Foundations Jumpstart EV Backtest: How a 2x EV Held Up Against a Real Opening

Box EV is a computed number, not a guess. Foundations Jumpstart is the clearest case of that, because there is nothing random about the pack contents. Wizards ships 121 preconstructed 20-card decks; a booster is one whole deck, drawn uniformly. SpellBook Finance reads those decks from MTGJSON, prices every printing, and sums the expected contribution from every card.

The audit unit is a single card contribution. Scythecat Cub sits in 2 of the 121 decks, so across 24 packs you expect about 0.397 copies per box; at $26.12 that is about $10.36 of display EV. The same arithmetic repeats for every priced card.

Expected copies are exact here, not a rarity estimate. A card in more decks shows up more often, so an uncommon that appears in a four-variant theme can out-contribute a pricier mythic that only lives in one deck. The EV follows the actual deck lists, not the sticker price.

The second question is whether the model resembles a real opening. This cohort used every public Jumpstart card row from the implied-box inventory. Read it honestly: 242 of 7,625 card rows are marked sold, and the sale-disposition log shows $1098.17 of realized cash across 286 sold units. The remaining 7,383 live cards are valued at current market, so most of the result is still a mark to market. The difference from the Play Booster backtest is that Jumpstart's realized cash has already recovered 0.83x of the box cost, because a grind-it-out product actually sells its many small cards. The realized portion is broken out first.

The box cost was $83.00. The live market price was $89.56. SpellBook Finance full EV was $185.26. The actual card result was $210.00 per implied box opened.

Part of the cost multiple is acquisition edge. The boxes averaged 7.3% below the $89.56 market price on 2026-07-05, because the buy came from hunting the EV rankings instead of paying whatever a random sealed listing asked. Marked at market, full EV is 2.07x and the observed card result is 2.34x market.

Two caveats keep this honest. The cost multiple leans on the $83.00 average across the implied 15.89 boxes, but only 8 displays carry explicit cost metadata, so read cost multiples as estimates and market multiples as firm. And about 16% of live printings appear across more than one intake flow; a conservative de-duplication lands remaining value nearer $1870.22 and the result nearer 2.09x market, right on top of full EV. The figure above is the higher end of that range.

A fair objection is that market value is not cash, and for a grind-it-out product like Jumpstart that objection has teeth: the value is spread across many small cards. So we modeled non-realized cards at 96% of market and applied marketplace fees to the whole gross result, including already-realized sales. At a 15% fee rate, the cohort still shows $1441.37 profit, 52.2% net margin, and 109.3% ROI on implied-box cost.

We used the production Jumpstart inventory cohort: every card row with set code J25. From the raw rows we excluded 2,029 archived duplicate shadows and 1,658 superseded re-scan rows (a box-rip re-scan marked the originals lost, but almost all still have a live copy), leaving 7,625 output cards. Each row was priced against the latest production card price for its finish. Sold rows were counted as realized sales, unsold rows as remaining card value at market, and the superseded rows were kept out of both the count and the value total.

The implied boxes opened count is card rows divided by a Jumpstart display footprint: 24 packs times 20 cards, or 480 cards per display, which comes to 15.89 boxes. Only 8 of those displays carry explicit purchase-cost metadata, so the paid-cost basis is an estimate applied across the rest. That is a floor, not a ceiling: cards that were discarded, given away, or sold and removed rather than left as rows are not in the count, so the true number of displays opened is at least this many.

The full EV comes from the live Foundations Jumpstart Booster Display row on SpellBook Finance, the same number the public box page shows, so a reader can cross-check it.

It proves that a real Foundations Jumpstart opening landed near the modeled EV. It does not prove every box wins. Variance still exists, prices move, and listed inventory is not cash until it sells.

Most of the result is unrealized. Only 242 of 7,625 cards have sold, and the rest is valued at current market using the same prices that feed the EV, so the close match is partly expected. The honest claim is narrow: the modeled EV is auditable, complete on coverage, and consistent with how this cohort is selling, not a promise of cash in hand. Jumpstart in particular pays out as many small sales, not one chase hit, so patience and listing volume matter more than luck.

If the math made you want cardboard instead of caveats, the SpellBook Shopify store is over here. No variance proof required at checkout.