An in depth analysis into the price of minerals

I’m a VR player, and I find it breaks immersion having to lift the headset up and look at the 3rd party tools all the time, I wondered if there was another way. So, I’ve tapped into the EDDN and started saving all the commodity data. Over the last few days I’ve collected over 4.5m datapoints for the prices of minerals; I’ve done some analysis on this dataset which I would like to share. None of the following takes into consideration the time it takes to mine, this is all about the price you get per tonne.

Hypothesis:

Commanders who contribute data to the 3rd party tools are only a subset of the commanders who use these tools to mine and sell minerals. As this maybe self-fulfilling, only a subset of stations that have high prices will ever be seen within the tools.

Therefore, can we identify likely stations that will buy minerals at relatively high prices, through in game tools and exploration alone?


Findings:

What are these charts all about?
  • These box whisker plots show the prices of minerals for each quartile by volume, i.e. 25% of the prices sit within the Grey box and 25% within the orange box.
  • This can be a better representation of the true value of the mineral, as the upper and lower quartiles are less skewed by extreme prices.
  • You may also find these are closer to the actual prices you'll see when travelling the galaxy.

1580227680022.png

  • Over the last week, Low Temperature Diamond prices have far exceeded Void Opals.
  • Both Void Opals and Low Temperature Diamonds have unusually long upper whiskers, indicating the 25% of the data they cover is spread over a much larger price range. Making it difficult to find these high prices.
  • Void Opals and Painite having a much lower average and price distribution over other minerals. Is this an effect of the player community.
  • Benitoite has the 3rd highest max price, it also has the 2nd highest upper quartile price and the 2nd highest average price. This should be much easier to find a market with a high price.
  • Metaalloys although being relatively low in price have a very consistent range in which they sell for, suggesting these minerals are not affected by supply and demand.

The rest of the analysis specifically focuses on Void Opals, Low temperature diamonds, Painite and Benitoite.


How does Station Economy affect price?
1580227722778.png

  • Refineries offer the best price for LTD however they also have the lowest average, the variation is too extreme to find a consistently high price.
  • Industrial and Extraction offer high upper quartiles, suggesting these economies could be the best place to sell your LTD’s.
  • High Tech and Tourism offers the most consistent price however this means there is little chance of getting a very high price for your LTD’s
  • This picture is very different for Void Opals, High Tech offer some of the lowest prices with the occasional high extremes.
  • Tourism economies have the best upper quartile price but relatively low extremes. However, based on this view Void Opals should be avoided.
  • Painite and Benitoite follow a similar pattern indicating they can be sold together.
  • High Tech and Industrial Economies have both strong upper quartiles and strong high prices.

How does the type of station affect prices?
1580227874838.png

  • Except for Benitoite, Asteroid bases seem to the place the sell your minerals, if you can find one.
  • Although Bernal’s or Ocellus Stations are rare, they do provide a good consistent distribution of prices for your minerals
  • With Orbis stations standing out for consistently high prices for your Benitoite.
  • Void opals should be avoided.

How does the controlling faction’s government affect prices?
1580227920112.png

  • The Corporate governments seem to pay the best prices for LTD's
  • Benitoite can be sold almost anywhere just avoid Democracies and arguably Patronage.
  • Painite can also be sold anyway although you won’t get a high price at dictatorship.

How does the factions allegiance affect prices?
1580228116577.png

  • LTD’s show some differences between allegiances, Empire offers the best prices most of the time with Independents offering extreme high prices.
  • Void Opals should be avoided.
  • Benitoite and Painite can be sold anywhere.

How does the factions current state affect prices?
1580228155590.png

  • Expansion states are the places to sell your LTD’s
  • Void Opals should be avoided.
  • Benitoite and Painite can be sold anywhere, maybe avoid factions in investment states for Painite.

