trade routes

[Magic_gorter]

If my city has build a settler I will not start a new one (I will do this before 1000BC). If there is nothing else important to build (when playing a landing game I hardly need other units) I will start building a caravan or a freight. When the city finish it I will look if I can get a nice demanding city. If not I will transport it to a far away city to get the best gold/beaker trade (wonder building goes first)....in the early game it is low. It sometimes help avoid rioting and it gives a small boost to science which is important to me. This way I like to reach the important techs as early as possible.

So roughly said I only do this if I can't build anything else important. When playing conquest I won't do it of course.

Most of the time I try to do this: caravan->Marketplace/Library->caravan->etc.
Hope I explained everything clearly.....

 

[Peaster]

If my city has build a settler I will not start a new one.

Why not ? Are you saying that you don't enjoy managing a big ICS-style civ ? If so, I can just accept that.

Or, are you saying this is good strategy ? If so, I want to press this point. It implies that a small stream of arrows is worth more to you than 50 shields [and you did say that your shields seem fairly worthless]. That would contradict the usual values, that a shield is worth approx 2 arrows, so I'd like to explore that some more.

Hope I explained everything clearly.....

Yes. Thanks!

 

[Magic_gorter]

I was not explaining well about the settlers. I meant that if a city has a settler (and most of the time no extra food) I won't start the next settler. If that settler has founded a city then I will build a new settler again of course.

For me it's a good strategy (perhaps not the best). For me the 50 shield value isn't important that much. In some GOTM games I win with this strategy and sometimes I won't but it is a strategy I like much and works most of the times. Perhaps I should look better for the demanded vans.....but I try to play my games pretty quick (due to lack of RL time).

 

[Peaster]

I meant that if a city has a settler (and most of the time no extra food) I won't start the next settler. If that settler has founded a city then I will build a new settler again of course.

I see. I know you've been quite succcessful in comparison games, one reason I am interested in your strategy. If it disagrees with mine, it becomes even more interesting ! A chance to learn.

I make lots of settlers in my games, but most of those build new cities within 10-15 turns [eg the time it takes to build another settler]. So, there usually aren't many cities like the ones you describe, supporting settlers for a long time, unable to build more. Trying to remember what I make with those ... probably war units, boats, or vans. But any vans would be intended either for good bonuses or for WoWs.

I really can't remember a city with such useless shields [in the first halves of my games]. So, for now, I'll stick to the general rules, that 2g = 1s, and that a bonus under 65g is usually a mistake, even in a landing game. Since I don't play for landing very often, I could be wrong in that context, and if so, I hope someone can explain why.

If not, maybe we should play a comparison landing game some time, maybe in an upcoming GOTM ? These take more RL time than ECs, but it might be worthwhile, especially if we can attract a few other players who are willing to keep stats + notes.

 

[Magic_gorter]

Sounds like a good idea to me. The next GOTM I will create will be a "normal" one. Have not decided yet what kind of map....

So if there are more players who are willing to keep stats and notes that GOTM can also be a comparison game.

 

[Terrapin]

Some of Peaster's comments are inconsistent. If you start trading early in the game, you most assuredly will not get 200g apiece for them.

Also, part of the value of trade routes is that they allow your cities to remain stable w/r/t happiness with lower % of luxuries. That allows a younger civ to stabalize in rep/Dem while setting their tax rate for more research. I think that is what MG is saying.

 

[Magic_gorter]

That's indeed more or less what I tried to say....I try to get that science slider as high as possible and getting the cities happy....

 

[Prof. Garfield]

No ... The history of the van [rush built or not] doesn't matter! This is clearer if you play using a value system. In your extreme case, where each van is RB'd and only gets a 130g average return, the van system is barely sustainable. So, aim for an average return of at least 200g, so that you are making a clear profit, and trade is helping you achieve your other goals in the game. If a few vans only return 130g, I wouldn't consider them to be mistakes, but they aren't really helping with the sustainabliity issue. I'd consider a 40g van to be a pretty clear mistake [some players seem to be happy with them, but haven't explained clearly].

I agree that once you have built the van, its history does not matter; sunk costs are sunk, after all. When you are deciding if you should build the van or not, whether or not you rush buy it does make a difference. What matters is the opportunity cost of the van, or what you give up in order to get it.

