Market Trading Vs Selling: An Analysis

People also aren't mentioning that gold prices early on in the game are actually very good, like you can sell 100 food for 60 gold or something, it's not at 100 food for 14 gold for the entire game.

Also, if you go very late into the game, people who are trading sometimes buy food with all of the gold they get from their trade carts, which improves the sell prices for the seller too. However by this point the trade will have paid off. It's just another thing to take into account.

 

I appreciate your comment. I'm 1715 on HD and so I would at least defend and say I am not a noob. But even a noob can show his math.

In competitive AoE2, 1715 HD is a very low level. I'm only 1618 on Voobly and that is barely an intermediate level, quite "noob" when it comes to competitive play.

 

[HyunaOP]

I knew Saracen Market was better than Spanish trade

Thanks Man xdxdxdxdxdxdxdxd

 

[JoMamma]

@RicoJay13

I watched this and your previous vid. Liked for effort. I'll try to give some fair feedback.

The vid itself came across as you trying to get back at people on reddit. I don't blame you because the aoe2 subreddit is not exactly a shining wealth of knowledge... Even still, if you're doing an analysis, focus on the analysis, not what people told you on reddit. If you would have posed this as "there seems to be a misconception on the time for a positive ROI from trade", it would have made more sense.

Some of your observations were worth considering, such as how tradecarts become very inefficient from convoluted routes on maps such as BF, and how long (perhaps unexpectedly so) it takes for them to give a positive ROI. Even still, these conclusions aren't exactly ground-breaking for anyone out of NPL... Unfortunately you only looked at BF, which isn't a map exactly known for its expert play (sry mbl). You did end the video by basically saying the break even time is longer than expected, but I think the way you were comparing the extra vils on wood versus trade mislead some people to believe you were a firm advocate of never making trade. Perhaps if you would have posed the test as more of a what-if scenario, this would have been more clear.

I appreciated the first couple Star Wars jokes, but I think there were too many.

Semi-related: if you would have titled your vid on Malay elephants something like "Examining the cost effectiveness of Malay battle elephants", the discussion would be a lot more valid.

I like the idea of examining the game from a mathematical standpoint, but you if you are going to try and extend analysis to expert play, you should do a lot more research.

Overall, there's potential, but you weren't quite there with this one.

 

[RicoJay13]

One thing that nobody else really addressed so far is that either you need to invest into trade later anyway, because selling wood for gold just takes too much population, so you need to invest the resources into trade anyway, so why not do at when you still have gold left that you're mining? So unless you can finish off a game with the investments saved by not building trade, it's worth it to make trade anyway, or you will lose a lot of resources later in the game.

What you say sounds perfectly logical, it's just that I do it often that I use the same population that I would have devoted to trade carts to simply chop wood. And it works great inside of that very long window before trading pays off.

 

[Cicero]

The math doesn't lie.

Maths doesn't lie - but it often misleads. A few observations:

- You say that it takes about 14 minutes for the trader to just break even on his investment. But actually, the trader has a massive advantage on the seller at this point, since he now has continuous gold income, and is in a position to constantly spam paladin, or the gold of his choice. The seller has probably mined out of gold by this point, and selling resources generates no where near enough gold to sustain gold based armies.* They are either forced to make trash - which has essentially no offensive power in team games, or must sell resources to buy gold to make tradecarts to get gold... which is a far greater investment than starting trade earlier.

- Another much simpler way of measuring the effectiveness of tradecarts is to simply calculate the amount of time it takes for one tradecart to pay off its cost. As noted in your video, it more than pays off after just one trip. There is no need for a player to produce 32 tradecarts all in the space of 6 minutes so this isn't really a fair comparison.

- You talk about the cost of setting up trade as being a massive investment, but it really isn't. At the point when players start trading, they are almost fully boomed, and they have already established their farming economies as well as many of the buildings they need. With their now huge economies, setting aside few thousand wood (plus gold) over a period of 15 minutes (probably more realistic than making 30+ carts in 6 minutes) takes relatively little out of their military production capabilities, but it has massive implications for the game after those 15 minutes. You say that those resources could be spent on 40 mangudai, but this is 40 extra mangudai over a period of 15 minutes. This is rarely going to decisively win a game before that time is up; after which there is a massive advantage to the team that set up trade.

- Players will be close to pop capped when they start trade. Its about transitioning your eco-balance to a point where you can maximise your military population. Selling food/wood for gold is highly inefficient, and will end up taking a huge number of villagers to support only a tiny gold-based army.

