This platform exists to replace hobby folklore with measured answers, and after a year of publishing those measurements, some rules repeat so often they deserve a starting page. None of these are opinions. Every one links to the study that produced it, and every study shows its work.
Ten rules, each one paid for in data
1. Never pay launch prices. Across fourteen Scarlet & Violet sets, chase cards settle roughly 90 days after release once the box-opening wave passes, and the cards that settled expensive went on to lose. If you remember one number, remember that recent Special Illustration Rare classes settled at $58 to $101 medians and sit negative today.
2. Advertised scarcity is pre-paid; quiet demand is not. Our SIR versus IR study of 403 cards found the rarer class returned a median of minus 7% since settle while the common Illustration Rares returned plus 76%, and the gap survives price-band controls. The market overpays for scarcity it can see and underpays for the $6 card a million people quietly want.
3. Respect the print clock. The single cleanest divide in the market is whether a set is still being printed. Our weekly Spread report tracks closed-print sets at more than double the trajectory of in-print product over the past year. While a set is in print, supply keeps arriving; when the presses stop, the clock starts.
4. A PSA 10 is only as scarce as its gem rate. Cards that gem above 80% of the time carried a median 2.2x premium over raw as of August 2026; cards that gem under half the time carried 4.2x. Before paying any slab premium, check the population page: a 10 that almost every submission achieves is a certificate, not a rarity.
5. Grading is a trade with a losing side. Our grading study priced both outcomes: on 106 of 219 chase cards, a PSA 9 sells for less than the raw card costs, and vintage carries positive expected value at a 9 while modern often doesn't. Run the math (gem rate times the 10, miss rate times the 9, minus fees) before sending anything.
6. Price from sold comps, never from asks. A listed price is an opinion; a sold price is a fact. Our entire pricing layer is medians of individual sold listings with lots, reprints, autographs, and wrong-set matches filtered out, and our audits still catch contamination. If a number looks too good, check what actually sold.
7. Populations only go up. PSA prints hundreds of new 10s of popular modern cards every week; our Registry Watch tracks the minting rate. The only PRICE decliner among our twelve biggest tracked populations was the one whose registry grew fastest relative to demand. Scarcity claims about modern slabs have an expiration date.
8. Seasonal folklore is smaller than advertised. The famous summer dip measured out as a July slowdown, not a decline: a +6.6% median tracked-card move this summer, with the damage confined to last spring's modern chase cards while vintage rose double digits. Calendar timing is a seasoning, not a strategy.
9. Liquidity is part of the price. A $2,000 card that sells twice a year is a different asset from a $200 card that sells daily. Every card page here shows 52-week sales counts, comp sample sizes, and how the current price sits in its 52-week range; a thin number wearing a big price is where bad buys live.
10. Keep score against an index, in public. Anyone feels like a genius in a market that rose 40%. Our own buy list is logged at entry price and graded weekly against the moonstone Pokémon Index on the Scorecard page, including when it loses. Do the same with your collection: what beat the index is skill, the rest was tide.
Start your reading here
Each rule cites its source study: the SIR vs IR scarcity study (403 cards), the moonstone Spread (190 set histories), Population Giants (115 registry-tracked cards), the Grading Frontier downside analysis (219 cards), the Summer Dip audit, the rip-or-hold pack simulation (400 packs), and the public Weekly Buys scorecard. All are linked from the Research page and all show their methodology inline. DATING, added 2026-09-22 on an independent recompute: every figure on this page is as of the publication date, August 5, 2026, or the date of the study it cites, and none of them are live. Two were re-checked against production on 2026-09-22 and still hold: 151 Charizard ex gems at 27.4% against the 28% quoted, and Moonbreon reads a $4,175 PSA 10 median on 506 recorded sales against the $4,250 quoted. The rest rest on tapes that have since been truncated or on studies with their own as-of dates; treat the numbers as rules of thumb with a timestamp, not as current quotes.