The Expected Value Framework

Every grading submission is a bet with a known cost and an uncertain outcome. You pay a fixed amount (raw card price + grading fee) and receive a variable return (the graded card's market value, which depends entirely on the grade it receives). Expected Value (EV) converts this uncertainty into a single number that tells you whether the bet is profitable in expectation — and by how much.

Our EV model evaluates every card in the Excellent Cards database using this framework. Currently, 2,314 of 61,370 English cards show positive EV — meaning the math favors submission when all grade probabilities are properly weighted. Here is how to run the calculation yourself, step by step.

The Basic Formula

For a single target grade, the simplified formula is:

EV = (Probability of grade x Graded price) - Raw price - Grading fee - Shipping costs

Example: You have a raw card worth $50. PSA 10 sells for $500. You estimate a 30% chance of achieving PSA 10. Your all-in grading costs (fee + shipping + supplies) are $35.

EV = (0.30 x $500) - $50 - $35 = $150 - $85 = $65

Positive EV of $65. This submission is profitable in expectation. But this simplified version ignores all other grade outcomes. What if the card comes back PSA 9? PSA 8? PSA 7? Each of those outcomes has a probability and a market value that should be included.

The Full Multi-Grade Formula

In reality, a card can receive any grade from PSA 1 to PSA 10. The complete calculation accounts for every possible outcome weighted by its probability:

EV = (P10 x Price10) + (P9 x Price9) + (P8 x Price8) + (P7 x Price7) + (P6 x Price6) + (Pbelow x Pricebelow) - Raw price - Grading fee

Where P10 is the probability of PSA 10, Price10 is the PSA 10 market price, and so on for each grade. Pbelow and Pricebelow represent the probability and average price for grades PSA 5 and under (typically negligible for cards in visually good condition).

Key insight: grades below your break-even point have negative contribution to EV. If a PSA 7 sells for $60 but your total cost is $85, every PSA 7 result costs you $25. A thorough EV calculation includes these losses, which is why the full formula sometimes produces a significantly different (and lower) EV than the simplified PSA-10-only calculation. Ignoring the downside scenarios is the most common mistake in amateur EV analysis.

Step-by-Step: A Real Example

Let us calculate the EV for Reshiram (Black & White), a full art card from the BW era:

Known data:

  • Raw market price: ~$51
  • Grading fee (PSA Value tier): $25
  • Shipping and supplies (amortized per card): ~$10
  • Total cost: $86
  • PSA 10 market price: ~$2,505
  • PSA 9 market price: estimated ~$150 (based on recent sales)
  • PSA 8 market price: estimated ~$80
  • PSA 7 or below: estimated ~$40 (you lose money at this grade)

Assume this is a pack-fresh card with good centering that you have examined under magnification. Estimated grade probabilities based on the card's era and your assessment:

  • PSA 10: 15% — the card looks strong but BW-era gem rates are moderate
  • PSA 9: 45% — the most likely outcome for a well-preserved BW card
  • PSA 8: 25% — possible if there is a flaw you missed
  • PSA 7 or below: 15% — unlikely but not impossible for older cards

Grade-by-grade contribution:

  • PSA 10: 0.15 x $2,505 = $375.75
  • PSA 9: 0.45 x $150 = $67.50
  • PSA 8: 0.25 x $80 = $20.00
  • PSA 7 or below: 0.15 x $40 = $6.00

Gross EV = $375.75 + $67.50 + $20.00 + $6.00 = $469.25

Net EV = $469.25 - $86.00 = $383.25

This card has a strong positive EV of approximately $383 per submission. Our database shows a blended EV profit of ~$196, which uses more conservative probability estimates (lower PSA 10 probability, accounting for the population-weighted average across all submissions rather than a hand-picked candidate). Either way, the submission is clearly profitable in expectation.

Notice that 80% of the EV comes from the PSA 10 outcome, even though it only has a 15% probability. This is typical for high-multiplier cards: the PSA 10 drives the math, and everything else is noise. This means your PSA 10 probability estimate is the single most important input — a 5-percentage-point error in that estimate changes the EV by $125.

Estimating Grade Probabilities

The hardest part of EV calculation is estimating the probability of each grade. This is where the science meets the art, and where experienced graders develop their edge over beginners:

  • PSA population reports. The gem rate (PSA 10 / total graded) gives a baseline probability — but it reflects the average condition of all submitted copies, not yours specifically. The gem rate includes cards submitted by optimistic beginners alongside cards submitted by expert graders. Your card may be better or worse than the average submission.
  • Self-assessment adjustment. After examining centering, corners, edges, and surface (see our PSA 10 identification guide), adjust the baseline probability up or down based on what you observe. Perfect centering and sharp corners? Adjust up from the population gem rate. Slightly soft corner or marginally off centering? Adjust down significantly.
  • Historical personal results. If you have graded cards before, your personal gem rate is the best predictor of future results. Track every submission. If your historical gem rate on WOTC holos is 8%, use 8% — not the 15% you believe each new card deserves. Your track record does not lie.
  • Era-based benchmarks. As a starting point before you have personal data: WOTC 5-15%, ex/DP 10-25%, modern 30-60% for pack-fresh cards. These ranges account for the typical condition of cards submitted within each era.