I'm 1715 on HD. I win team games without trade often.

I've got some prior experience of 17xx team games on HD... I've literally had people start trading with my market in Feudal. You need to test against proficient players before making any conclusions.

*Admittedly I've put this to the test this several times myself (i.e. I've been too lazy/overconfident to add trade). Few here will be surprised when I say that I've pretty much screwed myself over if the game doesn't end before I've mined out my gold. Selling food/wood at the market generates nowhere near enough gold to support a gold-based army. You can use maths to show the difference (I think tradecarts generate at least about 7 times as much gold as the same population of villagers on food/wood) but you really don't need to - experience here is enough to know this simply doesn't work.

 

[Athasos]

Man were the first answears harsh.
I would personally never "criticise" this way but you for sure made it very easy to "hate" on your video.

BF isn't seen as a competetive map, it also is normally a very much walled and stalemate map where siege and castles/bbt make it a super slow map.
Therefore a lot of units die on this map and trade is absolutely needed, as it basically is "free" gold income after the first trip.
This math you use here does not work for any kind of "better than 1750 HD" gameplay.
At first you don't only play BF, and as long as you don't play BF or Arena, wood is sooner or later a resource you'll fight for, therefore selling wood for a long time is a bad approach.
Most of the time trade is added with "extra" resources, which you don't spend, as your fully boomed eco gives you more than enough resources.
Traderoutes are cleared, so bumping isn't that much of a problem anymore (even on BF).
I have seen so many games where the team starting trade earlier won because of it.
I can't really recall any game where the "investment" into trade caused a loss because the other team started trading later or made less tradecarts.

Just another thing you forgot:
Trade needs 0 maintenance once started and with a cleared route, while microing vills for wood and selling wood takes some time, time where the extra 40 army might die to a SO.

Let me give you a tip here, it might be right calculations you do here, but you miss a lot of variables.
This game is not only about cost efficency, it's more about pop efficiency, especially in TG.
I had a guy telling me halbs and skrims kill all gold units, as halbs counter paladins and skrims most ranged units, on paper that might be true, but 50 paladins kill 50 halbs and you know how they do it?
They run away and kill eco, while the halbs either chace or die to castles or TC's attempting to raid, same goes for skrims, especially against mounted ranged units.
It is an incredibly deep game and you need much more than some "calculations" to understand how to make the best of your resources and population.

You said it yourself, you are 1750HD and while you might have played a lot of games, there is people here who played this game for almost 20 years, and all those games played during this time showed one thing:
You lose without trade.

To end this very long relpy I would like to give you some advice:
Don't force yourself in making new meta for very general situations.
Take a look at the small things, like how good is saracens market, to avoid making trade to early, or how can I abuse malay uptime or khmer bonus for my advantage.

 

[ChriSt]

If you dislike his video you could be a bit more diplomatic ?

Like saying this is the worst I’ve ever seen when someone is putting effort in something is not really motivating or helps him to improve..

Hum no. He is not entitled to diplomacy and as his replies indicate, he is after all a 1700x HD expert, thus he has no need to inprove or make trade for that matter.

 

Hum no. He is not entitled to diplomacy and as his replies indicate, he is after all a 1700x HD expert, thus he has no need to inprove or make trade for that matter.

The prime example of AoE2 elitism

 

[Clemensor]

Alright, so I pondered a bit about the whole idea and concept behind this and will now try to give some points that may havent been adressed yet.

1.) In your video you clarify, that your experiment is limited to maps with large amounts of wood (or as you said:"wood that lasts for an hour long game"), such as BF in the video. If we go with BF as the map of choice, being able to push fast is certainly not the norm, by the time you break through layers of walls and onagers behind them, the other team might already have a nice trade route going and will eventually push you back. On more open maps the distance you have to push to make an impact is quite significant and can take quite a long time if you arent already ahead.

2.) I think you are positing a false dilemma. At the stage, when trade is usually made, players in most instances will already have their desired vill number of ~120-130 villagers. So over time they naturally will accumulate a surplus of wood and food, thus you will see players supporting both army and trade production by selling resources and deleting villagers as they go. So its not as black and white as you make it seem in your video, you dont necessarily trade off carts vs army, you are aiming for enough army to not lose your position, while building up trade for the next stage of the game. Nobody will tell you to build trade before any army. This point will bring me to my last point.

