In speculating/investing in magic there's three numbers I like to always keep in mind for any particular card before evaluating whether its worth it to jump in.
1. The lowest cost of the card on ebay from a seller with decent feedback.
2. The buylist price from big name sellers --CFB, SCG
3. The sale price from the big name sellers.
Let's call the difference in price between what CFB sells the card for, and what it will buy the card back for, Sale Price/Buylist spread. (borrowing terms from finance)
Because the buylist of these stores represents their "demand" for the cards, to me it stands to reason that the higher the spread, the more fluff you're buying.
What do I mean? Well...Suppose hypothetically that you buy a card from CFB for $40. But suppose also that the buylist only values the card at $5.
In that case, you're buying something for $40 that the seller him/herself is only willing to pay $5 for.
On the other hand, if you're buying something for $40 --Liliana of the Veil, and CFB's buylist price is $35, then you're buying a card that CFB has to spend $35 to acquire.
In one of these cases, the price is "fluffier". It could be hype, it could be temporary speculation, it could be that the seller has too many of them in their inventory. But for whatever reason, if they need more, they're at this time only willing to pay a small fraction of the cost.
In fact the larger the spread (the difference between the sale price and buylist), the more likely it is that the money you're putting into the card is something for which you are not being compensated adequate value.
There has to be at least some spread of course, because that's how the sellers make their money. In other words CFB can't sell you LOTV at $40 and buy it back at $40 on their Buylist. It wouldn't make sense.
Enter now, the world of Ebay. Ebay is one of the primary venues through which the individual magic player can turn their cards into money again. In an efficient market, ebay sellers should only be able to sell when their cards are below the price of the big seller.
Because you're getting the card from ebay cheaper than the big sellers, you now have another salient data point. The difference of the ebay price over the buylist of the current big sellers. (ebay/buylist spread)
Suppose CFB is buying Liliana of the Veil at $35. (they are at the moment)
Suppose you see a reputable ebay seller selling NM copies at $35. Should you buy? For me the answer is always yes.
In fact, after ebay bucks (2% cash back), and additional cash back from your credit card, you will actually come away with some small profit.
These two data points can give us some information.
1. The ebay/buylist spread tells us something about absolute profitability. For me, its pretty much the final arbiter on whether or not I might buy.
2. The Sale Price/buylist spread on the other hand tells us something about inflated hype, how many more copies the seller has in stock, or even low sales volume(low liquidity) and gives us more information on how the seller views the card.
Dual lands are a particularly interesting case study in all this. Why? Well let's start off with something about new cards.
Theros, Innistrad, RTR. Does anyone really know how many of these cards were printed? How many boxes sold? I certainly don't. As an individual buyer, if I choose to speculate, I'm going in the dark here completely.
If I choose to go all in and buy up a bunch of Snapcaster Mages, I have no idea whether Starcity has 25 left in stock or 250. And so that is why the buylist price of the big vendor is pretty important-it reveals their demand for the card.
But what about dual lands? These are different. Being twenty years old, on the reserve list, big sellers are far less likely to have huge inventories in stock. They can't break open sealed product for it. They can't order more boxes from distributers. There is no secret reserve of these cards.
In fact they have to acquire it the same way you and I do. From ebay sellers, or one at a time. They have to acquire it from other players.
What does that mean? It means that Starcitygames or CFB can't buylist much much lower than the sale price of the card on ebay.
Look if you had an underground sea NM, would you sell it to SCG at $100 when you could sell it on ebay for $160. If SCG wants underground sea, it's going to have to buy the underground sea--like everyone else. It has to play nice.
It might be able to offer a slight discount, but it can't undercut by too much because it can only acquire them from ordinary players (or other vendors).
That means that when you buy dual lands and reserve list cards, the ebay/buylist spread is going to be better. You're going to get more 'value' for every dollar you spend. Did you just spend $170 on that new NM underground sea?
Dont feel bad because it just so happens that SCG must also spend somewhere in the same ballpark for underground sea.
That leads me to my final observation: We said before that there must always be a spread between what the seller buys the card for and what they sell the card for.
If SCG buys cards in the ballpark range for $170, it must sell it for say $250.
We know something else too. SCG exhibits an economic phenomena known as price leadership. It acts like a dominant firm in the area of magic. So when SCG sells underground sea at $250, other people feel pressured to abide by that.
