pop". It is that in a market where supply can respond, demand alone does not determine outcomes, and that the response can overshoot by an enormous margin, because the people producing supply are looking at the same rising prices everyone else is.

Modern and ultra-modern cards fell somewhere between 30% and 50% from their peaks, with tracked modern down over 30% in 2023 alone. Vintage material in quality grades, meaning the Ruths and Mantles and Mayses, held with reports of no meaningful pullback at the top end[4].

The distinguishing variable is not age, prestige or nostalgia. It is whether supply could respond. A 2021 rookie card can be printed again, and more copies can be pulled and graded. A 1952 Mantle in a high grade cannot be manufactured, and the number of raw copies still capable of grading well is small and shrinking. Same demand shock, same correction, opposite outcomes, because one side of the market could answer and the other could not.

Any statement about "the card market" in 2025 is therefore wrong about roughly half of it. This is not a subtlety to be noted and moved past. It determines whether a rising index is evidence of a healthy market or of a supply expansion in one half masking a contraction in the other.

1. 1.Ask whether supply can respond at all. A card with a large ungraded population behind it is a different asset from one where almost every surviving copy is already certified. Population at grade and population above grade are the two observable proxies.
2. 2.Look at the price move a submission lag ago, not the current census. The census reports what already arrived. The prior year's price move is what predicts what is still coming.
3. 3.Separate the response from plain mean reversion. This is where most naive versions of the argument fail: a price rise is followed by weaker returns for two quite different reasons. An unusually high sale reverts, and a price rise provokes submissions, and on a single card the two are indistinguishable. Reversion does not care how many ungraded copies exist. Submissions do.
4. 4.Expect the effect to be a drag, not a signal. The supply clock tells you what will weigh on a price, not when to buy. Used as a buy signal it will mostly be measuring reversion and calling it something grander.

The category's worst outcome was a supply shock that nobody was measuring, in a market that was, at the time, extremely popular. Certified supply grew by a third last year. That seems worth watching.