3.) I think the best analogy gameplay wise on a much smaller scale would be delaying eco upgrades for more army. For example a very aggressive fwd with skrims, vills and spears, where you get as much army out as possible to overwhelm your opponent while crippeling yourself in the long run. This can undoubtedly work and players like hoang are artists in neglecting eco upgrades for all-ins. These are very niche instances and I imagine they are much harder to pull off on a larger scale. Also All- ins are risky, instead of getting steadily ahead you are trying to finish the game quickly in a kill or die situation, and although all-ns certainly have their place in the game they most of the time arent the recommended approach to learning and understanding the game.

I would also suggest trying to analyse some TGs by higher level players to strenghten or weaken (depends on the outcome of the analysis) your argument and see how they approach the game. You can analyse their vill numbers compared to military and trade carts and also army production to trade production in combination with selling resources. These case studies could make your thesis come to life and could give the whole argument more depth.

In conclusion I think your suggested conclusions are too broad for the actual point you are making and most of you argumentation comes from a point of ignorance about how TGs are played at different lvls. I really tried to be as constructive as possible with my criticism here and I hope you will consider some of my points. @RicoJay13

 

[ZeroEmpires]

Watch aM vs Secret on Team Acropolis. In the end game Viper is down to 24 villagers, 75 trade carts and the wood buy price is 650+ gold per 100. This map doesn’t have a lot of Wood like bf, but it’s more representitive of the maps played competitively (unlike bf).
Wood is such an important resource in the whole game, selling it off so that you can avoid making trade for longer is just going to make you weaker when you inevitably have to make trade anyway. Unless you have a crystal ball and know that you can end the game without trade, then you need to prepare for it and set up a steady income of gold so that you don’t gimp yourself when you’re out of gold and can’t sustain military unit production and trade cart production later on.

And this is the problem of applying raw math to a game that has so many variables which you can't account for, and which in order to create a realistic simulation you need to have good game knowledge. Your numbers might be right, but if you're basing those numbers off of assumptions or understandings that are incorrect then your outcome will inevitably be misleading.

Why do we only take 1 sheep at a time under the TC when there's a negligible loss of food for being lazy and taking multiple? Looking at it from a math perspective only tells you it's okay to be lazy, but try landing a boar under the TC when you have 4 rotting sheep carcasses there and suddenly you can see why in the context of the game situation you'll end up tripping over your math.

I believe the secret to success in AoE2 is logic, not math.

 

[Memeluke]

wood buy price is 650+ gold per 100

Man then his strategy is even better! It was good when he got 14 gold then at 650+ must be amazing

 

Man then his strategy is even better! It was good when he got 14 gold then at 650+ must be amazing

11111

 

[_Mr_St4rk_]

Man that's awfull, i can almost feel ofended by how misleading those Infos are, u should do a video showing how essential is to place a market and the differente between a clear trade route and a bad one.. and how to fix that 11 instead of doing this market thing that makes no sense 11

 

[archxeon]

I believe the secret to success in AoE2 is logic, not math.

Way to favor Secret over AfterMath. jk :P

 

[_Melkor]

I think even if it was proven to be more resource efficient to sell resources rather than trade for it, I would still set up trade. Because when the fighting begins, the minimum amount of micro you need to spend on making sure you are always building trade carts the better. Having 30 early and forgetting about them is so much better than getting a huge stockpile and then having to micro army and trade for next 5 minutes.

 

[Henkdesupernerd]

I would like to compare your investigations of investing into trade (long-term investment) to adding multiple town centers (long term investment)


Your scenario:
A decent player has boomed to ~110vils in 30minutes and starts adding trade carts around that time. (the 'trader') The trader invests wood and gold (at a slow pace) into trade carts. It takes 14minutes to break even with the 'seller', and from 44 minutes onwards he makes considerably more profit than the 'seller'

the other player starts to sell resources during this time which profits him resources equal to the cost of 40 mangudai over the 14minutes he has the advantage, compared to the 'trader' that is still waiting for his carts to return with the gold.
(the 'seller')

The trader will have trade setup which does benefit him in the long run (gold income and population efficiency from trade is significantly larger than that of selling, in the long run)

Some nitpicking:
First of all, the saved resources might equal 40 mangudai, but this does not consider that it requires 2.5castles to get the 40 mangudai out at in those 14minutes that you do have the advantage. Which means 1950resources put into stone, which in itself is a huge investment. On top off that factor in the resources spend on upgrades (which for mangudai is HUGE), and the villager building time spend on building the castles and the buildings for upgrades.

Secondly, the first carts added will have a more efficient route, since there are less trade carts blocking their path at this stage in the game.

