Trade Strategies

Are trade strategies currently disabled? Across all my countries/enterprises the only multiplier to product price sold on the market is the quality of the product. I have been playing a while but never noticed until recently.

Hi Shibz,

I just took a quick look at your country Shibzerland Prime on GR. Of the 4 state corporations I checked, all had near 100% of their production reserved for contracts - which are defaulted to market value and ignore trade strats entirely.

I didn't want to trawl through all of your corporations. Are you experiencing this issue in any that don't have most of their production contracted? If so, please reference a name or two and i'll take a closer look.

All the best,
Dub.

Dub,

Thanks for taking a look. Yes, I am having the issue across all companies. In Shibzerland Prime the following companies have 0% contracted and are exhibiting this behavior: Madera Cattle Feed, San Rafael Adv. Effectivity, Puri Adv. Quality, Cadiz Airforce Maint., Jati Off Weapons Maint.

Since my companies also have over 100% production, everything over the contracted amount goes to the world market at market price multiplied by quality factor without the trade strategy factor.

This is happening across my three countries and enterprise on GR and my enterprise in FB.

I should also add that I have played with extreme strategies and slight changes to 100% market price. Neither had any affect.

This has happened for some time. I did a lot of experimentation in early 2013 and concluded that I was not seeing a difference in sales when setting a very high price but it did appear possible to receive less with a lower price. Therefore, I have, since that time, always set my sell strategy to "best price".

There is plenty left in country and corp management to keep me busy so I am content with the way it works.

Aries, thank you, I just wanted to make sure I wasn't missing something.

I believe it was explained around the time Aries refers to that the game will try to match bids for product to offers the best it can, so if you are asking high prices you have little chance of getting them but will receive a more reasonable price. Something like that.

I've been doing a simple experiment to look into this. I tracked 4 oil corps with similar welfare, quality, and tax, but which used different trade strategies.

High 1: Average Product Quality: 274. Strat: Sell at a high % above quality.
High 2: Average Product Quality: 265. Strat: Sell at a high % above quality.
Low 1: Average Product Quality: 267. Strat: Sell at best price/product quality.
Low 2: Average Product Quality: 272. Strat: Sell at far below best price/product quality.

All corps had welfare 109.98, tax 30%, production/quality process 225, 100% employment etc.

The graph:
http://imgur.com/HNRki2U

I'm not sure in terms of analysis now. What do you guys think?

I think I can't see your graph.

edit: wait. got it. could you list these corps though?

I would be interested to see how it compares to a similar "best price" example of mine. Perhaps Fullerton Oil in CC2 on LU would be a good example. A full year sales/expenses would be interesting.

I think what you have to add is how much higher than quality adjusted price corp 1 is. If the product is in demand then you are going to get above actual produced quality prices - up to a limit - so corp 1 may be asking 100 above Q but only getting 30 above compared to 3 and 4 which are happy to take whatever deal is asked for on the market (the best deal around or a deal at below Q)

Hence high 1 showing highest profit, (high 2 seems a red herring, the same as high 1 by your labelling, think you have made a boo boo somewhere), but low 1 is getting a reasonable profit by asking for the best deal it can get, and low 2 is happy to take anything, hence lowest profit.

In a red market this seems quite right to me. In a green market though, I think you would see low 1 probably outperforming high 1 as it is far more likely to sell each month than waiting for a high price being offered.

But I defer to your two greater study of how it works.

Responses! :o I just got in from work and won't be able to look at this and comment meaningfully for a while...

I almost certainly did make a booboo ToeCutter, though i'm not sure to which one you are referring :P The mistake I see right away is that, in my lunchtime hurry, I forgot to label which corp corresponds to which line....

High 1 is the highest red line. High 2 the second highest.
Low 1 is the highest black line. Low 2 is the lowest black line.

Darnit, you made me go find them. Next time just say that you named the corps that and they are in your Pax Romana enterprise on LU.

I looked. None of them are selling at best price. Instead, they currently have the following sell strats:


High 1: Start at 320% of the market price and lower by 15% every month that the product remains unsold.
The offered price is updated when the produced quality changes.

High 2: Start at 310% of the market price and lower by 15% every month that the product remains unsold.
The offered price is updated when the produced quality changes.

Low 1: Start at 171% of the market price and lower by 15% every month that the product remains unsold.
The offered price is updated when the produced quality changes.

