Study: These Factors Can Help You Predict Sports Card Prices

Research and data analysis by Ashley McKillips.

The sports cards craze has been in full swing for a couple of years now, and a number of lucky investors have already hit home runs. In 2021 alone, an astounding 14 different sports cards sold at auction for over $1 million. A Mickey Mantle card shattered records when it sold for $5.2 million in early 2021, only for a Honus Wagner card to take its place as the most expensive sports card ever sold, going for $6.6 million later in the year.

So how do you get in on the action? Knowing how to find the best sports cards to invest in can be tricky, so we combed our data to pull out some of the top characteristics most commonly found in these lucrative sports cards.

There were several significant commonalities amongst the top sports cards that can help you predict which cards are going to be desirable—and thus, profitable. Here's what we found.

• 72% of the top 200 sports cards are rookie cards.
• Number of championships won and MVPs held were the two most important predictors of sports card price, with a high number of championships being associated with the biggest price increases.
• Players featured on the top 200 sports cards have won an average of 3 championships and 4 MVPs.
• Basketball cards are the highest priced cards, with the average top basketball card being priced at $57,035. This is followed by football cards with the second highest average price, then baseball cards, and finally hockey cards.

In examining price data from these top sports cards, we found that basketball cards carry the highest price tag on average, followed by football, then baseball, then hockey. A breakdown of the average price of the top sports cards by sport is as follows:

• Basketball cards: $57,034
• Football cards: $43,194
• Baseball cards: $34,094
• Hockey cards: $18,036

Baseball cards, considered the OG when it comes to sports cards, have fallen far from their historical position as the most popular trading cards. Given the role that history plays in the hobby of sports card collecting, it may be surprising to see that baseball card prices fall so far below those of football and basketball. After all, the most expensive trading card of all time is still a Honus Wagner card, and many of the most popular sports cards are baseball cards.

Of course, it can be argued that both basketball and football have overtaken baseball in recent decades as America's favorite pastime. The former two certainly have more star-power and hype associated with them than baseball, which goes a long way toward influencing price. Athletes turned celebrity icons such as Michael Jordan and Kobe Bryant make up some of the best basketball cards. And since the beloved sports card brand Panini has become a top producer of NFL cards, they've managed to produce many of the best football cards in the game in a matter of years. Meanwhile, many of the best baseball cards are cards that have been valuable for decades, which begs the question of whether the top football and basketball cards will remain this highly-priced in decades to come.

The highest-priced cards included in the study by sport are as follows: 

• Basketball: 1986 Fleer Basketball Michael Jordan ROOKIE RC #57, PSA 10 in mint condition, priced at $456,455
• Football: 2000 Playoff Contenders Tom Brady Auto Rookie #144, PSA 10 in mint condition, priced at $555,988
• Baseball: 1952 Topps Mickey Mantle #311,PSA 8 NM-MT (PWCC-A) - Certified Top 30%, priced at $489,300
• Hockey: 1979 O-Pee-Chee Hockey Wayne Gretzky ROOKIE RC #18, PSA 9 in mint condition, priced at $200,000

Our study looked at the influence of three factors on the price performance of sports cards: the number of championships a player has won, the number of MVPs a player has won, and whether or not the card is a rookie card.

1. Championships

Of the three factors we looked at, the number of championships and MVPs the player has under their belt were the best predictors of price. A high number of championships won is associated with a greater price increase than a high number of MVPs.

Of all 200 sports cards studied (50 from each of the four sports), the top 20 highest-priced cards were of players who've won an average of 4.5 championships, with 6 and 7 championships being the most common. Across the 20 cards included in the study with the lowest prices, the players featured average 1.6 championships each, with 0 championships won being the most common. All the cards in the bottom 10 feature players who have won no championships.

Five of the top 20 cards feature players with 6 championships and five feature players with 7 championships. The only card in the top 20 featuring a player with no championships was the 2009 Bowman Chrome Gold Refractor Mike Trout Auto Rookie, BGS 9.5 (PWCC), priced at $97,609. Despite having never played in a World Series, Mike Trout cards have long been some of the highest-selling sports cards, and this being a rookie card does help its value.

2. MVPs

The players featured in PWCC's top 200 sports cards hold an average of 3.96 MVPs. This is quite high. For reference, Michael Jordan has 5 MVPs, Mickey Mantle has 3 MVPs, Tom Brady has 5 MVPs, and Wayne Gretzky has 9 MVPs.

As stated previously, the number of championships and MVPs are the best predictors of price we found, with championships increasing price a little more than MVPs. However, a player winning a significant number of MVPs still proved to be a significant price predictor.

Across the top 20 highest-priced cards included in our study, the players held an average of 5.5 MVPs. It was most common for players in the top 20 cards to hold 9 MVPs. For the bottom 20 lowest priced cards included, players averaged 1.5 MVPs each. It was most common for these players to have 0 MVPs.

It may also be the case that MVPs held is a more important factor in some sports than others. When we looked at the top 10 highest priced cards from each sport, these are the average number of MVPs held by the players featured:

• Basketball: 8.1 MVPs
• Football: 3.9 MVPs
• Baseball: 2.3 MVPs
• Hockey: 8.2 MVPs

3. Rookie cards

Rookie cards tend to be seen as more valuable than non-rookie cards, and our study confirmed this is the case. Of the top 200 cards studied, 72% were rookie cards.