If you rush-buy the van, the opportunity cost is about 120 gold (a bit more after losing the warrior and horsemen slots), assuming that the production in a city is not so important that you have to worry completing things every 2 turns.

If, however, you don't have a pressing need for the production of a city for some other purpose, the opportunity cost of a caravan decreases to whatever could be built with the city, some fraction of a military unit or building, or a settler plus a few more shields. The lower limit of a caravan's opportunity cost is 50 gold, because the shields could be sold off for that much by building and selling a building. If it is impractical for the city to build a settler, the opportunity cost must fall towards that lower limit, unless you have a need for military units or a particular structure. Under these circumstances, a 130 gold delivery is a decent return, because the cost of a caravan is lower.

In a city that is not very busy, there is a reason to rush buy a caravan (or other unit) anyway: you want the use of the unit sooner. In the case of the caravan, you want the delivery bonus sooner, and are willing to pay a premium to get it. In this case, however, (usually) only the gold matters; science can't be re-invested at interest, because science later is just as good as science sooner, provided that you get the technology in question before you have to use it (2 exceptions: if there is a pressing need/good advantage to getting a tech sooner [say explosives, or invention if someone else is building leo's], or you will later achieve 1 tech per turn, and have caravans to spare. In the latter case, however, it presumes the establishment of a sustainable and successful trade system.). Paying a premium upwards of 120g to achieve a 130 gold delivery doesn't seem like a good idea to me if the caravan will take more than 2 turns to get there, and then there is the cost of establishing and maintaining a shipchain for those kinds of deliveries, and the cost of establishing lots of new cities for the purpose. If the scheme is infinately repeatable (as with hides), it is a good move, because you pick up the extra science, and you always have a clear place to re-invest the gold. If not, you get no more science than otherwise, and you tie up your gold in order to achieve smaller returns.

There is another reason not to pay a premium to get your caravan sooner: the city grows while producing the caravan, so you can put your gold to other uses while the expected payout of the caravan grows.

Basically, I don't recommend rush-buying caravans just to deliver them faster, in order to get 130 gold deliveries. In order to pay that kind of premium on a caravan, I would expect larger returns, particularly if transportation will take a while. In order for transportation to be quick, you have to have a ship-chain, which is also expensive, both in initial cost and per-turn maintenance (particularly if you are rush-buying ships out of a small number of ports and not re-homing them). Establishing such a shipchain to send your caravans with such a small "markup" as 130 gold for 120 cost doesn't make much economic sense to me.

 

[Peaster]

Some of Peaster's comments are inconsistent. If you start trading early in the game, you most assuredly will not get 200g apiece for them.

Why is it so hard for people here to accept the truth ?
In the recent GOTM #105, I was averaging well over 200g per van by 1AD [I'd estimate 450g, but haven't checked that carefully]. See my log at

http://forums.civfanatics.com/showthread.php?t=347381

If you have doubts about the log, you can DL my 1AD save from the GOTM archives and play a few turns for yourself. If you still have doubts, you might also take a look at the GOTM Hall of Fame. Sorry if I am showing off a bit, but this thread is starting to frustrate me. Sometimes you just can't help people.

Also, part of the value of trade routes is that they allow your cities to remain stable w/r/t happiness with lower % of luxuries. That allows a younger civ to stabalize in rep/Dem while setting their tax rate for more research. I think that is what MG is saying.

I understood this, and an earlier poster said much the same thing. Did you understand my response, that there are probably better ways to play ? I admit that this is debatable, and hard to resolve by simple discussion, but maybe we [Magic Gorter, me, ...] can settle that in a comparison game. You are welcome to join us.

ProfG: I will read and answer your post ASAP.

 

[Peaster]

Prof: We may be nearing agreement. But I am not sure you understood all my posts, and not sure if you still object to my 130/65 guideline. Let's clarify which decisions we are talking about. I see two; 1) Whether to trade at all, and 2) Whether a given city should build a van for trade.

IMO decsion 1 should mainly factor in expenses [eg shipchains, WoWs...], expected average bonuses, and rate of van production. If you decide to trade, it's usually best to go all the way; pay the costs and then forget them. Decision 2 should mainly depend on the expected bonus of that city's van, not global startup costs.

You bring up the issue of RB-ing vs using shields, and we can talk about that, but I am not sure it is relevant to the 130/65 discussion.