Low 2: Start at 274% of the market price and lower by 15% every month that the product remains unsold.
The offered price is updated when the produced quality changes.

You have indeed identified a mistake (actually two)... Low 2 is trailing it's quality level (274) by 0% rather than being set to explicitly 'best price/product quality' as I stated in the original post. This is because I intended to discover the difference between a 0% hike over quality and the higher hike at the 'High' corporations - ie investigating trade strategies. Also in my lunch time hurry there was a naming issue and I listed the order of corps within the graph incorrectly.

I'll redo the graph later and name it to avoid confusion. This is what happens when i'm Simcountry-ing in the 10 minutes before a meeting. :P

Now I'm confused. Dubs original post said low 2 was well below market, now it seems its actually at market with 0% over Q produced, or Aries has low 1 and 2 round the wrong way. And Low 2 is the much below Q strat.

Yeah, understand your probs Dub, lets see a comparison with good labelling and I still think my ideas will show to be along the right lines. I sort of remember going over this with Border C and Q (who remembers them ?) a while back. Q was one of the best econ players that I ever ran across.

Okay, let's try this again. As the above numbers are bungled and mislabeled I did everything from scratch. The results are perhaps somewhat surprising, and I am thus posting them out of the kindness of my heart.

Same setup, 4 oil corps.

High 1: Average Product Quality: 274. Strat: Sell at a high % above quality.
High 2: Average Product Quality: 265. Strat: Sell at a high % above quality.
Low 1: Average Product Quality: 267. Strat: Sell at a high% below quality.
Low 2: Average Product Quality: 272. Strat: Sell at 0% above quality.

All corps had welfare 109.98, tax 30%, production/quality process 225, 100% employment etc.

The results:

Graph 1: http://imgur.com/5vqxvT7%2CEgLxiGf

Now onto more interesting stuff. High 2 and Low 2 have very similar levels of profit, despite the first selling at 40% over quality and the latter selling at 0% over market quality. However, there is a difference in their average quality level of 7.

To adjust for this I calculated the difference in average profits between High 1 and High 2, as they were both using the same trade strategy but have different quality (9 quality points). The mean profit of High 1 was 188(mil SC$) and that of High 2 was 150(mil SC$). This results in a difference of 4.2(mil SC$) per 1 quality point.*

As mentioned before, High 2 and Low 2 have respective quality levels of 265 and 272, a difference of 7. To bring the comparison between them in line - isolating the difference in trade strategies and cutting out the difference in quality - it was thus necessary to add 29.4(mil SC$) (4.2 x 7) to the results of High 2. See the following graph:

Graph 2: http://i.imgur.com/EgLxiGf

There is thus now a significant difference in the profit levels of High 2 and Low 2, attributable mainly to trade strategies. *I hear you shouting at the computer screen that there are variables such as supply price, number of products sold etc also in play here, and you would be correct. Therefore:

To demonstrate the validity of the quality adjustment procedure above I adjusted the quality of High 2 (265) to match that of High 1(274). If the plotted lines of their profits match closely after said adjustment, then it may be said the 4.2(milSC$) adjustment above holds true in the case of this experiment (as both corps would in that comparison have the same trade strategies and quality). It would also therefore follow that were High 2 and Low 2 to have the same quality, as modeled in Graph 2, High 2 would in fact be making more profit - as observed in Graph 2. The moment of truth:

Gaph 3: http://imgur.com/K7YNTHz

As seems fairly clear, both High 1 and High 2 have very similar profit levels when High 2's results are compensated for the difference in quality using my adjustment calculation. The same calculation that predicted if High 2 (on high strats) and Low 2 (on low strats) had the same quality, the former would make more profit.

The only conclusion I can draw is thus that trade strategies do, to some extent, positively affect profit. At least in the small case study performed on these 4 corps. I wouldn't be changing your strategies just yet, at least until someone with time performs a much longer and bigger (thus more reliable) study.

(in b4 Aries looking at the corps, they are now using diff strats)

What we are dealing with when you distill it down, is coded mathematics. I reckon there is a built in 'randomness' in the way most of this works. There are certain predictable, set, equations that have been either given by the GM over time or worked out by players for some of the game mechanics but I have never seen anything that is even semi-solid regarding the markets. If you had the code... you could reverse engineer it, but we don't. And of course, there is no true randomness in possible in current computing, so it would be aa approximate random element.

Dunno, something to consider, unless you really feel like doing a large study on it. Thesis idea for an under-grad ?