Conclusion:

Low Temperature Diamonds consistently offer the highest price per tonne, the places to sell your LTD’s are;
  • Industrial and Extraction Economies
  • Asteroid and Bernal/Ocellus Stations
  • Corporate Governments
  • Empire Allegiances
  • In my sample of 1,148 stations, 23 meet the above criteria, 3 of which are Ocellus stations.

Benitoite offers high and consistent prices in a variety of locations, these include:
  • High Tech and Industrial Economies
  • Orbis Stations for Benitoite
  • Not Democracies or Patronage.
  • 71 Stations meet the criteria.

Painite can be sold in many locations with the following criteria:
  • High Tech and Industrial Economies
  • Not Dictatorship.
  • Not Investment
  • 464 Stations meet the criteria.

But what does this all mean for Void Opals?
  • You can’t use the above method for finding stations to sell your Vopals.
  • This doesn’t mean you should stop mining them, you just need to use the 3rd party tools.

Next Steps:
My plan is to start book marking some of the stations for LTD’s and Benitoite that are closest to my favourite mining hotspot for that mineral, and if there is a good market price check on the 3rd party tools to see if it is there.
I will rerun all this analysis once I have more data, maybe a months’ worth and do some trend analysis based on the above criteria.
Future analysis to include trade runs, maybe with a travelling salesman approach and black market trading, there must be a way to make money here.

Hope you found it interesting.
o7
 
Nice work man! I knew LTD's were holding up better but that's telling. I'm also switching to the other 'ites'. Makes rocky rings worth a look these days and I made more money faster off of those than Painite.
 
Good work on the analysis, and great to see someone else looking seriously into market behaviours.


Void Opals and Painite having a much lower average and price distribution over other minerals. Is this an effect of the player community.
Yes - those were the two that were mined the most before 3.6, so the two where demand is most likely to have started out zero or near-zero at many stations. This may well even out over time, though another factor that will affect this is that absolute demand sizes of Painite and especially Diamonds tend to be much higher than the other gems ... but they can also be laser-mined which speeds up collection.

Metaalloys although being relatively low in price have a very consistent range in which they sell for, suggesting these minerals are not affected by supply and demand.
They almost certainly are, but it's so slow to get meta alloys (barnacle hunting rather than mining) compared with the size of the demand at a station that the chances of players knocking meta-alloys off the "maximum demand" point is minimal.


Some areas where I have a difference of opinion:

Economy, government type, station type, allegiance: I suspect any differences here are primarily due to small sample size - especially clear on Asteroid bases, where there are very few of them at all and the recent Public Holiday at Moore's Charm (https://cdb.sotl.org.uk/stations/64/trade/62?minrange=3306-01-20&maxrange=3306-01-28) was probably skewing the sample a fair bit on its own. But even high-tech and tourism economies are relatively rare. Which stations are more likely to get more player traffic is also a possible statistical issue depending on how you're processing the EDDN data.

The slight bias towards the Empire in the median prices is possibly because Aisling Duval gives a 10% price boost on "high value goods" in her control systems, and a lot of them are Imperial. Though I suspect that doesn't explain all of it. (Moore's Charm is Independent, for the extreme on the LTD prices)

States are the crucial factor (and states and demand explain almost all the variation in price, as far as I can tell) - things to consider here:
- a station can be in multiple states at once and their effects stack. The combined effect of this is what's causing most of the really high outliers in your measure.
- you haven't looked at Civil Liberty, Public Holiday or Pirate Attack, states which cause extremely significant increases in price.
- you also haven't looked at None, as a control.
The differences from economy, allegiance, government and station type may well reflect somewhat how likely it is that a system has enough player activity to get into these states in the first place.

The big thing where something is not quite right is you've said "not Investment" for Painite, despite Investment giving a ~2.5x increase in Painite prices compared to a neutral economy, and ~1.4x compared with Boom. (See for example: https://whdb.sotl.org.uk/stations/13/trade/63?minrange=3305-12-27&maxrange=3306-01-28 - hover points to see the applicable states)
 
Glad I interrupted my game to use this 3rd party too, nice one Cmdr!

