The Spike Report · Pokémon

Data cutoff: Sunday, 4 October 2026, end of day. Distributed: Tuesday, 6 October 2026. Source: JustTCG.

A standing section from this edition on. Every prediction this report makes gets written down and settled here the following week, whether it landed or not.

We said the Dusknoir archetype would show up in this edition if it moved anything we track. It cannot, and that is the finding.

Alakazam-Dusknoir took first and second at Frankfurt and Dusknoir appeared in six of the sixteen combined top-eight slots across both regionals. We said that if the archetype moved any card in our coverage, edition seven would report it.

There is no Dusknoir card in our coverage. There is no Lopunny card either. The only cards from those top-eight archetypes we track at all are two Alakazam prints from Fates Collide, and they moved -1.2% and -0.5% against a board median of -0.33%.

So we did not make a prediction that failed. We made a prediction about a population we cannot observe, and then spent a week waiting to measure it. That is worse than being wrong, because being wrong is informative. Before we write down another tournament prediction we will check that the archetype’s cards are in the universe first.

We said in edition five that we were holding a card off the board at +8.4%. Here is what happened to it.

Groudon EX (150 Full Art), Primal Clash. The sequence, in full.

The +8.4% we declined to print has been given back in full and then some. The card is now $171.08, below where it started on 8 September, and it appears on this week’s board at -8.5%, number seven. A single unconfirmed step, published as a gainer, would have been the worst call this report has made. The rule caught it.

We will take the other side of our own rule too, because it is the same card. The rule required four distinct prices after the 8 September step before this card could return to the board. It has now recorded five, but across twenty-six days, and it changed price on only five of the last thirty days against a universe median of eleven. Five prices in twenty-six days satisfies our test and does not describe an actively traded card. It is on the board this week partly because the clearing rule counts prices and never asks how long they took to arrive, which is a gap we put there ourselves.

Edition six’s three Early Signal cards, one week on. Mega Zygarde ex -3.2%, Greninja ex -2.2%, Team Rocket’s Mewtwo ex +1.4%. Two of three continued in the direction the pattern describes and one reversed. That is the whole record so far for cards named in that section and then re-measured a week later: three cards, two continued, one reversed. Far too small a sample to mean anything yet, and it will be printed here every week until it is not.

And edition six’s whole board, one week on. Nine of the ten movers decelerated or reversed.

One card out of ten accelerated. If you read this board as momentum, that table is the argument against it. The board ranks the size of a move that already happened; it is not a forecast. A week later seven of these ten were moving at less than two thirds of the previous week’s pace, one of them — Pokemon Center Lady, from -10.4% to -0.4% — at under a twentieth, and one had changed direction outright.

The decline did not rotate. It broadened.

Last week we wrote that selling had moved out of Mega Evolution and into Scarlet & Violet, and that the shift was the story. One week later that framing looks like a one-week artifact and we would not write it again.

One note on the numbers before the table. Our era, set, age and breadth tables were missing two filters that the board has always had, which we found and fixed this week. Every figure below is on the corrected basis, this week and last, so the “last week” column will not match what edition six printed. The full accounting — six corrections across editions one to six, including a 1.46-point error on edition five’s headline era figure — is published separately today as The Spike Report · Corrections. Nothing on any board we have published is affected.

Mega Evolution did not hand the decline on. It got worse on both measures, and at 91% its breadth is the widest we have recorded for any era in any edition — not a return to a previous level but a new one, against 83% in edition five and 74% in edition six. Scarlet & Violet barely moved, 82% to 80%. Sword & Shield barely moved.

What changed is XY, which went from 27% of its cards declining to 52%. XY has been at or near the healthy end of every table in this report. It is now at a coin flip. Sun & Moon at 25% is the only era still clearly holding, and on the corrected basis it was the healthiest era in editions five and six as well, at 16% and 28% declining against XY’s 37% and 27%. So the honest version is that XY was never quite the safest place. It was the second safest, and now it is not safe at all.