I agree that once you have built the van, its history does not matter; sunk costs are sunk, after all. When you are deciding if you should build the van or not, whether or not you rush buy it does make a difference. What matters is the opportunity cost of the van, or what you give up in order to get it...

If vans had different costs in different cities, I'd agree that costs would be an issue. Or, if your cities were pretty useless for production [see my discussion with MG] I might agree. But at least in my games, most cities have decent production options [a settler, a WoW van, a boat...], so that 1 shield = 2g makes sense [eg costs are fairly uniform]. So, the build method doesn't really affect "Decision 2".

I would agree that in certain cities production can have greater importance than in others, and that it makes more sense to RB in those [but this is not really a cost issue]. For example, I'd be more likely to RB a 400g van than a 130g van. So, 1 shield = 2g is just a useful approximation, but I think it works here.

On the other hand, we should forget about selling buildings. That's just bad. Maybe I shouldn't have even brought it up. Or did you do that ?

... there is a reason to rush buy a caravan (or other unit) anyway: you want the use of the unit sooner ... however, (usually) only the gold matters; science can't be re-invested at interest ... Paying a premium upwards of 120g to achieve a 130 gold delivery doesn't seem like a good idea to me if the caravan will take more than 2 turns to get there...

I mostly agree with this, based on interest theory [I suppose you have seen my thread on this subject].

There is another reason not to pay a premium to get your caravan sooner: the city grows while producing the caravan, so you can put your gold to other uses while the expected payout of the caravan grows.

Yes ... in special situations. It's usually a good idea to RB any relatively valuable van, so that you can re-invest your profits ASAP. But this is a fairly subtle problem in interest theory. You have to compare the city growth rate [measured in arrows] vs your overall econ growth rate. Ideally, you should include the growth of the destination city as well. Since everyone here seems so keen on general rules, I'll suggest

* Ignore city growth [in deciding whether to RB] unless the growth rate exceeds 10% per turn. This is likely to happen if celebrating in a republic, for example.

But I can see quite a few exceptions to this one, myself, and I guess somebody can have fun finding them for us ...

... and then there is the cost of establishing and maintaining a shipchain for those kinds of deliveries, and the cost of establishing lots of new cities for the purpose ... Basically, I don't recommend rush-buying caravans just to deliver them faster, in order to get 130 gold deliveries. In order to pay that kind of premium on a caravan, I would expect larger returns, particularly if transportation will take a while.

I agree that RB-ing [or even building] a 130g van can be a mistake in certain unfortunate situations, but IMO these should be rare.

Shipchain expenses are "Decision 1" issues, rather than Decision 2. If you have decided to play for trade, and paid the costs, you should normally be able to RB some "good" vans [much better than 130g and/or quickly deliverable] even if you can't afford to RB every van.

In my games, the very first vans [maybe 500BC?] are often the slowest and/or least profitable. But I usually still value beakers in that phase, so I'm marginally OK with a fast 65g van, or a slow 120g van. Not saying I'd RB those - that would depend on gold reserves and on my alternatives.

 

[Peaster]

So if there are more players who are willing to keep stats and notes that GOTM can also be a comparison game.

I thought I might learn something or prove something from such a landing comparison game. But yesterday I looked for past GOTMs in which I played for landing and could only find #63 [std map] and #75 [odd island map], where I did quite well [Gold and Silver, plus a Green and Blue]. If I made any strategic errors in those, they weren't caught in the comparison. So, I am even more convinced that my advice applies to landing games. I don't think I'd learn much by doing another one, and I doubt I'd convert more people [and landing can get tedious, at least the way I do it].

People following this thread might find the spoiler thread to gotm 63 interesting for its discussion of landing strategies, but there's not much about trade strategy IIRC.

 

[bear42]

Is there any way to cancel a food trade route once its been established?

 

[Peaster]

@bear: I think it can be canceled like any other trade route, by creating a new route that the city considers more valuable [assuming your city has 3 routes already]. But I'm not a great "game-mechanics" expert, and I hope someone else will answer you. Or, you might look into the "Early Landing Guide" by solo - I think he discusses this.