In other words, when SCG sets its price, the past has shown it usually isn't too long before other firms follow, and the price effectively becomes the new price.
This means that dual lands in particularly have this mild perpetual pressure of always rising. It's not to say that they must absolutely rise. Low liquidity reserve list cards all have the same argument after all.
But the perpetual rising pressure is limited by only one thing--the baseline liquidity of the dual lands as the big firms see it. (how quickly SCG and big firms are bought out)
" It means that Starcitygames or CFB can't buylist much much lower than the sale price of the card on ebay." <- Not true. The buylist price is almost always much lower than the sale price on eBay that I've seen. Otherwise, it may be worth it to trade in cards to them with the trade-in credit to purchase other cards.
This made me wonder how advanced the inventory software is for SCG and CFB. The amount of data that they have to parse on a daily basis to not only keep up but directly lead the majority of the secondary market has to be nigh unfathomable and I can't help but wonder if they have algorithms to determine how to set their buylist and retail prices. I guess they are probably literally just in meetings all day on the subject, but I can't help but wonder if there's more technology at work because the amount of information involved is just so mindnumbing when you think about tracking every card in the game.
Private Mod Note
():
Rollback Post to RevisionRollBack
Thanks to Gabgabdevo for the awesome sig image!
I'm always looking for foil Madcap Skills and Ghitu Fire-Eater, [trade thread link forthcoming]
It's also noteworthy that the price spikes (even for reserve list cards in eternal formats) don't always play out. I remember somebody trying to corner the market on Aluren a year or two back by buying out all of the available copies online, and the price jumped for a week or two, but it fell because there just wasn't enough demand to sustain those prices. I believe the same thing happened with Wasteland in 2011 or so. The price spiked and people got scared, but then it dropped back down again within a few months.
This made me wonder how advanced the inventory software is for SCG and CFB. The amount of data that they have to parse on a daily basis to not only keep up but directly lead the majority of the secondary market has to be nigh unfathomable and I can't help but wonder if they have algorithms to determine how to set their buylist and retail prices. I guess they are probably literally just in meetings all day on the subject, but I can't help but wonder if there's more technology at work because the amount of information involved is just so mindnumbing when you think about tracking every card in the game.
I'm sure that they have crawlers tracking all of this for them. If something spikes, they'll catch it and react.
I have heard vague rumors of a moustache-dispensing vending machine in a distant laundromat, across the street from a tattoo parlor. However, this information is shaky, and time is of the essence.
@onewheel, @adam: It doesn't involve that much technology and none of the sellers are that sophisticated that I know of. There's really very little need. It's like this:
Suppose you have 1000 copies of card X. You have no idea what the price is so somebody has to make an executive decision of what the initial price is - this part is manual and doesn't really matter much. Really, it doesn't. Suppose you set initial price to be $5. You decide maximum displayed quantity is 100.
You decide to list like so (quantity@$price):
900@$zz <-hidden
100@$5 <-visible
Then just display your best ask all the time, with a maximum displayed quantity. Automatically refresh inventory whenever your ask quantity (100) gets bought. Automatically back it off in increments of Y, i.e. next 100 @$10, next 100@$15, etc. but put in some sort of logic to back off in greater increments if it took 2 orders to sell the 100 vs 20. And corresponding logic to do the reverse if it takes too long to sell. Of course there's logic you can implement to improve or reduce your ask price as well when cards don't sell.
One should be able to see that when utilizing a crude algorithm of this sort, there is very little need to 'track prices of every card in the game'. Your inventory TELLS you what the market price is. Every once in awhile when you feel like it, you can manually go in and set some card prices if you feel like you want to strategically undercut your competitor or you want to hold some cards back in anticipation of buying out those cards and then listing on your own site.
But wait, what about cards that ppl discuss on market street? This is the part that requires manual intervention. Most stores are visiting market street and related threads on motl, they will know if this occurs. If it's happened already, the pricing system will know and should have flagged it. If it hasn't happened already or fully, somebody can adjust the prices manually.
Ofc if you're a smaller store you may be tempted to just 'crawl' scg's website and copy their prices. <shrugs>. It's pointless, most stores aren't going to be able to sell at SCG prices and it'll be in scg's favor if their smaller competitors aren't undercutting their prices. If you're really determined to know how everybody else's prices, sure you can try and build crawlers. But again, who cares? You know if your inventory is selling (fast) or not and can adjust prices accordingly. Besides, the list of 'hot' cards is a small list which you can see by going to most sites and just manually checking and adjusting your own prices if need be.
