Monthly Economic Report - July 2022

Well, that’s one way to completely miss the point…

Gotta look at the whole picture or you come off as delusional.

2 Likes

Unless my whole point was only last year

Sadly my memory is longer than that, and decisions made prior to that continue to negatively impact the econemy. Production in particular.

2 Likes

Greetings fellow spreadsheet makers

My spreadsheet could beat up your spreadsheet! :person_raising_hand: :cold_face: :snowflake:

2 Likes

=VLOOKUP() Battle of the decade!

2 Likes

Can you plot the graph to indicate the source of sleeper components ? ( i.e. how much worth of sleeper components are coming from each class of j-space )
Just curious.

Most from C5 and C6 after marauder change for sure.

My guess: C5 > C6 > C3 > C4 > C2 > C1

There’s cause for extremely guarded optimism re: activity in the last 2 months. ISK velocity, production, and destruction are all up ever-so-slightly.

That said… an increase in production from 2.52T to 2.56T over the last 10 days of July doesn’t offset the drop from 2.68T over the first 21 days of July. And that’s just the rolling average. If we smooth out the data to get monthly average totals, we wind up with:

June Production / Destruction:
Produced/Day: 2.75T
Destroyed/Day: 923B

July Production / Destruction:
Produced/Day: 2.51T
Destroyed/Day: 886B

That’s a drop of 8.73% in production, and 4.01% in destruction from June to July.

If you look at the ‘full’ year (ie: Jan 01-end of MER), the numbers are… a bit nastier. Your bump there, after all, is from the April low of 2.52T. May surged, and then we slumped again. But just to take the bigger trend, here’s the change from Jan:

Jan Production / Destruction averages:
Produced/Day: 3.26T
Destroyed/Day: 1.06T

So from Jan-Jul, we’ve got a loss of 22.92% of production, and 16.34% of destruction.

So… no, looking back over this year, it’s not an upwards trend. It’s about 50 days of growth, all lost in 10 days, with the rolling averages showing some uptick after. And even then, look at the actual numbers, not the averages:

That fainter line behind the averages? That’s the actual numbers, and that covers the period of the slight uptick you’re looking at after the May/Jun bounce. It’s about 2 weeks, and the rolling average is pulled up hard by the period from Jul 23 through Jul 25. (Side note: if you download the data, the file 0_produced_vs_mining_vs_destroyed.html is the one you want. It lets you zoom in on any part of the graphs, and gives actual #s on the mouseover.)

To make it clearer, here are those production numbers from the .csv for those last 2 weeks:

Jul 18: 2.33073E+12
Jul 19: 2.68767E+12
Jul 20: 2.78507E+12
Jul 21: 2.36567E+12
Jul 22: 2.38207E+12
Jul 23: 2.83069E+12 **
Jul 24: 2.85237E+12 **
Jul 25: 3.08177E+12 **
Jul 26: 2.49574E+12
Jul 27: 2.64891E+12
Jul 28: 2.3825E+12
Jul 29: 2.50312E+12
Jul 30: 2.55364E+12
Jul 31: 1.84907E+12

** - these 3 days support the entire stretch after them. Without them, your ‘upwards trend’ vanishes. They’re a weekend, and I don’t have the time to do a deep dive into the numbers to see what got built where (and it is possible, if you know how to read the data), but it’s possible, given the 500b differences between those days and the days around them, that there were 15-18 titans in build in various places.

2 dozen supercapitals(at most) over a single weekend shouldn’t be enough to make people overly optimistic.

7 Likes

Ahem.

=iferror(transpose(split(textjoin(“,”,TRUE,ARRAYFORMULA(if(exact(‘Character Sheet’!AD56,“Orange”),if(isblank(‘Experience WorkSheet’!F52:F85),vlookup(‘Experience WorkSheet’!F52:F85,‘Experience WorkSheet’!$F$52:$G$85,1,0)),iferror(if(gt(vlookup(‘Experience WorkSheet’!F22:F39,‘Experience WorkSheet’!$F$22:$I$39,4,0),0),‘Experience WorkSheet’!F22:F39,))))),“,”)),)

:stuck_out_tongue:

Do you have a list of definitions for the data fields in these reports? For example, the first line in economy_indicies_details.csv is 11/1/2003 - Ammo with a price_change field, a total_value, and price_change_weighted.

What are the definitions of price_change, total_value, and price_change_weighted?

It may seem that the definitions are obvious, but they are not. When you are using the field price_change_weighted, is that value the value of the weighting for that particular iteration, or is it something else?

I am very interested to see how you are calculating some of these values because it’s not entirely clear.

How are entry_sink_value and entry_faucet_value calculated? And are these floors/ceilings?

1 Like

my-spreadsheet-can-beat-up-your-spreadsheet

Wow… 2.5 Billion m3 roughly in 1 region …in freaking HIghsec…highest number…???

:clap:

Between Rattati’s idiocy and CCP price hiking and lack of content…Job Well Done bois! you are doing GrrrEAT!

4 Likes

Don’t forget this is summer on the most of the mudball that is Old Earth.

Many a people prefer to take their time away from landlines. Let’s see this graph in January.

Adding to that, consider expiring subscriptions of people who is unable to pay for technical reasons.
This, too, is adding to the dwindling numbers in statistics.

1 Like

Hey Gozer,
I think this could be a useful addition to the MER data dump, a description of the files in the data dump. Will add this to the MER todo list!

But for the ones that you asked about, here is a short description of those:

price_change: The monthly price change of the sub index.
total_value: This months total trade value (volume*price) of the sub index.
price_change_weighted: The monthly price change of the sub index weighted by the trade value of indices within the main index.

entry_sink_value and entry_faucet_value: Are just the actual values, no rounding/flooring/ceiling. In the simplest terms, this is calculated by aggregating transactions between players and NPCs. Transactions from players to NPC is a sink, NPC to players is a faucet.

1 Like

Its very strange to see no mining or destruction in poch I would of thought the bigger groups in their would have at least done some.

But it looks like poch is the place to be for condensed fight’s hmm looks exciting.

Also outter passage is out trading Jita wtf lol thats awesome, or is that from fire sales and burning everywhere? xD

things are getting out of hand again .We need more blue loot to help the the player base. printing isk in nul sec is still out of hand.remove asp the oversheer effects from dead escalations remove the npc t1 loot drop to make mining a thing again .all isk from a ratter should go to the ess .we need more scarcity!!!

1 Like

CCP Estimate,

Thanks for the data definitions. However, I was hoping to get some insight into the data collection methodologies and the formulas you are using to come up with these results.

For example, is the data sampled at a rate of (for example) 1-minute and stored? It’s important to know these things because they affect the analysis of the data. Right now, the intervals are fairly well defined, but I assume (and this is bad) that the value published is an average of data, but I don’t know the sample resolution or the intervals used to create the average.

I would actually like to see the raw data at its smallest sample intervals so I can run a proper economic analysis of the data. I prefer to start with the raw data and then build models from there. How the data is structured significantly influences what you can do with it, and I don’t have a very good understanding of the data structures.

I assume (again, bad) that you’re pulling this data from a SQL database, hence the .csv file and field names at the top of each report (references to “dungeon” were interesting).

Unfortunately the data right now can’t give full insight into the market. I might do some covariance calculations and assess whether I can build ratios, but there’s little more that can be done.

But thank you for providing those definitions. I would like to get closer to the data if that is at all possible.

No, only on half of it. :wink:

2 Likes

While you are technically right, human population isn’t distributed evenly.