I assume apart from this all your maths and testing was done correctly


Comparing the scenario of 'trader' and 'seller' to 1TC push or boom.
I think your scenario can be very well compared to a similar scenario of a player who booms (puts investment into villagers from multiple TC’s) versus a player who goes aggressive on 1 TC.

The player that decides to go with 1TC aggression does not invest into additional town centers, villagers and farms, but instead uses the resources he saves with this to invest into army or a fast imperial age, which in any way gives him a power spike. The idea is to deal enough damage quick enough to neutralize the opponent, resulting in a momentum swing into his favor which could lead to victory.
Less ideally, the push is not strong enough but cripples his opponent enough to allow him to add his own economy or granting him crucial map control, which could net into an advantage or at least equal chances in the late game.

If the aggressor does not inflict enough damage, the investment into economy from the boomer will pay off quickly. This allows the latter to get a significant economy lead, which can translate into powerful army that can steamroll the 1TC aggressor. Like we see from many (pro)games on several maps the 1TC push could work out but must be quick and decisive. (I consider Arabia and arena, where this strategy choice is most prominent, on BF it happens occasionally too)

Momentum
In age of Empires momentum swings are an extremely important factor in the outcome of a game. To win a game, players must control their momentum. If they are attacked and lose ground or units, that is bad for them, but they only lose when they lose so much momentum that they can't come back from it. (read: losing control over resources, losing a significant number of buildings/farms/houses, or losing a majority of army causing them to be significantly outnumbered). If a player can manage to hold their ground, even when taking bad engagements, but meanwhile can make an investment pay itself off they are likely to win the game later. If a 3 TC boomer can survive long enough their superior economy will pay off and they will most likely push back the guy that went 1TC aggression.


How to utilize the power spike from the 'seller'
In your scenario, the 'seller' would have 40 additional mangudai out after 14minutes after that -> at 44:00minutes.
In the time he has the advantage he is slowly building up his forces. He must do damage to the opponent before 44minutes, when trade starts to pay off. Let's say he prepared his mangudai and moves out with 30 mangudai at 40minutes, hoping to use his advantage to the opponent who was investing resources into trade at that point. How big is the chance that the 'seller' who is using his power spike can indeed inflict significant damage (preferably destroying some of the trade carts of the opposing team) when the opponent that decided to add the additional investment would most likely try to turtle up with walls, defensive castles and defensive units (all take minimal investment and have the primary purpose of delaying the aggressor, which buys time for his long-term investment to pay off).

I don't think in a map as easy to wall as black forest the aggressor will come far with his 30mangudai push in this situation. However, there might be situations that are more prone to aggression, like if you were able to sneak into enemy territory, or a teammate is doing a fast imp/siege monk push. in that case having those additional mangudai out early will be very significant.

A hybrid
Just like with a failed 1TC push, when the 'seller' can't inflict enough damage or swing the momentum into his favor considerably, all is not lost. Hopefully as a 'seller' you have used your additional resources/army to at least deal damage to the enemy, delaying the time it takes for his investment to pay off (maybe you killed a few trade carts or forced him to defend which caused him to take unfavorable trades, or maybe you killed some other economy like buildings/villagers)

In this case you can still add the investment. Just like with the 'failed' 1TC push you might have gained control over important areas which are easily defended and you crippled your opponent enough to catch a breath and start adding some investment yourself (if your push can't finish the job, you DEFINITELY want to prepare for late game, which means adding trade or in the 1versus3TC scenario adding more town centers and villagers)

Conclusion
Funny enough the conclusion is very similar to what most of us here would consider to be common sense:

- selling instead of trading gives you a power spike for the first couple of minutes (roughly between 30-44minutes)
- the resources you save can be to aggressive use.
- the game is decided by momentum swings. A power spike can be such a swing. However, if the short-term power spike is not inflicting enough damage or resulting in other forms of control, the momentum will very likely swing back into the favor of the player who did the long-term investment
- (In my opinion) The seller versus trader approach compares well to the 1TC push versus multiple TC boom.
- going for the selling approach might be the best in certain situations. In all these situations you maximize the potential of your power spike and you try to either win very fast, taking advantage of a huge momentum; or you want to bring enough damage to the opponent to equalize him, which allows you to add your own long-term investment afterwards. For black forest team game, a situation like this is not often seen but an example might be your flank going for early game aggression (a sneak, siege monk push or fast imp) in that case your power spike will add on top of that of your ally which hopefully gives you enough momentum to steamroll the opponents. On top of that such scenario will allow your military units to be utilized more efficiently (you ally might already have breached opponents walls)
- if you go for the sell approach and you feel like the game is not going to be finished early you could always fall back to the trade approach. This could be rough, unless your aggression got you into a position to safely add more economy/opportunity to invest in trade (long term investment).