@onewheel: Going by your logic, you would think that executives at big trading firms and investment banks do nothing but sit in meetings to discuss what how to set stock prices You know this is silly once you think about it and I assure you this does not occur, at least, the meetings are not about what price to set but the methodology for setting them.
@Dresden: I work in ecommerce - I can assure you that crawling is incredibly easy. So easy that there's no reason not to do it, especially when pricing is volatile. What you're describing would work fine for something like presales, but falls short in the case of reactibg to buyouts like with Dark Depths, Bitterblossom, etc.
@Tomcat: I think you're essentially correct, but you've failed to account for the fact that consumers value their time. What I mean by this is, say I could sell a Liliana of the Veil on eBay or TCG. I would have to research the price and list at ~$50 if I'm a motivated seller. Then I give 10-15% of that back to the vendor, and invest say $4 into shipping costs and packaging materials. So I might walk away with $42 from the sale. But I've had to spend my time listing the card, driving to the post office, organizing my records, communicating with the buyer, etc. maybe I don't want to do that. Especially if I am just going to use that money to buy other cards - from this perspective, I'll cone out ahead if take the 25-30% trade bonus. So it's not as direct of a relationship as you make it out to be.
I have heard vague rumors of a moustache-dispensing vending machine in a distant laundromat, across the street from a tattoo parlor. However, this information is shaky, and time is of the essence.
Thanks for the input on crawling, Adam. I agree it's not rocket science and for tcg-based sites it's pretty easy. For awhile though, SCG kept changing the formats of its pages so as to not allow competitors to easily crawl for its prices, I think they have since decided not to do that anymore but I digress.
Anyway, the point is that it depends on your end goal. What I described is the (very rudimentary) basis for market-making in the financial world, where pricing is infinitely more volatile, 'buyouts' happen in milliseconds and the stakes are significantly higher. So I would argue that rather than falling short, it's a tried and proven model for making money
Let's take a look at the bitterblossom example you mentioned. Sure, we all see tcglow from ~$15 before to ~$30 now. As a STORE, your price of acquisition should be ~50% list price, so say, $7.5. On the day of a 'buyout', if you're SCG and want to retain your reputation, your initial quantity will probably get hit and you have to sell at $18 or whatever premium to tcg you listed at. That's FINE - you MADE MONEY. If you're a smaller shop and you're the first to get hit, there's absolutely nothing you can do except cancel the order if you don't want to profit and get blasted on the forums, like most small stores do.
So as a smaller store, if you want to prevent your small quantity, let's say 10 of bitterblossom to be VERY generous (most stores look like they have a quantity of 1), being sold for $15 ($75 gross profit) you can invest time and resources into crawling webpages. The end result of this is that you can (maybe) avoid being bought out at $15 and instead try to list&sell at $32. [Keep in mind that whoever is sweeping tcgplayer will probably sweep your prices if you're anywhere in line with the low, this isn't avoidable no matter how fast you're crawling.] Fine, your theoretical profit is now $245, a whole $170 more than $75), this is theoretical - do you ACTUALLY sell any/all your quantity at this price? Since you were so cognizant of the market initially, maybe you're hesitant to sell it now and lose out on more potential profit.
So you hold. And bitterblossom can drop right back to $15 like it did last year or the year before that (this card has spiked and dropped before) resulting in no sale for you. Or you can be absolutely lucky at this and pick the exact top to sell at ($170 yay!) but do you want that to be your sustainable business model, picking market tops to sell? $170 theoretical max for a card that doubled, how often are you going to perfectly execute this (and have 10 in stock execute with)? how much do you pay developers to keep up to date?
You can spend an infinite amount of time and resources on maneuvers like this. Or you can do what I said, list and sell. If a card sells out, list more at higher prices (and buylist more if you're confident). At NO POINT in between are you losing money with each transaction and since you're willing to be 'on the market' listing your prices, you will get sales that your fellow stores aren't getting by not participating. At some point you will have to decide if you want more transactions at a lower profit or much much fewer transactions at a potentially higher profit.
