I hate to knit pick, but I do enjoy statistics and I believe for any algorithm or formula to work you need at least 30 data points. The last time I did any analysis was a couple yrs ago so I may be wrong. What you laid out is a thesis sry :/ But I do largely agree with your thesis. With mathematics, making a claim is a lot different than other disciplines (maybe you know this? Idk your background).

great read, thanks!

I would like to say this should be stickied. The amount of detail and concentration to conjure up this should be respected.

Jone, I agree with your findings. Although it does not come into play when it reflects rarity.

For example, I've been on the hunt for a sealed Mighty Final Fight...forever. They have all ranged from 80-85+ gold selling for 700, 900, 500, 600. The 85+ being the 700...and the 80 being the 900.

Would be really cool if we can tape into Ebay's data to better understand different variables; seaons (both climate and economic), trends (when a youtuber puts out a 'hidden gem' video and price spikes) and so on.

Great read. Hope more of these types of posts follow in NA

Yes, he's still out there lurking in the shadows.

 

aaah. call in donkey kong jr. too much math here.

This might be one of the single-best threads/posts I've seen on this site. I'm on my way out at the moment, but I'm going to weigh in when I return later! Excellent work and post, jonebone! 10/10!

I have a math degree myself.  And I've been in collectibles long enough to consider this theory intuitive, but I did gather up some data points for you.

First, I can't really relate to loose carts so I'm not too concerned with those.  I'm looking at them as collectibles, so I'm mostly focused on the VGA stuff.  Unfortunately games are still relatively new as collectibles, so there aren't a lot of graded items out there.  But still, by using a small sampling of some "Good" titles, you can see how the bell curve peaks around VGA 85, tail skewed to the left slightly.  All of these population reports had been requested by me sometime in the last year, may be a couple of adds since then:



I could do it for a lot more titles, but the graph becomes "too busy" at some point.  

I also come from cards where there's a lot of data to go on.  I'd tend to pick rookies since they are the most iconic cards and the most sought after.  You also have to stick within an era... the newer the card, the more copies that have been subbed.  And here's a quick look at the two best cards from the 60s, the Nolan Ryan 1968 Rookie and the Reggie Jackson 1969 rookie... which may as well be exact Bell Curves.



Does this give my theory any more credibility

Great read. I wanted to add my $ 0.02 on the subject as it was a part of my position in a panel a couple guys and I did with at PAX last year.

The topic of the panel was specifically on our position speculating the future trending of "retro" video game collecting. There are a number of philosophies/definitions of what makes a game "retro" but for the sake of simplicity, let's call it something 4 generations or older (I.e. In the Wii U, Wii, Game Cube, N64, SNES, NES timeline, retro would include anything N64 and older). Another factor which directly impacts supply (as a factor of cost) is the relevant console of someone's youth as applies to an audience entering into an age of fiscal capabilities (let's call it 28...out of college with a few years of professional experience). Since I am a NES collector, I will use that console for my example. I will roughly age bracket the audience that recognizes the NES as a component of youthful nostalgia as anyone that was 6-14 during its lifespan (approximately those born between 1971 and 1988). I'm on a bit of a tangent but stay with me. That would make the nostalgic resurgence for NES start in 1999 and theoretically will peak out about 2016. Unfortunately Pricecharting only has 6 years of data but if you look at the trend, it is on a steep upward climb since 2008. Looking at the SNES and N64 trends, there are following suit. I theorize a huge factor impacting the recent price hike has been directly a result of the aforementioned raise in nostalgia audience. If that were the only factor then the prices should start to drop off around the end of next year and onward...but it's not the only factor.