BTW @all - to follow up on the previous discussion in this thread, in which I proudly claimed to know all about trade strategy [if not the mechanics] ... well, we have started playing GOTM 110, mainly to test our various opinions on that, and on EL. I think that so far most of mine have been confirmed ! Some games are still going on, so I can't say too much here [see the spoiler]. The less clear claims are:

* that trade vans are pretty safe from AI attacks. I did lose about 4 vans in my game, to AI sneak attacks, more than expected. But I delivered many many many vans successfully, so "pretty safe" still seems fair.

* that vans with small bonuses, just for the ongoing routes, are a bad idea. It still seems hard to measure the benefits exactly, so this is hard to prove or disprove [but for now, it seems to me that the best games do not include these routes].

* that even the early vans should pay over 130G or so. In this game, I had to settle for a mediocre trading partner for a long time, not very far from my homeland. IIRC my first three vans averaged only about 80 or 90 gold, but vans were averaging over 130G before 1ad and stayed at that level, or above, until the end. For some reason [relatively short distances to the AI on this map ??], I couldn't get many bonuses over 200g until much later ... maybe 500ad or so. This pattern hurt my plan pretty badly, but IMO the plan still worked.

 

[CharlieChuck]

@bear42
Although it doesn't cancel it, building an opposing food van and delivering it the other way will cause each cities to effectively cancel out. I know some players use that trick to boost their cities in size at the end of the game (because of the 50 food it also delivers). I think it's not allowed for GOTM's but am not completely sure of that.

 

[timtofly]

Is there any way to cancel a food trade route once its been established?

To get your trade route back, you need to have an open trade item in the city that is getting the +1 food bonus. Build a new van with this item. Send it any where you want, but right before you establish the trade route go into the city that is sending the + 1 food. Click on all of the workers changing them into Elvis. Establish your trade route and then go back into the "Elvis" city and put the citizens back to work.

Although it doesn't cancel it, building an opposing food van and delivering it the other way will cause each cities to effectively cancel out. I know some players use that trick to boost their cities in size at the end of the game (because of the 50 food it also delivers). I think it's not allowed for GOTM's but am not completely sure of that.

This will work if there is no trade items available; however you will loose two trade routes. You have to decide if loosing trade routes is more important than starving. Also it is allowed in the GOTM. The thing that is not allowed is the rehoming of Freight. You could not rehome a "goods" van from one city to use if there was no open goods to create a new route.

Something to try though in the future if you just want to deliver food and not an ongoing trade of food. If you change all of the workers to Elvis before establishing the route, it may not displace a trade route. I have never tried this, so I do not know if it is possible.

 

[Ali Ardavan]

timtofly is right on all accounts. Another way to replace the food route is to deliver a van from a high-arrow city to the city with a food route. As timtofly mentioned, right before delivery go to the city at the other end of the food route and reduce its arrow production as much as possible. A new route will be established replacing the food route.

 

[WildPony]

* that vans with small bonuses, just for the ongoing routes, are a bad idea. It still seems hard to measure the benefits exactly, so this is hard to prove or disprove [but for now, it seems to me that the best games do not include these routes].

I have always felt the same way. Better to build another settler than a van that will provide a weak bonus and to trade later once a trade bonus is more valuable than another city. Also, the small per turn route is quite unsubstantial to me and does not compare to the value of an additional city.

 

[Magic_gorter]

You guys forget two important things why accepting a low revenue could be ok. I've seen it with the last GOTM (#110) I played again. One delivery can be the difference between celebration and no celebration or sometimes needed for hapiness control (when having science at 100%). If there is no good (far away) city who demands the van or freight you could use it to deliver it to your SSC city. This kind of vans are not RB (mayby the last 10 shields) of course otherwise it will cost me more money then they will give me.

The other important thing is that this van or freight can unblock the supply list of a city (best is SSC of course). So when 1 or sometimes 2 low revenues makes me build a new van in the SSC city with a big revenue coming up soon.

And of course if it's possible to get better revenues and the possibility to unblock the SSC with it...that option is most of the times a better choise.

 

[Peaster]

Glad to see that Wildpony agrees with me against low-paying vans. And also to see Magic's reasoning for them. Let's see if we can sort this out.

You guys forget two important things why accepting a low revenue could be ok. I've seen it with the last GOTM (#110) I played again. One delivery can be the difference between celebration and no celebration or sometimes needed for hapiness control (when having science at 100%)....

