These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

Market Discussions

 
  • Topic is locked indefinitely.
 

How to read the in game EvE market charts

Author
Vaerah Vahrokha
Vahrokh Consulting
#1 - 2011-12-27 01:53:34 UTC
Preamble

A slightly different text, better, illustrated and zoomable version of this tutorial is available on my website, at this location.


Intended audience

- Beginner EvE traders who opened an EvE market chart for the first time and wondered what is it about and what are all those "lines".

- Experienced EvE traders who want to push their experience a bit farter. They will be covered after the basics sections.

- EvE players who cannot / don't wish to commit themselves to learning advanced, RL trading alike methods like Experiment #01: RL finance analysis applied to EvE.



How do I show a market graph?

a) Open this screenshot on another browser window / tab.

b) Click on the Market button on your buttons bar (below your character's portrait), marked as "1" on the screen shot.

c) Search / browse / use quickbar for an item you want to check out. In this tutorial, we will look at Megacyte. It's marked as "2" on the screen shot.

d) You should see the desired item's name at the top (marked as "3") and below it, the list of current sell and buy orders.
Sorting the orders, placing buy / sell orders are beyond the scope of this article, there are other tutorials which extensively cover this "Market Data" tab.

e) You are in the Market Data. In order to see the graph, you have to click the "Price History" tab, marked as "4" on the screen shot.


What's a market graph?

It's a visual representation of a market prices history.

Please open this screenshot on another browser window / tab. I will refer to this graph as "Window 1".
You may look at (and switch to / from) "dry numbers" by actually clicking on the "Show Table" button, marked as "1". The table looks like this other screenshot. I will refer to this graph as "Window 2".

The table and the graph contain the same information about the only market protagonist: price. Nothing else matters.
Both the table and the graph show the price for that market in the last selected period of time. To select which period to look at, click on the "Time" selector at the bottom of the window, marked as "2". It's available on both Window 1 and 2.

Each recorded trading session is 1 day long (see the first column in the table) and comes with the following information: Date of the session (time is always 00:00:00), number of orders executed, number of units traded, lowest price for that session, highest price for that session and "sanitized" average (CCP employs algorythms to clean stray / purposedly deceiving orders. They are applied on next day, so it's possible to see weird values for the current session).

Let's get back to the graph (press "Show Graph" if you are in the table view / Window 2).
Vaerah Vahrokha
Vahrokh Consulting
#2 - 2011-12-27 01:57:40 UTC
Elements in a market graph

Looking at the graph screenshot again, most relevant elements have been marked with numbers:

1) As discussed above, this button lets switch between graph and table format.

2) This selector ("Time") lets select the time frame for the graph and table. By properly setting up the time frame we may gather additional information, the most useful settings range from Year down to 3 months.

The "Right click graph to configure filters" option next to "Time" is also useful but beyond the scope of an introducing tutorial. Most of the time the graph is good as is, you might want to disable some features in case some market manipulators put in fake orders to make the graph become a thin horizontal line (to hide their activity).

3) and 4) The red and green lines / curves on the graph are labelled respectively as 5 and 20 days simple moving averages (SMAs). A moving average is a so called "lagging indicator". Its points are plotted by calculating the prices average of N days in the past. 5 days moving average means 5 days were considered in the computation. They are called "simple" because they are based on a basic average.
More advanced moving averages involve applying a bigger importance (weight) on the last few days, by using exponential averaging and other. EvE only comes with the simple kind (the least powerful). Moving averages are called "lagging indicators" because, since they are based on past days prices, they lag behind the current price action.

What a moving average does is to smooth the price variations so that you may see the price motions in a simpler way, like a game of "connect the dots with a line".
Multiple moving averages have different "speeds", a general rule of thumb of "ancient" technical analysis is to buy when the red (5 SMA) crosses the green (20) SMA from below and sell when the red SMA crosses the green SMA from above. But remember, those are ancient advices used decades ago. They still work sometimes in EvE because its markets behave like old markets but it's no way a reliable way to trade.

The only more or less reliable way to trade with moving averages has been explained in the first part of the old forums first part of Experiment #01: RL finance analysis applied to EvE.

5) Donchian channel is another ancient lagging indicator. It somewhat helps determining whether the price is pushing up or not.

6) Median Day Price represents the median (not the average!) of the trades prices day after day. It's the "yellow dots" that form the price graph.