Football has the highest concentration of rookies at 88%, followed by hockey at 80%, basketball at 60%, and then baseball at 58%.

Interestingly, the average price of rookie cards and non-rookie cards included in the study is almost the same, the former being $37,439 and the latter being $37,772. However, there are a disproportionate number of hockey cards categorized as rookie cards, and their lower-than-average price drags down the average for all rookie cards. When split up by sport, it becomes a little more clear when rookie cards are significant predictors of price and when they aren't.

Basketball

• Average price for rookie cards: $66,344
• Average price for non-rookie cards: $43,070

Football

• Average price for rookie cards: $37,406
• Average price for non-rookie cards: $22,740

Baseball

• Average price for rookie cards: $30,848
• Average price for non-rookie cards: $38,578

Hockey

• Average price for rookie cards: $20,574
• Average price for non-rookie cards: $7,885

Based on this data, it seems to be the case that rookie cards are significant price predictors for every sport except baseball. Surprisingly, the top baseball cards categorized as rookie cards actually had a lower average price than those categorized as non-rookie cards.

This doesn't mean rookie cards will always be more expensive than non-rookie cards. When we looked at the top 20 highest-priced cards included in the study, 75% were rookie cards. In other words, a quarter of the 20 highest priced cards (all priced near or well over $100,000) were not rookie cards. These included highly sought after favorites such as two 1952 Topps Mickey Mantle #311 cards, and two 1997 Metal Universe Precious Metal Gems Green Michael Jordan cards. 

Despite not being rookie cards, these cards all had other unique qualities. For example, the Michael Jordan card is one of only 10 produced with the green background. The Mickey Mantle card is considered one of the most valuable sports cards in the world largely because it's part of the first annual set of cards Topps ever produced.

Our data shows that championships won, MVPs held, and rookie designation are all very important factors in predicting the price of a sports card, with championships won having the biggest influence on price. Based on our findings, the highest priced sports cards are likely to be rookie cards from players who have won at least 4.5 championships and hold at least 5.5 MVPs. The sport matters too—basketball cards currently rake in the most cash.

Of course, it's rare that one card has all these characteristics, and there are always exceptions. For example, one hockey card made the top 20 highest priced cards (a 1979 O-Pee-Chee Hockey Wayne Gretzky rookie card in mint condition). At $200,000, it's certainly an outlier when compared to the sport's average of $18,036 per card. A 2000 Playoff Contenders Championship Ticket Tom Brady with Auto is in the top 5 highest priced cards, at $400,100, despite not being a rookie card.

That said, these generalizations are still helpful for card collectors who are looking to make educated guesses about their next sports card investment.

We collected player data on PWCC's top 50 most bid on baseball, football, basketball, and hockey cards as of March 2022. Player data was collected from StatMuse.com. We chose to take the top 50 cards from each sport category, rather than the top 200, in order to provide data that could be applied to any of the four sports. If we had chosen overall sports data and only had 200 data points it could have been skewed toward a particular sport. 

We chose not to remove repeated popular cards that have been bid on and sold multiple times from the list. If they are popular enough to make the top 50 most bid on list multiple times, then their characteristics must be significant.

While the data is inherently random, as literally any cards could have been the top traded, it is also clustered by sport in groups of 50 so the sampling isn’t truly random. While this gives us balanced data across all sports, it can lead to potential problems in the prediction model, as the variation from the true value is greater with clustered data. This does not impede overall takeaways about the characteristics of sports cards,

We used a simple linear regression (SLR) model in the program R to try to find the best player statistic to predict the price of a sports card. The predictors we wanted to test were rookie status, number of MVPs, number of Championships, and if the player is in the hall of fame. We started by running a correlation test on the variables so we could see which would create multicollinearity in the model. 

Next, we used stepwise regression to find the best model out of all of the predictors. Stepwise regression is a kind of prediction strategy where you systematically add and remove predictors from the overall model to create the most accurate nested model. Finally, we checked that the model met the assumptions of simple linear regression. Most of the data points fell on the line, so we can assume normality. The residuals were also mostly linear with the exception of a few outliers, and similarly, the variance was also mostly constant. Based on this, we can say it's safe to use simple linear regression.

If we were to do this study again, we would run it without clustering to see if the sport is a good predictor of price. It would be interesting to run the model without having any card repeats, or to run it again with the top 1,000 cards for each sport. We'd also like to run a deeper analysis on specific sports. That way, we could look at sport-specific statistics as predictors of price (i.e. RBIs, or lifetime touchdowns). We could also look only at cards that have been traded in the last two years, look at card brand as a categorical variable, year minted as a predictor, and player as a categorical variable.

1. Ashley McKillips (2022, March). Proprietary MoneyMade data [Sports card prediction models built by our research team]. MoneyMade.
2. PWCC Marketplace | Leading the Trading Card Industry. (n.d.). PWCC. Retrieved March, 2022, from https://www.pwccmarketplace.com/
3. StatMuse | Search StatMuse, save time. (n.d.). StatMuse. Retrieved March, 2022, from https://www.statmuse.com/