If the markets are truly fluent you should see your bookmarked station prices go down but if this underlying data is right it at least gives a basis for best sells.

Purely from a RP perspective, I like the data that different economies etc do want more specific commodities and minerals.
 
Last edited:
Too bad the ingame tools are not up to the task.

They should really allow multiple filtering options in galmap (economy types, multiple states, various commodities/demands, etc) and they should allow the results coming in not based on personal experience (visiting every station to find market info is so middle age) but at least by a certain radius.
I mean ship's AI is mandatory and it should be able to automatically convey/extract informations to/from a central database regarding any populated world.
 
I’m a VR player, and I find it breaks immersion having to lift the headset up and look at the 3rd party tools all the time, I wondered if there was another way. So, I’ve tapped into the EDDN and started saving all the commodity data. Over the last few days I’ve collected over 4.5m datapoints for the prices of minerals; I’ve done some analysis on this dataset which I would like to share. None of the following takes into consideration the time it takes to mine, this is all about the price you get per tonne.

Hypothesis:

Commanders who contribute data to the 3rd party tools are only a subset of the commanders who use these tools to mine and sell minerals. As this maybe self-fulfilling, only a subset of stations that have high prices will ever be seen within the tools.

Therefore, can we identify likely stations that will buy minerals at relatively high prices, through in game tools and exploration alone?


Findings:

What are these charts all about?
  • These box whisker plots show the prices of minerals for each quartile by volume, i.e. 25% of the prices sit within the Grey box and 25% within the orange box.
  • This can be a better representation of the true value of the mineral, as the upper and lower quartiles are less skewed by extreme prices.
  • You may also find these are closer to the actual prices you'll see when travelling the galaxy.

View attachment 160435
  • Over the last week, Low Temperature Diamond prices have far exceeded Void Opals.
  • Both Void Opals and Low Temperature Diamonds have unusually long upper whiskers, indicating the 25% of the data they cover is spread over a much larger price range. Making it difficult to find these high prices.
  • Void Opals and Painite having a much lower average and price distribution over other minerals. Is this an effect of the player community.
  • Benitoite has the 3rd highest max price, it also has the 2nd highest upper quartile price and the 2nd highest average price. This should be much easier to find a market with a high price.
  • Metaalloys although being relatively low in price have a very consistent range in which they sell for, suggesting these minerals are not affected by supply and demand.

The rest of the analysis specifically focuses on Void Opals, Low temperature diamonds, Painite and Benitoite.


How does Station Economy affect price?
View attachment 160436
  • Refineries offer the best price for LTD however they also have the lowest average, the variation is too extreme to find a consistently high price.
  • Industrial and Extraction offer high upper quartiles, suggesting these economies could be the best place to sell your LTD’s.
  • High Tech and Tourism offers the most consistent price however this means there is little chance of getting a very high price for your LTD’s
  • This picture is very different for Void Opals, High Tech offer some of the lowest prices with the occasional high extremes.
  • Tourism economies have the best upper quartile price but relatively low extremes. However, based on this view Void Opals should be avoided.
  • Painite and Benitoite follow a similar pattern indicating they can be sold together.
  • High Tech and Industrial Economies have both strong upper quartiles and strong high prices.

How does the type of station affect prices?
View attachment 160439
  • Except for Benitoite, Asteroid bases seem to the place the sell your minerals, if you can find one.
  • Although Bernal’s or Ocellus Stations are rare, they do provide a good consistent distribution of prices for your minerals
  • With Orbis stations standing out for consistently high prices for your Benitoite.
  • Void opals should be avoided.

How does the controlling faction’s government affect prices?
View attachment 160441

  • The Corporate governments seem to pay the best prices for LTD's
  • Benitoite can be sold almost anywhere just avoid Democracies and arguably Patronage.
  • Painite can also be sold anyway although you won’t get a high price at dictatorship.