7) This is the price graph with all the elements: the daily sequence of yellow dots representing the prices median superimposed by Donchian channel and moving averages.

There are also some other very important elements: above and below each price dot, there are two very thin "barbs" or "wicks". They represent the minimum and maximum for the prices of that day and are a very useful and not lagging instrument to "see" the market sentiment and the amount of buyers and sellers in any given day.

As shown by this screenshot, the barb sitting above the yellow price dot is called "shadow", the one sitting below is called "tail". It's possible for some sessions to show just one of those or even none.

8) Price levels. These lines are automatically calculated and rescaled. Sometimes they help, since some traders place buy and sell orders in their proximity, some times they don't and only add to confusion.

9) Volume. This is another non lagging indicator that may help determine whether a price increase is due to true demand or is just a speculative bubble ready to burst and crash. Volume also indirectly helps determining whether a market is liquid or not. If it is, it will tend to be smoother and speculations will be harder to setup. Markets with just a couple of items traded a day are dangerous!



In the next days more chapters will be added, showing how to read some of the many subtle hints the markets may give us about their past, present and future behaviors.
bubble trout
Brutor Tribe
Minmatar Republic
#3 - 2011-12-27 12:04:03 UTC
Thank you very much for this. I'm an extremely new player who has been trading. Getting an in-depth explanation of each component of the graph, and furthermore how to read and use this information helps greatly. I look forward to whatever you add to this.

Two quick questions.

Firstly, can you explain how the Donchian channel is calculated? You don't seem to think it important, but I would like to know how it is calculated, if only to satisfy my own innate curiosity. Your explanation of the other features was very in informational ( I hadn't any clue as to what the "barbs" were extending from the median day price).

My other question was if "dynamic range markets" could raise instead of fall in eve. In your example graph the three you pointed out were all "falling". I assume that this is a period where no external factors are effecting the supply or demand and traders are competing, overall lowering price to sell their stocks of whatever in a slow but controlled manner. Is it possible (in eve) for demand to slowly outstrip supply, resulting in a gradual but stable raise in prices over either or both the short or long term? Or do people speculate too much in eve for this to take place and instead all we see are exponential or semi-exponential raises followed by a crash to whatever the true price people are willing to pay? Perhaps I don't understand the term dynamic range markets, or there may be factors I don't know/am not considering.

Thanks again for this explanation. Smile
malaire
#4 - 2011-12-27 12:50:04 UTC
bubble trout wrote:
Firstly, can you explain how the Donchian channel is calculated?

In short, Donchian channel is area between highest High-price during last 5 days and lowest Low-price during last 5 days (High and Low prices being those which are shown as lines on Market History Graph or in High and Low columns in Market History Table.)

New to EVE? Don't forget to read: The Manual * The Wiki * The Career Options * and everything else

Vaerah Vahrokha
Vahrokh Consulting
#5 - 2011-12-27 13:04:21 UTC
bubble trout wrote:
Thank you very much for this. I'm an extremely new player who has been trading. Getting an in-depth explanation of each component of the graph, and furthermore how to read and use this information helps greatly. I look forward to whatever you add to this.


Thank you for the appreciation!


bubble trout wrote:

Firstly, can you explain how the Donchian channel is calculated? You don't seem to think it important, but I would like to know how it is calculated, if only to satisfy my own innate curiosity. Your explanation of the other features was very in informational ( I hadn't any clue as to what the "barbs" were extending from the median day price).


I am one of those traders who don't like indicators (they are lagging = imprecise and often confusing), it's why I focus more on what a market is about: price. Price. And price.
Anyway a Donchian Channel is just one of the "channel" family of indicators (Keltner bands, STARC bands, Bollinger bands...) that give buy and sell signals depending on whether price goes above or below it.

To calculate it, you choose a number of past days (5 fixed value in case of EvE) and apply this formula:

High bounding line = Highest price(H1, H2, H3, H4, ... Hn)
Low bounding line = Lowest price(L1, L2, L3, L4, ... Ln)

where H1 / L1 etc. are the high / low for the indicated past day.


bubble trout wrote:

My other question was if "dynamic range markets" could raise instead of fall in eve. In your example graph the three you pointed out were all "falling". I assume that this is a period where no external factors are effecting the supply or demand and traders are competing, overall lowering price to sell their stocks of whatever in a slow but controlled manner. Is it possible (in eve) for demand to slowly outstrip supply, resulting in a gradual but stable raise in prices over either or both the short or long term? Or do people speculate too much in eve for this to take place and instead all we see are exponential or semi-exponential raises followed by a crash to whatever the true price people are willing to pay? Perhaps I don't understand the term dynamic range markets, or there may be factors I don't know/am not considering.

