If we’re going to talk MTG Finance and Cardsphere finance, I think it’s important to talk credentials first.

I currently have a little over 2,800 cards sent on Cardsphere, and a little under 2,000 cards received. I’ve sent cards worth more than $50, and received cards worth $250 or so at the high end. Based on that experience, as well as a fair bit of experience looking at data involving TCGPlayer, Channel Fireball, and Card Kingdom, I think I’m reasonably well equipped to talk about how to figure out what a card is ‘really’ worth.

Like the title says, there are several pieces of price data that go into determining this.

Cardsphere List Price

First things first, let’s discuss the most basic element of Cardsphere prices, the list price. This price is important because we talk about other prices in terms of percentages in list price, but it’s not the end all be all, because buying or selling a card at 100% of list price is a rarity

So, given that it’s the basic element against which all user prices respond, how does Cardsphere come up with their list price for a card?

I don’t know and that’s a good thing. But in order to explain why it’s a good thing, I have to talk about TCGPlayer for a second.

Several years back the standard price that everyone used in paper trades was TCGPlayer mid, and back then TCGPlayer mid was a mean of all of the listed prices of a given card. Everyone knew how that price was made, so it was seen as a trustworthy pricing method for paper trading.

But let’s talk hypotheticals. Let’s say Jeremy the Jerkbag knows you want to trade into Thragtusk. Now, at the time Thragusk was an expensive card, but James knows you use TCGPlayer mid for all of your pricing, and Jeremy has a TCGPlayer store. Knowing this, and knowing how TCGPlayer does their prices, Jeremy lists a Thragtusk at $65,000 dollars. There are a lot of listings for Thragtusk, so all his offer does is raise the buy price by $5, but that means you’re out an extra $20 when Jeremy trades his playset to you.

TCGPlayer, being a smart site, decides to change this by switching to the median instead of the mean. The problem here is that it only minimizes the amount of damage that garbage listings do, it doesn’t eliminate them. Now, Jeremy the Jerkbag is listing a few thousand cards at five times their list price, and a bunch of jerks just like him do the same. If any of those cards get bought out, the listings that only exist to bait out crazy people are the only ones left to make the median. So then if there’s a buyout Jeremy can trade that card in paper for a crazy price until enough people who actually want to sell the card relist it on TCGPlayer, and he has the added upside of maybe tricking some poor fool into buying the card for the crazy price he listed but never expected to get.

Currently, TCGPlayer uses Market price, which is built on past sales but they don’t say how. That’s because TCGPlayer has gotten wise to the fact that if they disclose how they make their prices, then people can do things to mess up those prices. If the Jeremys of the world knew how long prices were used in the Market calculation, they could do things like using fake purchases to screw up the Market calculation. Keeping the specifics of how the sausage is made under wraps helps prevent this sort of foul play.

Similarly, Cardsphere’s pricing data is based on information gathered from other sites' pricing data. That’s all I know. I’d like to know more, but I value the Jeremys of the world not knowing more than I value knowing. If Cardsphere gets prices from Ebay, for example, I don't want some jerk with an Ebay store poisoning Cardsphere price data so they can try to sell me overpriced cards.

What’s really important for the Cardsphere index price is that it’s somewhat consistent relative to other prices across all cards. With some slight latency for price spikes, this tends to be reasonably true. There’s inevitably some bumpiness relative to one site or another, but that’s true of any site. On the mean, Cardsphere’s index price tends to be at the low side a bit below TCGPlayer Market, and at the high side somwhere between TCGPlayer Market and TCGPlayer Median.

Lowest and Highest Historic Sales

So, now that we know that we can see Cardsphere index price is at least a good enough price point against which to base % offers, what do those % offers look like? Let’s take a look at a card. At a glance, MTGGoldfish suggests that Ancient Stirrings is a very prevalent Modern staple, so let’s use the original printing of that.

At time of writing, the lowest and highest offers in the last 10 completed trades were 69% and 81% respectively. This trading took place over the course of a little bit more than the last 2 weeks and contains only trading for the nonfoil Masters 25 printing of Stirrings. Thankfully, there’s no major noise in terms of non-English or non-NM trades in this history.

So, what does this tell us? Well, if you’re buying, it tells us that an 81% offer on a Masters 25 Ancient Stirrings is about the highest you’ll ever need to offer, and should get it for you pretty darn quickly. That makes sense, because at $2.01 it’s a tad bit higher than the cheapest you can get a copy on TCGPlayer as of writing. If you were trying to turn cards you’d shipped out into copies of Ancient Stirrings in as fast a way as possible without bleeding money to buylisting, the 81% offer would represent that strategy.

Alternately, if theperson offering 81% were trying to get a large volume of Ancient Stirrings, there are only about a playset worth of copies from TCGPlayer around that price point. Buying more than 4 copies on there would likely involve eating more costs, so it would ultimately be easier and cheaper to just grab however many copies from Cardsphere.