The age gradient held its shape and lost its refuge.

Rank correlation between set age and weekly change is +0.445, against +0.424 last week. Four of the five buckets got worse. The cliff at four years is still there: everything younger is declining at 72.7% or more, everything older at 53.8% or 43.0%.

But look at the bottom row, which is the one that matters. The over-8-year bucket has been the refuge in every edition that has printed this table, the group that was flat on the week and up over a month while everything newer fell. Its share of declining cards went from 31.8% to 43.0%, and its 30-day median went from +1.17% to +0.13%. It is still the best place in our data to be. It is no longer a place where prices rise.

That is the same event as the XY row above, seen from a different angle, and it is this edition’s finding.

The same caution applies as last week. Age and era are the same variable in our coverage, so “old cards are holding up better” and “XY and Sun & Moon are holding up better” are one observation, not two.

Set detail. Chaos Rising is the worst set we track at -7.24% on an equal-weight average and -6.53% on a median, five of five down. Journey Together -6.33% and -5.93%, five of five down, and it put two cards on the board. Prismatic Evolutions -4.89%, five of five down. At the other end Celestial Storm finished +3.94% and +2.83% with one of five down, and it holds this week’s only gainer. Roaring Skies +2.08%, XY base +2.03%, Lost Thunder +1.82%.

Breadth: the median card moved -0.327% across 293 priced chase constituents, against -0.153% last week on the same basis. 59% declining, up from 53%. Seven cards up 5% or more, twenty-nine down 5% or more.

We ran a test this week designed to fail, because a report that only publishes the patterns it finds is not measuring anything.

30th Celebration’s own Gengar ex (154) fell -36.9% in seven days. It is excluded from everything above, because the set is eighteen days old and our floor is ninety, but a 37% fall in a famous card’s anniversary reprint is the kind of thing that ought to drag its other prints with it if anything does.

It did not. The ten Gengar cards in our measured coverage ran a median of -0.10% with 50% declining, against the board’s 59%. Mega Gengar ex was the worst at -5.8%, Gengar & Mimikyu GX (Secret) the best at +4.5%, and Gengar EX (114 Full Art) at $765 did not move at all. There is no Gengar story. The name does not travel.

We are reporting that because we went looking and found nothing, and the next time we report that a theme moved together, this is the comparison it has to beat.

Eight cards cleared every test this week except the $10 floor. Two of them are the steepest monthly declines in the entire dataset.

Nothing above $10 in our coverage fell further over thirty days than those two Pikachu VMAX prints. Both repriced on twenty-nine or thirty of thirty days, which makes them among the most actively priced cards we hold. They are invisible to this report because they are cheap.

The third row is the one that should bother you. Rayquaza VMAX (101) is nine cents under the floor. Its sibling print, Rayquaza VMAX (102), was number four on last week’s board. Same card, same set, same week, one of them published and one of them not, separated by nine cents of price.

The floor exists so the board is not permanently filled by $11 cards moving thirty cents, and we are not removing it this week. But it is a threshold, not a measurement, and this is what it costs.

One list. Gainers and decliners ranked together by the size of the move.

Seven day window, 27 September to 4 October 2026. Prices are ungraded Near Mint. 97 of 350 tracked cards qualified and 87 cleared the guard. No qualifying card was pulled off the board by the review rule this week.

One thing to notice before the card notes. The two most dramatic entries on this board are the thinnest cards on it. Rayquaza GX at +8.5% repriced seven times in thirty days. Groudon EX at -8.5% repriced five. The universe median is eleven. Meanwhile the biggest decliners are the most actively priced cards on the board: Mega Clefable and Pikachu (Secret) both at thirty of thirty, Hop’s Zacian and Mega Greninja at twenty-nine. Size of move and quality of evidence are running in opposite directions this week, and the card notes say which is which.

Context for each of the ten. Where the shape of the week matters, the daily closing price is printed.