October 2005:
Following changes to the rules in 2005, Contraband no longer varies in price in the way Gabor describes below. It's proportional to population, so only varies very slowly and can't be manipulated like this. Unless your race has contraband dustoff pods, it's no longer even worth sending population off to gather it from high-contra worlds. But I've left the page below, because I hate to throw away data
- Paul H



September 2004: Gabor Toeroe has been analysing how contraband works, and sent me the following notes on how to make money from contraband:


Manipulating seems not to be possible anymore at least it is difficult. But you can anticipate the price and you know if a type is cheap or expensive.

It seems that correct trading at the beginning of a game can give you a boost so big that if your opponents don?t do that they have a poor stand against you.

CWF (type) = SUM [player=1 to N; RCWF (type, race (player))*colonists (player)]

CWF= SUM [type=1 to 12; CWF (type)]
CMP=SUM [player=1 to N; IFF [race (player)<>106; colonists (player);0]]
Start price (type) = 2 + (1+ CMP/10^9)* 25 * CWF (type) / CWF
Asymptotic price (type) = 2 + (1+ CMP/10^9)* 100 * CWF (type) / CWF
Mean price (type) = Start price (type) * (1- a)^(trn-1) + Asymptotic price (type) * (1-(1- a)^(trn-1))
Actual price (type)= Mean price (type) * (1+Random)
Random ranges in +-80%. Most of the times it is small, sometimes big and it drives deviations from the mean most of the time the next turn back to the mean but this could also be an effect of missing parts here. Also deviations under the mean are more likely then above the mean.
RCWF (type, race (player)) is the race contraband want factor of the race stats.
N is the player count of the game.
a is an experimental parameter and has to be fitted to the results. Currently I?m working with values like 1/20 or 1/25.

What is missing are two things: The happiness dependency and a dependency on the price of the other types. All formulas are derived from observations during game play and simulations. So I give no guarantee for correctness. I use it to find out the best types for investing in the beginning of a game because later in the game the formulas become false for some values. But this was before I added the (1+ CMP/10^9) part. Maybe now it fits. We will see.

A simple way to make cash with this information
If money is the bottleneck you will have to buy the type with the most expected relative growth because there you will get the biggest return for the invested money.

If your population is the bottleneck (-> sell limit!) then it becomes interesting. First one have to decide if one could afford to bind money to contraband over the sell limit or if one wants to have full ?cash flow? ability.

If one can afford binding mc then one can buy the cheapest type (cheapest means the type has the smallest percentage part of its predicted mean price for the nex turn)

If one want to have the full cash flow ability then on have to check if there are types, which have a high absolute price (most of the times that will be sorts like Kerria or GBW) and are expected to grow next turn.

Then one distribute the money for buying CB so among the best relative grower and the high price type that the expected sell limit = actual colonists at main base /1000 * actual growth factor is reached.

A detailed algorithm will follow. Because the last described method holds not always. And there seem to be far more investing strategies.

GFM GToeroe





Here is a related, detailed description of the gathering mechanics which Gabor posted to the mailing list - a "real life" example:

?&The first ever 16 turns played as Rebel have passed by and I decide tell something because you can?t find it in the guides.

1)      Get pod pads before Falcons.
2)      Get hyper tech 3 for your Bullrout support fleet.
3)      Ensure gathering of all contrabands within the range of jumper before the build up of 10000 cantinas starts at higher population levels.
4)      Setup worlds with 100k-400k for gathering CB. The more expensive and higher the amount of CB the more colonists.
5)      Build Bullrouts from all available and suited hyper drives to collect the surface, which is farer away stuff with docked pods. The nearby surface stuff you can send with boosted pods.
6)      Setup on all bases in your neighbourhood via pod boosting little stuff gathering bases. Choose 4 in north south west an east to setup a 100k world to supply a city and 5 cantinas to spread out your colonists again from these points and again and again. I didn't do that at 100% but I think it's worthful.
7)      Don't build mines at the start. Don't build farms unless required due to host config. Bad investments. You will collect masses of surface stuff later.
8)      Good investments at the start: Cities, 5 cantinas on each world with more than 100k colonists, contraband gather stations, contraband trading, setup and defend bases for gathering Bovinoids, (I made the experience the only bovinoid will enter bases at reasonable rates before you do the job via Falcons), (7/1/2) wings to defend important native planets and attack with ~40 of them the first base. If you made it right a Sage can take out the base shield and ~10%-20% of the AAguns. So buy at the start 230 type 1 fighter if you can-> 280/40/80=40 (7/1/2) wings. This is a tactical force valuable for many things. ~3 of the can take out Migs, 5-10 of them can take out the MBR, and so on. And if you manage to destroy the base shield you destroy with this 40 wings almost all important bases structure of an opponent?s ground base. (Note: Your wings of course will die in this attack).
9)      Bring back the excess in colonists or pod them to new gather worlds.

