Adjusted EV — Calibrated Pack Expected Value Methodology
Adjusted EV is the realistic expected value of a TCG pack after correcting headline EV for thin tape, liquidation costs, and pool depletion. The formula, calibration inputs, and worked examples used by GapSense to flag overpaid packs.
Why Claimed EV is misleading
Most pack EV calculators in the TCG space publish a single headline number. Pack costs $4.99, advertised pulls average $5.40 of value, EV +8% — buy the pack. The problem is what's hidden in the average. A pack EV is the expectation over a random draw from a finite pool, but most calculators score every chase card at full market price as if you can liquidate it instantly. In reality:
- Top chase cards have thin tape — the last sale was one buyer, weeks ago, on a single platform.
- The pool is depleting — every unboxer pulls and lists, suppressing realized prices for the next seller.
- Liquidation has execution cost — platform fees, redemption gas, KYC delays, FX.
Claimed EV does none of these adjustments. Adjusted EV does all three.
The Adjusted EV formula
For a pack p with possible cards c ∈ C, draw probability P(c), and observed market price price(c):
Adjusted_EV(p) = Σc∈C P(c) × price(c) × tape_weight(c) × (1 - exec_cost(c)) - pool_drag(p)
Three calibration multipliers on top of the textbook EV:
tape_weight(c) ∈ [0, 1]— confidence in the published price. A function of sample size (how many recent comparable sales), recency (days since last comp), and platform diversity (one venue or many). See tape thickness.exec_cost(c) ∈ [0, 1]— fraction of headline price lost to fees + redemption + grading + FX during liquidation.pool_drag(p)— expected price decay across the remaining pool as more packs are opened. Modeled per platform from depletion curves.
Worked example
Real numbers from a Phygital oripa pack tracked by GapSense in April 2026:
| Metric | Claimed | Adjusted |
|---|---|---|
| Pack cost | $4.99 | $4.99 |
| Top chase headline price | $1,200 (PSA 10) | $1,200 |
| Top chase tape weight | 1.00 (assumed) | 0.32 (1 sale, 41 days old, 1 platform) |
| Execution cost on chase | 0% | 22% (14% platform + 5% redemption + 3% FX) |
| Pool depletion drag | $0 | -$0.18 / pack remaining |
| EV per pack | +$0.41 (+8.2%) | -$0.06 (-1.2%) |
Same pool of cards, same probabilities. Claimed EV says buy. Adjusted EV says skip. The next 100 unboxers acting on Claimed EV collectively realize the loss.
When it changes a decision
Adjusted EV diverges most from Claimed EV in three regimes:
- Long-tail chase pulls. One historical $5,000 sale dominates Claimed EV. Adjusted EV downweights it to ~$1,500 and the pack flips negative.
- New oripa pools. Pool depletion is steepest in the first 100 packs sold — Claimed EV ignores the rate of pull, Adjusted EV models it.
- Cross-currency packs. JP packs priced in JPY but liquidated in USD eat 4-6% on FX alone, never reflected in headline EV.
vs naive EV calculators
| Approach | Tape weight | Execution cost | Pool depletion |
|---|---|---|---|
| Spreadsheet EV | None — uses last sale | None | None |
| EVCharting | None published | None published | None published |
| UnopenedPokemon | Volume gates only | None published | None published |
| GapSense Adjusted EV | Sample-size weighted | Per-platform fee + redemption table | Per-pack depletion model |
Read the live calibration record at /pack-ev or via the agent guide: /guide (mode pack-ev).
Related concepts
- Tape thickness — sample-size weighting and freshness
- Cross-platform gap — turning Adjusted EV into a trade
FAQ
How is Adjusted EV different from Claimed EV?
Claimed EV multiplies probability by the last published price. Adjusted EV multiplies by price × tape_weight × (1 - exec_cost) and subtracts a pool depletion term. The two diverge most on packs with thin chase tape or high redemption cost.
Where does the tape weight come from?
Tape weight is a function of sample size (how many recent comparable sales), recency (days since last comp), and platform diversity (single venue or many). A single 41-day-old sale on one platform scores around 0.3 — meaning the published price is downweighted to 30% of its headline contribution.
Does GapSense publish the pool depletion curve per platform?
Yes — depletion drag is calibrated per platform from observed price decay as packs are opened. Live calibrations are exposed via the /pack-ev endpoint and refreshed continuously from the platform feeds tracked by GapSense.
When should I trust Claimed EV instead?
When all three correction terms are near their no-op values: thick tape (many recent comps across platforms), low execution cost (e.g. liquidating on the same platform you bought from in the same currency), and a non-depleting pool (mature secondary market with stable supply).