Sunday, December 14, 2014

Trade: Test run

Went ahead and did it. Ran the tool, went through its' recommendations, and picked a few that looked decent. Shipped ~5b of stuff from amarr to jita using a pushx rush contract.

The items were as follows: 70x silo, 10x prowler, 2x pithum a-type invulnerability field.

The a-type invulns sold exactly as expected, making me 156m.

Silos were selling for 18m in jita compared to 6 in amarr, presumably because of patch speculation. I sold 23 before prices crashed, sold many of them for a small profit, and also sold a dozen or so for sort of an intermediate number.

I sold 3 prowlers for a profit of 50m each, and then the price crashed to about buy price plus 10m. There are still 5 left on the market that I haven't spent enough time 0.01ing to sell, but I've decided to pretend I sold them at their present price and see how much I made.

The answer - after tax and so forth - seems to be that I made a profit of about 750m. This is not bad for the amount of work involved.

My takeaway is that in the case of the prowlers and silos, I bought too many -- silos were about 1/3 of jita's daily movement, while prowlers were about 1/2. In both cases they crashed to much lower profit margins before I could sell more than a fraction of my stock. In the future I should use more like 1/6th of the buying system's daily movement. The a-types which sold properly were 1/10th.

Also, when there's a profit to be made, expect to spend time 0.01ing, because everybody else is going to want in on the action too. Also, things that are dramatically out of line with historical values will probably tend to correct quickly. Unless there's an actual reason for it.

In terms of doing this more than once... yeah, probably. It would have to be my primary profession, if I was going to spend time 0.01ing. I might decide to take a month, do nothing but trade, and see how it goes.

Tuesday, December 9, 2014

Trade thing 5ish: Progress

I redid the equation I use to decide how many of an item are likely to be sold per day.

The first version was okay for high-volume items, but was bad at identifying items which sell less than one unit per day, which I don't really want to bother with. When I ran the trade finder many of the results were low-volume items that it thought I could trade and make a huge profit on, while actually doing so would have involved days of 0.01ing to sell a single unit.

The second version had the opposite problem -- it successfully ignored low-volume items, but thought it would be a super good idea for me to haul four trillion units of tritanium from jita to amarr.

So I put away the automatic curve fitter and did the variables by hand. Version three successfully ignored low-volume items, and thought jita moved about 15 billion units of tritanium per day. Huzzah.

Version four is version three with a flatter curve to make mid-high-range numbers more accurate. When I ran it for the first time the list of most profitable trades contained largely mid-range production stuff like ice products and PI or reaction materials, so I sighed and went to the game to get real numbers to compare with eve-central values... and it turned out the equations's numbers were correct. So I'm calling it good.

Also, I hadn't included taxes before, which resulted in the "profit" numbers it returned being inaccurate. If you move 4 billion isk worth of stuff for a before-tax profit of 150 million, the tax eats half of it. If you can move 1 billion for a profit of 149 million, that's a much better trade.

I guess this is the point where I do a trial run and see what happens.

Saturday, December 6, 2014

Indy contributor: November


Another passive income month, refreshing orders every 90 days so they don't expire.

It looks like goons moved their staging point 1 jump away closer into my teritory (market hub proximity?). I dunno, it looks like there are more active players in the area where I am, including docked and undocked scouts. I'm probably not going to get my profit from these custom offices from PI this month, but it doesn't matter. 10k m^3 of p2 products is the limit of storage on my planets, so I'll first reach this number before I really get into it.

Blueprints haven't changed so far, but I am thinking about moving my off-character on my main account due to the price of PLEX.

Sunday, November 30, 2014

Thinking about trade 4: It's happening?

So, I got the trade finder to run. Upon running it with only the first 150-ish items (out of about 3600), and only using amarr and jita as buy/sell systems, and only taking the top 10 results, it produced this:



As a sanity check I went to amarr and jita, found the real prices of the items and the quantities I would really buy, and got this:



As expected the quantities are pretty badly off, but the tool has successfully identified profitable trades and even put them in more or less the correct order. I'm calling this a win.

Thursday, November 27, 2014

Thinking about trade 3: In which I build a list of items the easy way

Turns out my shell service wasn't banned from eve-central after all. The issue was a bug which made it look exactly as though it was -- complete with realistic timeouts on connection attempts.

Anyhoo, I'm trying to build a list of item types which are worth trading. My eventual plan, if I can get it to work, is to move items between trade hubs, which is a high-volume, low-margin proposition, so I've decided somewhat arbitrarily that I can ignore any item which Jita moves less than a billion isk worth of per day. We know The Equation can be off by at least a factor of 9, so I'm having it drop items it thinks move less than 100m per day.

Which turns out to be quite a lot of them. My list of items with market groups, minus blueprints, comes out at around 9700 items. Watching the process run, I'm expecting the number it thinks are worth trading to be less than half of that.


Tuesday, November 25, 2014

Thinking about trade 2: A list of typeIDs which can be found on the market

This is a list of all typeIDs which can be found on the market, taken from the latest static dump. It was very hard to get, so I may as well share.

In other news, I've gotten a free shell service banned from using the eve-central API. Which is exactly why I wasn't using my home IP.

Sunday, November 23, 2014

Thinking about trade

I'm very lazy. I would never be able to compare hundreds of item types every day or maintain hundreds of buy and/or sell orders. But there's a lot of money in trade, so I keep thinking about it, trying to find an easy way to do it.

It would be easy to write a program that uses eve-central data to compare item prices in different market hubs. However, if you only look at prices, all the program's top results  would be very expensive items, like capital ships and deadspace modules. Eve-central doesn't track how many units are sold per day, so it's impossible to calculate profit potential.


And then I thought, "or is it"? Eve-central does track how many units of an item are on the market, and I'll bet there's a correlation between how many units are on the market and how many units are sold per day.

Of course, the market being what it is, there will never be a particularly strong correlation -- but it doesn't have to be strong. It just has to tell me that moving a bunch of nocxium at a profit of 10 isk per unit is more profitable than moving a single unit of whatever at a profit of one million per unit.

So I took a sample of item types and blah blah spreadsheets and quadratic regression blah and got this:


The sample size is way too small, but I can always go back and redo the equation with better numbers. The point is that it more or less works and I can tell how many units are being sold to within about an order of magnitude.

Next step? Sorcery.