By contrast, the 69% offer has pretty good odds of being someone who doesn’t strongly care about having Ancient Stirrings, but could use some eventually, or would be all right with them as an add-on to an existing trade. If you were looking to get Stirrings for as low a price as you could easily get it for, something around the 69% offer would a good idea.

For this printing of Stirrings, we don’t have any completed listings in recent trades that look like ‘buylist-y’ offers on a sub $5 card, or offers in the 30-60% range. Trades like those represent the strategy the user who traded into Stirrings at 69% did, but an order of magnitude larger. Buylist users on Cardsphere want just about everything and almost always have the funds for it, but their offers have to be bad enough to avoid going broke. When you see those sorts of prices in a price history, they represent the lowest you could possibly get the card for, but any given buylist user isn’t really likely to ever get any specific card. They just end up with copies of whatever people want to move. If we look at the Trade Prices table again, we might do something like define ‘buylist’ prices as prices at or below the bottom bound of the “50% of trades” field for a given price range, or roughly the bottom 25% of trades.

If you’re selling, the 81% offer represents the best offer you could get. If you see someone buying Stirrings at or around 81%, you should likely jump on that offer. The 69% represents the lowest offer others were willing to tolerate. You can consider sending at 69%, but a better offer is reasonably likely. Any buylist offer you should treat as noise, unless you’re specifically eager to get rid of cards quickly or without sending out a ton of packages.

At their best these price points can tell you what the lowest and highest you can expect to pay for a card, and also the least and most you can reasonably expect to get for it. That said, there are some hurdles to consistently using this price data.

One of the few areas I think Cardsphere could use some serious improvement is in the price history they keep and display. The last 10 trades can, depending on the card, include the entire trade history of the card or a single package worth of that card committed less than a week ago. Additionally, they can be crowded by bad condition cards and foreign languages, giving even less than 10 trades worth of history. Rat Colony is a highlight case for the way buylists can blind an entire trade history in a single transaction. While the most recent purchase of Rat Colony was at a non-buylist price point, it was, like most Rat Colony purchases, about a billion copies.

Similarly, when people buylist a card they sometimes buylist a ludicrous number of copies of the card. This ends up causing the entire price history of that card to be covered up by a single transaction, hiding useful pricing data people would otherwise use. I don’t need to know that someone bought a zillion copies of Dominaria Opt at $0.05 each, if I want a playset of my own I need to know what sane offers look like.

Offer to Beat

Historic price trends aren’t the only thing you have access to when it comes to trying to get copies of a card. Offers on Cardsphere are public, and on any given cards page the top offers will show you the offer you have to beat to be on top.

Where the historic highest sale will show you the highest offer a reasonably sane person was willing to give in order to get the card quickly, the offer to beat will show you the bare minimum you need to pay to be ‘next in line’ on the card.

Obviously, you aren’t literally next in line, but for the most part a rational person will choose you over any of the competition. This is a common price setting strategy for good reason.

The only exception to this is when other users have wants that include other cards. Let’s say I want an Overgrown Tomb, and I beat the Tim with the next top offer by 5%, or about $0.33. But the person who’d trade either of us the Tomb also has about $3 in cards to trade to Tim. If they trade to Tim they get to move those other cards when they otherwise wouldn’t be worth the stamp. Sure, they lose out on $0.33, but they’re cycling through their collection faster and getting rid of stuff they don’t need moreso than they do if they trade with me.

For foreign users there can be other issues. International stamps are about twice as expensive and some foreign mailing systems are bad enough that lost packages are a real threat. For an individual card, you’ll have to beat domestic (read: American) traders by at least the price of an International stamp in order to be likely to get the card, but you can get around that somewhat by having a more wide variety of wants.

Despite those slight exceptions, beating the current top offer by a bit and consistently rechecking to make sure you’re still top offer is the best and easiest way to guarantee that you’ll get frequently traded card without breaking the bank.

Final Notes

In context, often the best strategy is to use several of these price points in concert. Maybe the person with the top offer beats out the highest recently completed trade of a card. In that case, you could be second in line and be reasonably confident they would get it quickly and you’d be on top after that. Maybe you know that you have enough wants to get a card in a package despite losing to the top offer, you might set your offer % to something more like one of the lower historic sales. You could also do things like looking at the want list of the person making the offer to beat. If their wants are more diverse than yours, they’ll likely be able to get whatever card you’re looking at in a package with other cards, so you might try and beat them by a healthier margin than you otherwise would, in anticipation of that possibility.

What I’m trying to do by giving you this pricing information isn’t to tell you what to think, it’s to tell you how to think. Now that you know what the data is, it’s up to you to decide what you want to conclude. Smart trading is getting what you want at the speed you need it for the lowest price you think you can. I hope I’ve given you good tools to determine the ‘lowest price’ part of that equation, but ultimately all three of those elements are up to you to try and calculate.