Conservative probability estimates protect you from overconfidence. Most graders overestimate their PSA 10 probability by 10-20 percentage points, especially early in their grading career. Use the lower end of your estimate range until your track record proves otherwise.

Break-Even Analysis

The break-even grade is the minimum grade needed for the graded card to sell for more than your total cost. This tells you the grade threshold below which you lose money:

Break-even selling price = (Raw price + Grading fee + Shipping) / (1 - Selling fee rate)

For a $50 raw card with $35 grading cost and 13% eBay selling fees:

Break-even selling price = ($50 + $35) / (1 - 0.13) = $85 / 0.87 = $97.70

If PSA 9 sells for $150, you break even comfortably at PSA 9 and profit at PSA 10. If PSA 9 sells for $80, you need PSA 10 to be profitable — a much riskier submission that requires high confidence in your card's condition.

Break-even analysis tells you how much margin for error you have. A card where you break even at PSA 8 is much safer than one where you need PSA 10. The former tolerates a wider range of outcomes; the latter is an all-or-nothing bet.

Selling Fees and Hidden Costs

Many graders calculate EV using the selling price but forget that the selling price is not the money in their pocket. Include every cost:

  • eBay fees: ~13% of the final sale price (12.9% final value fee + $0.30 per transaction). On a $500 sale, that is $65 in fees.
  • TCGPlayer fees: ~12-15% depending on seller level. Slightly less than eBay for most sellers.
  • Shipping cost when selling: $4-$8 for a graded card in a padded mailer with tracking. $12-$20 for Priority Mail with insurance for high-value cards.
  • PayPal/payment processing: Usually included in platform fees now, but 2.9% + $0.30 if selling independently through a website or social media.
  • Insurance: For high-value cards ($500+), insured shipping adds $5-$20 depending on declared value. Some sellers self-insure by accepting the risk; others buy insurance on every shipment.
  • Time cost: Listing, photographing, packaging, and shipping each card takes 15-30 minutes. If you value your time at $30/hour, each sale costs $7.50-$15 in implicit labor costs.

A card that sells for $500 on eBay nets you approximately $420-$430 after all fees, shipping, and packaging. Your actual ROI is based on that net figure, not the headline sale price. Overlooking these costs is how graders convince themselves they are profitable when they are actually breaking even.

ROI vs EV: Different Metrics for Different Decisions

EV tells you the expected dollar profit per submission. ROI tells you the percentage return on your invested capital:

ROI = (Net EV / Total cost) x 100%

A $10 card with $5 EV profit has 50% ROI. A $1,000 card with $200 EV profit has 20% ROI. The first is a better percentage return; the second generates more absolute profit per card.

For grading portfolios, both metrics matter. High-ROI submissions maximize your return per dollar invested — critical when you have limited capital and need to allocate it efficiently. High-EV submissions maximize total profit per card — important when you are constrained by the number of cards you can source and submit rather than by capital.

Most graders should optimize for ROI in the early stages (when capital is the constraint) and shift toward maximizing total EV as their capital base grows.

Portfolio-Level Thinking

Individual submissions are noisy — any single card can grade well above or below expectation, and even a perfectly assessed card can receive a surprisingly low grade on a bad day. The power of EV analysis is in the aggregate. Over 50-100 submissions, your results converge toward the expected value as the variance averages out.

This convergence has practical implications:

  • Do not overreact to individual results. A PSA 7 on a card you expected to be PSA 10 is disappointing but statistically normal. If your process is sound, the next batch will compensate.
  • Track your cumulative EV vs actual results across all submissions. If you are consistently underperforming your EV estimates by more than 10%, your self-assessment is too optimistic and you need to recalibrate your probability estimates downward.
  • Diversify across price points and eras. A portfolio of 20 cards at $30-50 raw is lower-risk than two cards at $300-500 raw, even if the total investment is similar. The larger sample size reduces variance around the expected outcome.
  • Set a stop-loss. If a batch of 20 cards significantly underperforms expectations, review your assessment process before submitting the next batch. The data is telling you something about your accuracy.

Sensitivity Analysis: What If Your Estimates Are Wrong?

Because grade probabilities are estimated rather than known, it is valuable to test how sensitive your EV is to different assumptions. For any card you are considering:

  • Calculate EV with your best-case probability estimates
  • Calculate EV with 50% lower PSA 10 probability (your pessimistic scenario)
  • Calculate EV assuming no PSA 10 at all (pure PSA 9 and below scenario)

If the card still shows positive EV in the pessimistic scenario, it is a robust submission. If it only works with your best-case estimates, it is marginal and you should consider whether your optimism is justified.

Use Our Calculator

Our grading analysis tool automates this entire calculation for every card in our database. Enter any card to see its current EV, break-even grade, and probability-weighted returns across all grade outcomes. The data updates daily as market prices change, so yesterday's marginal submission might be today's strong positive EV — or vice versa.

For a broader view of which cards offer the best grading returns right now, see our best cards to grade in 2026 ranking. And for the condition assessment skills that underpin accurate probability estimates, study our PSA 10 identification guide.