PS: just like with a 1TC push, your 'selling approach' can have great outcome in lower level play, because the player investing in the long term (read booms on multiple tc's) is less likely to succesfully defend with minimal resources/investment. Pro players are very good in defending against aggresive plays (for example surviving the trush on arena, or making army while booming on arabia) and they can often defend untill their longterm investment will pay off.
If you try to do the 1TC push (assuming you practiced it) on your 1750HD skill range you might find yourself often very succesfull. Similarly your 30-44minute powerspike by not adding trade on Blackforest might in practice work on your skilllevel, since your opponents are not experienced with defending against such aggresion.

 

[ChriSt]

The prime example of AoE2 elitism

I didn't tell you to get out of the sandbox yet

 

[Socksyy]

Trade is for tunas, just spam trash until loss

 

[Modri]

A funny observation: in any atleast intermediate voobly BF tg (boom game --> no significant rush happened) you can actually judge a player's strength quite correctly from the amount of trade carts he has at lets say min 40. 2k players will have more carts than 18xx players who will have more carts than 16xx players etc. Down at the bottom will be noobs who aren't knowledgeable or skilled enough yet to start trading in time and will keep selling w/f and when eventually they have no real army left and one of the better teammates will ask them why is it so, they will cry out in despair: "no gold man!"

(few people like Memb, MBL and some DM players like Christ are notable exceptions to this rule)

 

[kw1k000000]

Setting up no trade and going for selling wood food for long time (30 to 45 min) is an All-in strategy with limited upside and huge down side.

Even on Arabia if the game is even i.e. no team is significantly behind at that point then it never gets decided in early imp and it almost always goes to post imp with trade situation. The problem with relying on selling is if it doesn't work then you are basically left with nothing, no proper army and no eco.

Also its not like you can not make trade and army at the same time. Producing trade carts from 3 markets and fully upgrading + maintaining an expensive army like Mangudies is easily possible if you had a good boom. There is no reason to not set up the trade because there is no real upside to it.

Maths can explain everything but not everything can be explained by simple maths which has been applied here.

 

I didn't tell you to get out of the sandbox yet

Continue digging your hole further bud

 

[NachoAoE]

I don't think numbers are the problem here, it's how they're being interpreted. To get better numbers do this:

Play a significant amount of games ( I would recommend a classic 5% type 1 error ), so n should be high enough. And then see if you win more or lose more with and without trade. Go directly to the problem

Easy example:
Play 500 games without trade
play 500 with trade
~ Players arround same level of course
If by then you find that trade isn't worth it, then I will definitely reconsider looking at numbers and applying it to good level games. Thing is I don't believe you have that 1000+ games experience, be it from actually paying close trade atention to trade carts in Recs or playing.
I can assure you that won't happen, I don't have numbers or arguments, just the knowledge and experience to know that most the TGs my team loses is decided for having less/unefficient or late trade. I even know that one of the most powerful TG strategys is to camp a good hill on enemys corner, so their trade gets worse. Having 3 corners is practically insta-win if nobody titanics

 

@RicoJay13 have you also considered the fact that maybe pros have tried this in the past but it didn't work? The meta is what it is for a reason (albeit it does change a lot, just a year ago Eagle Warriors weren't common in castle age until DauT tried them out successfully).

 

[dodageka]

Just my two cents

I like your videos in general with respect to production quality and humor, that’s great

The general idea also isn’t wrong, there are some expert team games as well were they delay trade (because they are behind and need full focus on defense or feel the enemies have started trade and want to gain momentum through a short power spike) but generally you will see trade

I think you are ignoring two points in your analysis that could drastically change the outcome. First, when you start investing into trade carts, you won't necessarily stop vill production, so as long as you have pop space you will benefit from the "seller" effect as well, converting your surplus food and wood into gold.

At the same time, if you forgo trade you might not be able to actually create 40 additional units as you should be pop capped as well. In this case making use of your additional resources from not investing into trade would require you to force your opponent to take a fight and trade units (that you might find easier to replenish). However, this puts a lot of pressure to make something happen, and this pressure can often be abused by the other side (when you have to make something happen you have to take bad fights as well, like with Mangudai against Skirms, which is bad)