@Dresden: I work in ecommerce - I can assure you that crawling is incredibly easy. So easy that there's no reason not to do it, especially when pricing is volatile. What you're describing would work fine for something like presales, but falls short in the case of reactibg to buyouts like with Dark Depths, Bitterblossom, etc.
Crawling accomplishes two things in this scenario. First, it gives you data that is useful to predict when to sell. A card like Dark Depths fits certain criteria, and you can use that criteria to project how you think it's price trajectory will go.
Second, that projection (as well as the card availability data, which is also crucially important) helps inform how you set your pricing structure. Does it make sense to undercut? Should you put Dark Depths out there at 40 to test the market, or should you just list everything at 50? Should you set your reserve at 5, 10, 50?
You could do all this blind. But why would you if you don't have to? SCG stays on top by maximizing its profits. Taking the attitude of "I paid 10, so selling at 20 is fine even if I could wait a few days and get 50" is not how a business does that.
I'm not saying that the SCGs of the world set their prices 100% based on crawling the marketplace. But between the ease of setting it up and the value of the data you gather, they'd be insane not to at least be doing this to gather the data.
I have heard vague rumors of a moustache-dispensing vending machine in a distant laundromat, across the street from a tattoo parlor. However, this information is shaky, and time is of the essence.
Ah ok. Well I agree that having market competition data can be useful, as long as it doesn't cost much to do relative to infrastructure (and I'm sure an SCG-like store does this) and it's not the primary basis of setting prices. For a smaller store, not sure if it's as viable.
Btw, it's not "I paid 10, so selling at 20 is fine even if I could wait a few days and get 50". If it's always a GUARANTEED and SELLABLE 50, then yea, why would anybody do that. In reality, we know it's not always this cut and dried - the rise is ACTUAL sellable price certainly is not guaranteed, any store can list at whatever price they want (see the $250 mana crypts on tcgplayer), it doesn't mean they will sell.
To revise your statement a bit, the way I would think of it is "I paid $10 for 10, so selling 1/4 or whatnot quantity at 20 is fine to realize profits because I don't know in a few days if it's going to be 30 or 15. Also, I don't know if my competitors are going to be selling these and we start undercutting each other so selling some out now to capture guaranteed profit, reducing portfolio risk may be a good idea"
I hadn't mentioned this in my simplified model but there's also the buy-side of the equation, if you're getting hit on your bids (buy-list prices) you will probably know you should drop your ask (sell prices) and vice versa. So you're not doing this 'blind'. And even if you were, price discovery should be very quick and just a few transactions to determine where you can realistically capture profit.
What you're describing is constantly checking the market and fitting your prices to it. What I'm saying is that you really only need to check the market once when you set initial prices (and for a new card from a new set, if you're SCG, you just pick a number you think is reasonable) and utilize actual transactions to make most of the adjustments thereafter - if a card is selling at a price, you will know it and if a card is not selling at a price you will also know it. Again, the goal is to realize 'maximized profit'.
Btw, in all honesty for cards that spike or may spike, since it is a judgment call I'm sure Ben just manually adjusts the prices to the best of his judgment, utilizing whatever sources of market information he has combined with actual sales and buys. And a lotta judgment. No amount of graphing and analysis is going to guarantee you price trajectory accuracy. Read over the prediction articles on any site and you will see this. Ben@SCG has himself made some trajectory predictions that didn't work out (FoW >100 by end or 2012), etc., of course he has made many correct calls as well but I'm just saying that it's iffy.
@Tomcat: I think you're essentially correct, but you've failed to account for the fact that consumers value their time. What I mean by this is, say I could sell a Liliana of the Veil on eBay or TCG. I would have to research the price and list at ~$50 if I'm a motivated seller. Then I give 10-15% of that back to the vendor, and invest say $4 into shipping costs and packaging materials. So I might walk away with $42 from the sale. But I've had to spend my time listing the card, driving to the post office, organizing my records, communicating with the buyer, etc. maybe I don't want to do that. Especially if I am just going to use that money to buy other cards - from this perspective, I'll cone out ahead if take the 25-30% trade bonus. So it's not as direct of a relationship as you make it out to be.
