As someone who makes predictions using data and mathematics for a living, I always try to think about pricing cards (both as a buyer and as a seller) through that lens by asking myself 'what would I predict the next one of these to sell for?' If you are only considering one data point, then past sales are without question, the best predictor of future sales prices, so it makes for a great starting point. However, it is certainly not the only factor one should consider. If it's a relatively rare card that sells infrequently, then it gets much more complicated, whereas something like a PSA 9 modern card that sells with high frequency doesn't really need any data points beyond what the last handful of them sold for on average. But if you're looking at a card that only sells once or twice per year, then you would benefit greatly from analyzing other related market factors that are highly correlated with what you are selling (how have other cards in that same set moved since your card was last sold? or how has the market for the player you're selling moved since the last time this card was sold? or how has the same card moved in other grades?). Other aspects that are especially relevant with vintage obviously, might be things like eye-appeal or when the card was graded (is it an old PSA label from the days when modern 8s got 10s and modern 6s got 8s?, etc.) Obviously, you can make it as complicated as you want, but if you start with past sales prices and then make market adjustments from there, you'll be pretty close to how the broader market values that card and what it's most likely to sell for in a future sale.
|