You have brought in several ideas here, some old, some new.

1) The method of making the van is not very important. If you have some very useless cities, which can't make anything better than low-paying vans, then I guess making the van from shields is OK. But, as Wildpony mentioned, there are almost always alternatives, such as making settlers, or using the van for a WoW, or even for food.

2) Celebration. I am not convinced that a) celebration is as important as gold, and b) that ongoing routes are the best way to do it. Can you estimate these numerically, maybe based on your GOTM 110 ? Or, show us a game where ongoing routes from low paying vans made a clear difference ? Since it is your idea, I ask you to justify it - also because I find it hard to analyze myself (see below). After looking at several ELG games recently by the old masters of trade [solo, zenon...], I don't see many low-paying vans. This doesn't prove you are wrong, of course, but maybe I can use this to lay the burden of proof on you. Here are some of my thoughts, which you can use or abuse:

a) I assume you want to celebrate in repu/demo, to grow your cities a few more sizes. I've estimated (in the gotm 110 spoiler thread) that raising a city one size is normally worth about 40g, or about +2g per turn. I guess that a typical ongoing route pays about +3 arrows per turn (though some are much more) ? So, if the route makes your city grow at least 2 sizes higher than it would have otherwise, then the celebration is more important than the route's ongoing arrows. Does that seem about right ? Can you estimate how many sizes an ongoing route will add ? I won't try to make any grand conclusion yet.

b) Let's assume that celebration is important, and look for the best (cheapest) ways to do it. Ali claimed (110 spoiler, I think) that entertainers are cheaper than using the lux bar. I guess you are saying that ongoing routes are even cheaper ? My rough estimates:

* An Elvis removes a city worker, but probably not a great one. Let's say it was getting 1s + 1f + 2a. In repu/demo, food is not very valuable, but I'll say 1f=1g (and 1s=1a=2g) so the worker was worth about 5g per turn. The Elvis adds 2 lux, I think (or is it more complicated than that?). So, 2L for 5g.

* The lux bar converts taxes to directly to lux. So, 2 lux costs 2g. This assumes all your cities are alike - that you want celebrate all of them.

* Ongoing routes add into your city's total arrows, which get split into TLS. I'm guessing you raise Lux to 50% to celebrate, so half the ongoing arrows become lux for celebration, and this matters for about 2 turns. I'm assuming your low-paying van costs 125g minus bonuses/etc. Wild guess is it that costs you 40g, compared to alterntive investments. So, you get about 3 lux total, for about 40g.

* Temporary conclusion: the cheapest way is the lux bar, 2g/2L = 1. Next is an Elvis, 5g/2L = 2.5. Next is an ongoing route, 40g/3L = 13. I do NOT feel confident that I got all this right, but unless I made a pretty big mistake, ongoing routes are not very effective for growth celebrations.

3) Preventing unhappiness while science = 100%. Do they affect happiness at all, when Lux =0% ?

The other important thing is that this van or freight can unblock the supply list of a city (best is SSC of course). So when 1 or sometimes 2 low revenues makes me build a new van in the SSC city with a big revenue coming up soon.

OK. I didn't have this special-purpose van in mind, since it is not made mainly for its ongoing routes (is it?).

 

[Ali Ardavan]

Glad to see that Wildpony agrees with me against low-paying vans.

I am getting lost in the details of this discussion here. Obviously low paying vans are low priority. But they are not worthless.

Here is my typical way of using them: My strategy is to win by spaceship but not necessarily Earliest Landing date possible (that is I am willing to delay landing a bit for additional growth) . I am already in Democracy and my old cities are maxed out or near maxed out. All my main cities have Marketplaces, some have banks. My lux rate is enough to keep my cities at the border of celebration (with some additional help at times from entertainers) but not more since I am concentrating on science (For me this is typically 20% to 30%). I build a new city in a newly discovered part of the world. I would like this city to grow quickly without increasing my overall lux rate which would be very costly. As soon as the city is built I will concentrate on food to help it get to size 3. I do this by irrigation, building a harbor, and when I am super rich by rushing a supermarket. My next priority is trade, I want the city to have trade routes as soon as possible so when it reaches size 3 it can grow incessantly. This means delivering to/from a city of size 2-4 which would result in a low payout. However, the trade routes assure the growth of the city and their value grows with the growth of the city.