Absolutely. I did not account for time and transaction costs such as shipping. Also not included is the very non-zero risk that the NM card you buy on ebay is actually Ex-. But for purposes of simplicity to better see cause and effect forces in the magic market, I didn't put then in there.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
1. The lowest cost of the card on ebay from a seller with decent feedback.
2. The buylist price from big name sellers --CFB, SCG
3. The sale price from the big name sellers.
Let's call the difference in price between what CFB sells the card for, and what it will buy the card back for, Sale Price/Buylist spread. (borrowing terms from finance)
Because the buylist of these stores represents their "demand" for the cards, to me it stands to reason that the higher the spread, the more fluff you're buying.
What do I mean? Well...Suppose hypothetically that you buy a card from CFB for $40. But suppose also that the buylist only values the card at $5.
In that case, you're buying something for $40 that the seller him/herself is only willing to pay $5 for.
On the other hand, if you're buying something for $40 --Liliana of the Veil, and CFB's buylist price is $35, then you're buying a card that CFB has to spend $35 to acquire.
In one of these cases, the price is "fluffier". It could be hype, it could be temporary speculation, it could be that the seller has too many of them in their inventory. But for whatever reason, if they need more, they're at this time only willing to pay a small fraction of the cost.
In fact the larger the spread (the difference between the sale price and buylist), the more likely it is that the money you're putting into the card is something for which you are not being compensated adequate value.
There has to be at least some spread of course, because that's how the sellers make their money. In other words CFB can't sell you LOTV at $40 and buy it back at $40 on their Buylist. It wouldn't make sense.
Enter now, the world of Ebay. Ebay is one of the primary venues through which the individual magic player can turn their cards into money again. In an efficient market, ebay sellers should only be able to sell when their cards are below the price of the big seller.
Because you're getting the card from ebay cheaper than the big sellers, you now have another salient data point. The difference of the ebay price over the buylist of the current big sellers. (ebay/buylist spread)
Suppose CFB is buying Liliana of the Veil at $35. (they are at the moment)
Suppose you see a reputable ebay seller selling NM copies at $35. Should you buy? For me the answer is always yes.
In fact, after ebay bucks (2% cash back), and additional cash back from your credit card, you will actually come away with some small profit.
These two data points can give us some information.
1. The ebay/buylist spread tells us something about absolute profitability. For me, its pretty much the final arbiter on whether or not I might buy.
2. The Sale Price/buylist spread on the other hand tells us something about inflated hype, how many more copies the seller has in stock, or even low sales volume(low liquidity) and gives us more information on how the seller views the card.
Dual lands are a particularly interesting case study in all this. Why? Well let's start off with something about new cards.
Theros, Innistrad, RTR. Does anyone really know how many of these cards were printed? How many boxes sold? I certainly don't. As an individual buyer, if I choose to speculate, I'm going in the dark here completely.
If I choose to go all in and buy up a bunch of Snapcaster Mages, I have no idea whether Starcity has 25 left in stock or 250. And so that is why the buylist price of the big vendor is pretty important-it reveals their demand for the card.
But what about dual lands? These are different. Being twenty years old, on the reserve list, big sellers are far less likely to have huge inventories in stock. They can't break open sealed product for it. They can't order more boxes from distributers. There is no secret reserve of these cards.
In fact they have to acquire it the same way you and I do. From ebay sellers, or one at a time. They have to acquire it from other players.
What does that mean? It means that Starcitygames or CFB can't buylist much much lower than the sale price of the card on ebay.
Look if you had an underground sea NM, would you sell it to SCG at $100 when you could sell it on ebay for $160. If SCG wants underground sea, it's going to have to buy the underground sea--like everyone else. It has to play nice.
It might be able to offer a slight discount, but it can't undercut by too much because it can only acquire them from ordinary players (or other vendors).
That means that when you buy dual lands and reserve list cards, the ebay/buylist spread is going to be better. You're going to get more 'value' for every dollar you spend. Did you just spend $170 on that new NM underground sea?
Dont feel bad because it just so happens that SCG must also spend somewhere in the same ballpark for underground sea.
That leads me to my final observation: We said before that there must always be a spread between what the seller buys the card for and what they sell the card for.
If SCG buys cards in the ballpark range for $170, it must sell it for say $250.
We know something else too. SCG exhibits an economic phenomena known as price leadership. It acts like a dominant firm in the area of magic. So when SCG sells underground sea at $250, other people feel pressured to abide by that.
In other words, when SCG sets its price, the past has shown it usually isn't too long before other firms follow, and the price effectively becomes the new price.