As you will learn day 1 of any Econ class, supply is the leading cause in price establishment and fluctuation. Supply on NES games has many factors working against it:
1. Age - Wear and tear take their toll and some games are no longer played, lost, destroyed, etc.
2. Collectors - Since I started collecting, I have never once sold or traded a game in my collection (not including dupes). Collectors like me (as I assume many of you are) are not likely to reintroduce a game back into the market outside of extenuating circumstances (economic duress, wife aggro, children, available space, relocation, etc.).
3. Increased audience - Nostalgia doesn't affect everyone at the same time. It hit me about 5 years ago and I've been high-speed into collecting ever since. My go-to guru started about 15 years ago and he is 3 years younger than me. Nostalgia combined with disposable income is also not necessarily the only reason someone gets into collecting. In addition to readily available financial means, social affinity has an impact on the size of the audience. "Retro culture" is quite popular today. There were kids at the panel that were not old enough to have lived during the NES lifetime yet it still appeals to them as a result of social impact and popularity.
4. Fixed volume – The last licensed NES game was released 21 years ago and the total number of games ever produced will NEVER increase.

Now let's get back to the point of the thread. I like the bell curve for a number of reasons and I have a couple thoughts I'd like to open up for discussion. With strict regard for condition vs. demand, there are a few additional variables that I think need to be mentioned. Demand acts as a calculable value impacted by the following forces:
1. Fiscal capability – Whether or not someone is able to afford a mint CIB SE has an apparent impact on demand.
2. Availability – In this capacity, I am specifically referring to the production count of the game.
3. Need – This one is tricky. A game like Top Gun has a low demand because the gameplay sucks (see point 4), it is very common (see 2), and I would speculate that most "collectors" already have it. I "need" 193 games to complete my NES collection but I am not focusing on all of them with the same vigor. Because I believe the price will continue to rise and there will always be a demand to keep the prices stable, I am trying to get the most rare/expensive games out of the way to minimize the impact of inflated price on my wallet. I.e. My CIB copy of Little Samson purchased for $730 from Penguin back in 2012...now going for about 2x. That is much worse than the inflated cost of Addams Family, a common I still need (2012 cart price - $5, today - $10).
4. Gameplay – SMB3 is not a rare game by any means but because it is highly regarded as one of the greatest games in the NES library (or overall), the demand stays pretty high as it appeals to almost every new collector and extends to casuals as well.
5. Niche of a Niche of a Niche – I am going to assume most (completionist) collectors are "Cart Only" collectors. I am a CIB collector (niche) including unlicensed titles (niche). I also collect one-off and other rare items (NTF2 Test cart, ET dig Atari cart, certain protos and repro's (niche). "Niche" could also refer to items outside of a collector's typical set (i.e. I have CIB version of many of my favorite SNES games as well as NIB copies of both original and Player's Choice Zelda LTTP because it is my favorite game of all time). I could care less about most CIB SNES games because it doesn't affect me the same way NES does. It is important to factor in niche items when evaluating realizable cost as "value". This leads into condition as a "niche".

Most games in the library fit this model but there are outliers. Rad Racket II, is a very rare game, especially the box, but not highly priced. It could be a combination of being unlicensed (niche, see point 5), poor gameplay, or one of many crappy sports games, but nonetheless, doesn't compare in value to other "rare" titles. My copy is pretty average (a 7 for me) and there is a very small probability that I will ever intentionally seek out another copy in better condition. That is not the case for all games. My Contra Force is near mint. The box has no visible markings/damage, the manual looks like it was never handled and the cart has probably been in a NES less than a hand full of times (mostly because it's a total shit game). Because of the pristine condition, I paid $275 for it back in 2013 when CIBs were going for between $125-200. I don't have a good reason for paying the premium other than the simple fact that I knew I wasn't going to come across another copy in that condition for a long time if ever and I felt it was worth the price. I am probably not the best baseline as I have a very specific rule about replacing items in my collection with better condition copies (will actively seek out a replace for any item that is less than a 7 on my scale; 7 being some visible wear, box is fully intact but has some visible blemishes).

The point I'm getting at is that, while I think the bell curve of cost/demand is accurate, I would include all impacting factors when evaluating "why" a game would exist in the 3rd sigma subset above std dev and the "score" is result of a collection of all attributes.