Don?t spend more than 400k for gathering CBs. The reason is the gathering works like this: First some abbreviations:
CMod=IFF[Col<400k; Col/400k;1];
Q = ((PSI + 30) / 400) + .2 ;
P=1-Q ;
CBMin=Col/100k ;
gap0=(CB/25+5)*Col/400k ;
gap1=MAX[gap0 ; MIN(CB;200)*CMod] ;
CBMax=CMod*IFF[CB<2000/7 ; CB ; 3/10 * CB +200]

CB is the amount of contrabands of one type on the planets surface. The chance now for gathering a amount lying between CBMin and gap0 is P and the chance for gathering a value between gap1 and CBMax is Q. The mean gathered value can be calculated with (CBMin+gap0)/2 * P +
(gap1+CBMax)/2 * Q
This what I extracted of sims of thousands of bases. It?s only guess and only for Rebels. But it fits quite well.

An economic investment can be sorted by its percentage return and the time it took to get a full return of the initial costs. So in the beginning I never would spend money into mines and farms if the food stockpile doesn?t cut growth or surface food will ensure the needed food for max growth.

If you have Bovinoids then the costs of cities is 150mc and the return is 100mc or 66%. After two turns you exceed the initials costs. The second effect of Bovinoids is that you don?t have to bring the factory count to the usual level. Say if one needs a level of 500 then only 250 are needed and
750 additional mc are free for other purposes, i.e. CB trading.

Five cantinas on a 100k+ world produce in the mean 21kT contrabands. At the start we had a mean price of 5mc. I think this is also in most other games at the start. So five cantinas will cost 52.5mc with Bovinoids and have a mean return of 105mc or 200% and this will increase turn by turn.

If you build more then 5 cantinas then the productions changes and you have a mean output of 1.2*cantinas*Col/10M*mean price If you start with 1,000,000 colonists then we have as output for one cantina
1.2*1*0.1*5=0.6mc at costs of 10.5. You see that would be no good investment. There for you have to wait for better prices and more colonists. Build 100k-400k worlds to collect the contrabands. Build one to 4 cities there (due to the time it took to collect the CB. The cities must bring a return) maybe a government centre and 5 cantinas. Theoretically the most efficient way would be to spread the whole colonists in 100k worlds which are a little space from each other to fill the gaps of unsettled planets with the nearby colonist exceed. But unfortunately you have to place 100 foods on each world to hold the population happy&

Later in the game when cantinas become interesting you have to decide carefully if you send more than 100k to a new gather mission. For this you need the above mean gather formula and the actual prices to decide if the colonists produce more money in the HW cantinas or via the gathering.

When Falcons are available go first for Ghipsoldals. In most cases you form an alliance with a race which can use this engines for your advantage or you get alien hull plans where you need good engines and also your TCCs will take advantage from Tylium Thrusters! The earlier the better spread them in
portions of 50k around your HW to produce much, much engines. That?s better than the return of amorphous. After this get the Amorphous. 100k will produce 200 spice what gives at the start with a price of ~3 an additional income 600mc. Then get one pod of Siliconoid. With Bovinoid ordnance will
cost you ¼ of the usual price. I have not Amphibians collected so far, but from what I heard now one can forget the LW production. Good natives are Bugs (x3 tax) Avian and additional Bovinoids (x2 Tax).  If you are ready with most of them dust off Reptilians bring them home and distribute them in
suited portions to your mining worlds.
Of course if there are Borg for example take care of the humanoids. If there are Bots take the non-Insectoids first. I?m ready.