How does the factions allegiance affect prices?
View attachment 160445
  • LTD’s show some differences between allegiances, Empire offers the best prices most of the time with Independents offering extreme high prices.
  • Void Opals should be avoided.
  • Benitoite and Painite can be sold anywhere.

How does the factions current state affect prices?
View attachment 160447

  • Expansion states are the places to sell your LTD’s
  • Void Opals should be avoided.
  • Benitoite and Painite can be sold anywhere, maybe avoid factions in investment states for Painite.

Conclusion:

Low Temperature Diamonds consistently offer the highest price per tonne, the places to sell your LTD’s are;
  • Industrial and Extraction Economies
  • Asteroid and Bernal/Ocellus Stations
  • Corporate Governments
  • Empire Allegiances
  • In my sample of 1,148 stations, 23 meet the above criteria, 3 of which are Ocellus stations.

Benitoite offers high and consistent prices in a variety of locations, these include:
  • High Tech and Industrial Economies
  • Orbis Stations for Benitoite
  • Not Democracies or Patronage.
  • 71 Stations meet the criteria.

Painite can be sold in many locations with the following criteria:
  • High Tech and Industrial Economies
  • Not Dictatorship.
  • Not Investment
  • 464 Stations meet the criteria.

But what does this all mean for Void Opals?
  • You can’t use the above method for finding stations to sell your Vopals.
  • This doesn’t mean you should stop mining them, you just need to use the 3rd party tools.

Next Steps:
My plan is to start book marking some of the stations for LTD’s and Benitoite that are closest to my favourite mining hotspot for that mineral, and if there is a good market price check on the 3rd party tools to see if it is there.
I will rerun all this analysis once I have more data, maybe a months’ worth and do some trend analysis based on the above criteria.
Future analysis to include trade runs, maybe with a travelling salesman approach and black market trading, there must be a way to make money here.

Hope you found it interesting.
o7

Blimey, early candidate for Post of the Year! Brilliant work, Commander!

Source: https://media.giphy.com/media/NnGGHE0muVqpO/giphy.gif
 
Good work on the analysis, and great to see someone else looking seriously into market behaviours.
Thanks for the feedback, it's much appreciated, see my responses.

Economy, government type, station type, allegiance: I suspect any differences here are primarily due to small sample size - especially clear on Asteroid bases, where there are very few of them at all and the recent Public Holiday at Moore's Charm (https://cdb.sotl.org.uk/stations/64/trade/62?minrange=3306-01-20&maxrange=3306-01-28) was probably skewing the sample a fair bit on its own. But even high-tech and tourism economies are relatively rare. Which stations are more likely to get more player traffic is also a possible statistical issue depending on how you're processing the EDDN data.
I'm not sure what the number of stations out there is, there is conflicting data on google, I've only collected 1,000 so far and 7,000 total factions. I'm literally taking everything from the commodities schema and all station data from the journal, on a side note I do want to expand this into bodies within the star systems to see if there is any correlation there.

The slight bias towards the Empire in the median prices is possibly because Aisling Duval gives a 10% price boost on "high value goods" in her control systems, and a lot of them are Imperial. Though I suspect that doesn't explain all of it. (Moore's Charm is Independent, for the extreme on the LTD prices)
Interesting, I will look into that specifically to see if I can factor something in.

States are the crucial factor (and states and demand explain almost all the variation in price, as far as I can tell) - things to consider here:
- a station can be in multiple states at once and their effects stack. The combined effect of this is what's causing most of the really high outliers in your measure.
I was looking at the states individually even when there are multiple. Once I have a larger sample size I'll be able to factor in multiple states, which will definitely add another dimension.

  • you haven't looked at Civil Liberty, Public Holiday or Pirate Attack, states which cause extremely significant increases in price.
  • you also haven't looked at None, as a control.
Because I've only used a weeks worth of data, I was getting low sample sizes for these faction states, so they were deliberately excluded. I'll re-run this later and hopefully get something more meaningful. And I was ignoring None's, I'll put them back in next time I look at it.