This means that dual lands in particularly have this mild perpetual pressure of always rising. It's not to say that they must absolutely rise. Low liquidity reserve list cards all have the same argument after all.
But the perpetual rising pressure is limited by only one thing--the baseline liquidity of the dual lands as the big firms see it. (how quickly SCG and big firms are bought out)
The rest of what you said is pretty right though
Thanks to Gabgabdevo for the awesome sig image!
I'm always looking for foil Madcap Skills and Ghitu Fire-Eater, [trade thread link forthcoming]
I'm sure that they have crawlers tracking all of this for them. If something spikes, they'll catch it and react.
Suppose you have 1000 copies of card X. You have no idea what the price is so somebody has to make an executive decision of what the initial price is - this part is manual and doesn't really matter much. Really, it doesn't. Suppose you set initial price to be $5. You decide maximum displayed quantity is 100.
You decide to list like so (quantity@$price):
900@$zz <-hidden
100@$5 <-visible
Then just display your best ask all the time, with a maximum displayed quantity. Automatically refresh inventory whenever your ask quantity (100) gets bought. Automatically back it off in increments of Y, i.e. next 100 @$10, next 100@$15, etc. but put in some sort of logic to back off in greater increments if it took 2 orders to sell the 100 vs 20. And corresponding logic to do the reverse if it takes too long to sell. Of course there's logic you can implement to improve or reduce your ask price as well when cards don't sell.
One should be able to see that when utilizing a crude algorithm of this sort, there is very little need to 'track prices of every card in the game'. Your inventory TELLS you what the market price is. Every once in awhile when you feel like it, you can manually go in and set some card prices if you feel like you want to strategically undercut your competitor or you want to hold some cards back in anticipation of buying out those cards and then listing on your own site.
But wait, what about cards that ppl discuss on market street? This is the part that requires manual intervention. Most stores are visiting market street and related threads on motl, they will know if this occurs. If it's happened already, the pricing system will know and should have flagged it. If it hasn't happened already or fully, somebody can adjust the prices manually.
Ofc if you're a smaller store you may be tempted to just 'crawl' scg's website and copy their prices. <shrugs>. It's pointless, most stores aren't going to be able to sell at SCG prices and it'll be in scg's favor if their smaller competitors aren't undercutting their prices. If you're really determined to know how everybody else's prices, sure you can try and build crawlers. But again, who cares? You know if your inventory is selling (fast) or not and can adjust prices accordingly. Besides, the list of 'hot' cards is a small list which you can see by going to most sites and just manually checking and adjusting your own prices if need be.
@onewheel: Going by your logic, you would think that executives at big trading firms and investment banks do nothing but sit in meetings to discuss what how to set stock prices You know this is silly once you think about it and I assure you this does not occur, at least, the meetings are not about what price to set but the methodology for setting them.
@Tomcat: I think you're essentially correct, but you've failed to account for the fact that consumers value their time. What I mean by this is, say I could sell a Liliana of the Veil on eBay or TCG. I would have to research the price and list at ~$50 if I'm a motivated seller. Then I give 10-15% of that back to the vendor, and invest say $4 into shipping costs and packaging materials. So I might walk away with $42 from the sale. But I've had to spend my time listing the card, driving to the post office, organizing my records, communicating with the buyer, etc. maybe I don't want to do that. Especially if I am just going to use that money to buy other cards - from this perspective, I'll cone out ahead if take the 25-30% trade bonus. So it's not as direct of a relationship as you make it out to be.
Anyway, the point is that it depends on your end goal. What I described is the (very rudimentary) basis for market-making in the financial world, where pricing is infinitely more volatile, 'buyouts' happen in milliseconds and the stakes are significantly higher. So I would argue that rather than falling short, it's a tried and proven model for making money
Let's take a look at the bitterblossom example you mentioned. Sure, we all see tcglow from ~$15 before to ~$30 now. As a STORE, your price of acquisition should be ~50% list price, so say, $7.5. On the day of a 'buyout', if you're SCG and want to retain your reputation, your initial quantity will probably get hit and you have to sell at $18 or whatever premium to tcg you listed at. That's FINE - you MADE MONEY. If you're a smaller shop and you're the first to get hit, there's absolutely nothing you can do except cancel the order if you don't want to profit and get blasted on the forums, like most small stores do.