I will attempt to prove my theory with the following item examples:
Example 1: Blades of Steel CIB (Mint Condition)
- Cheap; High Fiscal Capability
- Highly Available
- Average Need (good for new collectors)
- Good gameplay
- Somewhat niche (b/c CIB)
The bell curve would serve correctly because when using "comparing value to buyer" over "price", it would be a very small population that would be willing to pay a premium for this game, regardless of condition.

Example 2: Little Samson CIB (Mint Condition)
- Expensive (Out of the range of sanity for many collectors; See Wife Aggro)
- Low Production
- Average Need (highly desired to those of us "in the know" but lost to most folks)
- Fantastic gameplay (one of the best games on the NES)
- Niche (b/c CIB, lacks nostalgia for most, mostly desired by completionists only)
The curve would peak early and drop off fast, regardless of condition.

Example 3: Pro Sport Hockey (Mint Condition)
- Higher than average price among all games
- Low Production (Late NES release, Released on both NES and SNES)
- Low Need
- Poor Gameplay (There are better hockey titles on NES)
- Niche (b/c CIB, lacks nostalgia for most, mostly desired by completionists only)

Despite a quite a few differences in attributes, Pro Sport Hockey's curve would resemble Little Samson. While condition might have an influence on price (mostly because the box has a rarity of 9 on NA), I don't believe it is the prime influence.
So this has been a REALLY long rant but, in case it wasn't obvious, this is a subject that really interests me. I would love to see data to either support or disprove my theory. If I can find a little time, I'll put together a spreadsheet to collect purchase information so I can have some points to compare against.

What do you guys think? Am I crazy? Will condition still have a grip on impacting price even if demand significantly dropped off?

This data is good, though I think there is an obscured factor involved when considering VGA graded games. I have a number of NIB games that aren't graded, mostly because it doesn't mean anything to me and I can't justify the price. Additionally though, the conditions aren't especially good so I don't see a lot of value from getting them graded. I believe the numbers you see for VGA graded games is a subset of the available NIB games in that only those that are already perceived as being in "better than average" condition are likely to be submitted. I'm not going to send them my Back to the Future II&III because it's sealed, sure, but it's not a great copy. The only game I have even thought about getting graded is my Ninja Gaiden because it is absolutely lovely and I'd be interested to know just how lovely they think it is. 

I'm sure my math minor can't hold water to your major but I do think this is a much more complicated equation and there are probably a bunch of factors I'm still not including (perhaps certain regions received more copies of a certain game and/or avg AGI of residents of some states impacts the price of games, and so on).

Yep. Condition and collectibility will always be king. Even the most common of common sports games on the NES still command a premium over their cart only, boxed, and CIB bretheren if they are mint or near mint. It just means that cart only games in the worst shape become worthless, acceptable and good condition CB and CIBs go down in price a decent amount, and mint and near mint CIB copies goes down in price at a slower rate, if at all, because they still have rarity and desirability.

 

Good writeup, but my Bell Curve is completely separate from pricing. If you were coming up with a pricing equation (probably can't be modelled that accurately though), then yes you'd need to consider many of those factors. You'd need to factor in demographics of the buyer market (income, age, location (domestic / international), macroeconomic impacts, etc.) as well as demographics of the supply (condition, rarity, availability, popularity, gameplay, publisher, platform, etc.)

And then you have the arbitrary variable that you simply cannot factor... hype. Hype influences a game's value far above and beyond anything else. Youtube videos, forum speculating and internet articles do wonders for a game once it catches on. Stuff like Earthbound, Hagane, Little Samson, etc. Once the hype train takes off, you'll never get it for cheap again... and then it'll be targeted for counterfeiting as well.

So interesting comments, but those are really outside the scope of what I was doing here. If you want to get into a pricing theory article, I could do one of those later on too. The fundamental idea there would be distinguishing between condition sensitive and pure rare items. Something like an NWC is pure rare, where the cost of ownership is high, and the delta between Mint and poor will not be that great (on a percentage basis). Compared to something like a sealed Mario on NES... where the cost of ownership will be substantial (but not nearly as high), yet the delta between Mint and Poor will orders of magnitude higher.