The big thing where something is not quite right is you've said "not Investment" for Painite, despite Investment giving a ~2.5x increase in Painite prices compared to a neutral economy, and ~1.4x compared with Boom. (See for example: https://whdb.sotl.org.uk/stations/13/trade/63?minrange=3305-12-27&maxrange=3306-01-28 - hover points to see the applicable states)
This could just be a sample size issue, due to me excluding the other states and when I have enough data I assume this will change how it looks.

Thanks again for reading.
 
I dream about being able to buy a 'Best Buy' and 'Best Sell' computer that gives all prices for the system you are currently in. Sometimes I even dream I have played a game with this feature...


What was that... some letter and a number or something???? :)
 
Thank you so much for this drill down analysis. Awesome stuff. Really makes a difference to see high level trending from game data and not third party tools (which have been terrific to date). As a new player, I have found the market economy very bland within ED. Loads of potential but bland. Given the market is not linked to any crafting systems and no other commodities compete with mining as a viable source of income, it's no wonder the economy system is largely useless. IDK, perhaps I just don't know enough to leverage off other commodities. Definitely an aspect of the game Frontier can and should focus on at some point.
 
I'm not sure what the number of stations out there is, there is conflicting data on google, I've only collected 1,000 so far and 7,000 total factions. I'm literally taking everything from the commodities schema and all station data from the journal, on a side note I do want to expand this into bodies within the star systems to see if there is any correlation there.
EDDB reckons just short of 70,000, which seems about right, counting dockable planetary bases, and mostly - except where they've been manually catalogued for it - not counting non-dockable planetary settlements.

Because I've only used a weeks worth of data, I was getting low sample sizes for these faction states, so they were deliberately excluded. I'll re-run this later and hopefully get something more meaningful. And I was ignoring None's, I'll put them back in next time I look at it.
You'll probably always have low sample sizes for Public Holiday and Pirate Attack, just because of how those states work - short duration, long minimum time between recurrence, blocked by a lot of other states.

The way I analyse states is to look at the price at a particular market with no state and state A, to work out the effect of applying state A. Once I have that, I can then use both None/B and A/A+B to work out the effect of applying state B.

There were significant changes last Thursday so I haven't got enough data to redo the whole analysis in 3.6 yet but https://whdb.sotl.org.uk/effects gives what I found in 3.5
 
@Cmdr Green Skin Excellent work. Well done. Really impressed. Not sure I can say that enough. #GeekLove.

Some of the variation we see is weird, probably because we're not looking at the true determinants, rather we are observing other patterns. This occurs to me in the station type and the unusual behaviour of asteroid bases. I don't believe station types are evenly distributed in populated space so we might be seeing regional effects, reflected in these variables. It would be good to see these prices grouped by region: Bubble(Power)/Pleides/Colonia or some similar category like distance from Sol: 0-300 ly, 300-600ly, etc. Though possibly as these are markets we might want to take other variables from EDDN like traffic. Not to mention average demand. Plenty more suggestions in the thread above.

With more data what you would attempt to do is trace the relationship between supply/demand and other characteristics with the intention of scoring all sites for their value as sales locations. If you published that. Well... the formula for forum immortality. If the supply/demand mechanism in game is sensitive enough to player actions, and players act on those findings, it would change again, but hopefully less so.

Research always tends to pose more questions than answers.
 
OP, an impressive method of displaying the data visually. Not too much information at a time to become overwhelming, but enough to make me think I can understand it (y)

I'm sure between Ian Doncaster and yourself, we'll all gain an improved understanding of this part of the game mechanics. Probably some of the devs are interested too :geek:

But it makes me feel FD missed a constructive opportunity to communicate with the player base here. Why not provide us with a "tech reveal" of the main features of the mining economy, such as in one of Paige's livestreams ?
 
Back
Top Bottom