So as a smaller store, if you want to prevent your small quantity, let's say 10 of bitterblossom to be VERY generous (most stores look like they have a quantity of 1), being sold for $15 ($75 gross profit) you can invest time and resources into crawling webpages. The end result of this is that you can (maybe) avoid being bought out at $15 and instead try to list&sell at $32. [Keep in mind that whoever is sweeping tcgplayer will probably sweep your prices if you're anywhere in line with the low, this isn't avoidable no matter how fast you're crawling.] Fine, your theoretical profit is now $245, a whole $170 more than $75), this is theoretical - do you ACTUALLY sell any/all your quantity at this price? Since you were so cognizant of the market initially, maybe you're hesitant to sell it now and lose out on more potential profit.
So you hold. And bitterblossom can drop right back to $15 like it did last year or the year before that (this card has spiked and dropped before) resulting in no sale for you. Or you can be absolutely lucky at this and pick the exact top to sell at ($170 yay!) but do you want that to be your sustainable business model, picking market tops to sell? $170 theoretical max for a card that doubled, how often are you going to perfectly execute this (and have 10 in stock execute with)? how much do you pay developers to keep up to date?
You can spend an infinite amount of time and resources on maneuvers like this. Or you can do what I said, list and sell. If a card sells out, list more at higher prices (and buylist more if you're confident). At NO POINT in between are you losing money with each transaction and since you're willing to be 'on the market' listing your prices, you will get sales that your fellow stores aren't getting by not participating. At some point you will have to decide if you want more transactions at a lower profit or much much fewer transactions at a potentially higher profit.
Second, that projection (as well as the card availability data, which is also crucially important) helps inform how you set your pricing structure. Does it make sense to undercut? Should you put Dark Depths out there at 40 to test the market, or should you just list everything at 50? Should you set your reserve at 5, 10, 50?
You could do all this blind. But why would you if you don't have to? SCG stays on top by maximizing its profits. Taking the attitude of "I paid 10, so selling at 20 is fine even if I could wait a few days and get 50" is not how a business does that.
I'm not saying that the SCGs of the world set their prices 100% based on crawling the marketplace. But between the ease of setting it up and the value of the data you gather, they'd be insane not to at least be doing this to gather the data.
Btw, it's not "I paid 10, so selling at 20 is fine even if I could wait a few days and get 50". If it's always a GUARANTEED and SELLABLE 50, then yea, why would anybody do that. In reality, we know it's not always this cut and dried - the rise is ACTUAL sellable price certainly is not guaranteed, any store can list at whatever price they want (see the $250 mana crypts on tcgplayer), it doesn't mean they will sell.
To revise your statement a bit, the way I would think of it is "I paid $10 for 10, so selling 1/4 or whatnot quantity at 20 is fine to realize profits because I don't know in a few days if it's going to be 30 or 15. Also, I don't know if my competitors are going to be selling these and we start undercutting each other so selling some out now to capture guaranteed profit, reducing portfolio risk may be a good idea"
I hadn't mentioned this in my simplified model but there's also the buy-side of the equation, if you're getting hit on your bids (buy-list prices) you will probably know you should drop your ask (sell prices) and vice versa. So you're not doing this 'blind'. And even if you were, price discovery should be very quick and just a few transactions to determine where you can realistically capture profit.
What you're describing is constantly checking the market and fitting your prices to it. What I'm saying is that you really only need to check the market once when you set initial prices (and for a new card from a new set, if you're SCG, you just pick a number you think is reasonable) and utilize actual transactions to make most of the adjustments thereafter - if a card is selling at a price, you will know it and if a card is not selling at a price you will also know it. Again, the goal is to realize 'maximized profit'.
Btw, in all honesty for cards that spike or may spike, since it is a judgment call I'm sure Ben just manually adjusts the prices to the best of his judgment, utilizing whatever sources of market information he has combined with actual sales and buys. And a lotta judgment. No amount of graphing and analysis is going to guarantee you price trajectory accuracy. Read over the prediction articles on any site and you will see this. Ben@SCG has himself made some trajectory predictions that didn't work out (FoW >100 by end or 2012), etc., of course he has made many correct calls as well but I'm just saying that it's iffy.
Absolutely. I did not account for time and transaction costs such as shipping. Also not included is the very non-zero risk that the NM card you buy on ebay is actually Ex-. But for purposes of simplicity to better see cause and effect forces in the magic market, I didn't put then in there.