Thanks again for this explanation. Smile


They can, they can. Rising RMs are rarer because traders are driven by greed and fear. Greed makes ascending markets go up steadily (thus few RMs) and many just jump on the bandwagon.
Fear makes traders try hold on their stock during falling markets while buyers want to buy low, not realizing their low could be just an intermediate stop before a further fall. This results in steep price descents mixed with RMs.


Zothike
State War Academy
Caldari State
#6 - 2011-12-27 15:31:15 UTC
nice tutorial +1
Thoraemond
Far Ranger
#7 - 2011-12-27 17:07:08 UTC
A lot of handy information.


Vaerah Vahrokha wrote:
6) Median Day Price represents the median (not the average!) of the trades prices day after day. It's the "yellow dots" that form the price graph.

When did this change?

The last time I tested this was about three years ago, at which time the problem was reproduced by a Bug Hunter: Back then, the chart's label definitely said "median" but it definitely actually displayed "mean" price information. The table then had a column for "Average" which included the "mean" price.

(Note that since "mean" and "median" are two types of "average", using the word "average" without describing a type of average is often over-vague and can lead to confusion.)

My hypothesis was that maybe historically the information had been a median, but since a mean is computationally easier to calculate than a median on a rolling basis through the day, the system had been changed to capture means and the labels had not been updated.


Vaerah Vahrokha wrote:
Each recorded trading session is 1 day long (see the first column in the table) and comes with the following information: Date of the session (time is always 00:00:00), number of orders executed, number of units traded, lowest price for that session, highest price for that session and "sanitized" average (CCP employs algorithms to clean stray / supposedly deceiving orders. They are applied on next day, so it's possible to see weird values for the current session).


Is it not a bit inappropriate to talk about trading sessions as a analogous to RL trading sessions? In RL, a session is only a small fraction of the day (e.g., 7 h), whereas in New Eden, a trading day is typically approximately the entirely of each day. Especially since down-times are in the middle of each calendar day, there is no expected difference between what you might call the "closing" price one day, and the "opening" price the next day, since those two periods are separated by only a single clock tick (rather than, to match the above example of 7 h of trading time, 17 h in RL).

Surely any method of analysis that requires opening and closing prices is doomed to have a certain increased error rate over its application in RL when applied in New Eden, since there is no "open time" and no "closed time" like in RL?
Vaerah Vahrokha
Vahrokh Consulting
#8 - 2011-12-27 19:13:12 UTC  |  Edited by: Vaerah Vahrokha
Thoraemond wrote:
A lot of handy information.

When did this change?

The last time I tested this was about three years ago, at which time the problem was reproduced by a Bug Hunter: Back then, the chart's label definitely said "median" but it definitely actually displayed "mean" price information. The table then had a column for "Average" which included the "mean" price.


This would definitely deserve further research. For what I have learned by toying with EvE's cache it looks like it has one behavior for past days vs current. The current days seems "simplified" since it's managed in real time, while the past days are fully featured. We also have the additional variable of the data being sanitized, so who knows which is the real minimum and maximum? Were the "dumb" purchases sanitized as well? (IE the guy buying a badger for 250M)?


Thoraemond wrote:

Is it not a bit inappropriate to talk about trading sessions as a analogous to RL trading sessions? In RL, a session is only a small fraction of the day (e.g., 7 h), whereas in New Eden, a trading day is typically approximately the entirely of each day. Especially since down-times are in the middle of each calendar day, there is no expected difference between what you might call the "closing" price one day, and the "opening" price the next day, since those two periods are separated by only a single clock tick (rather than, to match the above example of 7 h of trading time, 17 h in RL).

Surely any method of analysis that requires opening and closing prices is doomed to have a certain increased error rate over its application in RL when applied in New Eden, since there is no "open time" and no "closed time" like in RL?


Markets are not just stocks or futures. The Forex market is officially open 24 hours a day from 5pm EST on Sunday until 4pm EST Friday yet most large brokers allow for week end trading as well (see Oanda).

It's a quite complex matter to discuss but here's a brief.

1) Actually what influences a "continuous" market like EvE or Forex are not paper-declared hours but actual market interactions. I.e. you may look for autorithative price action before the UK open, before the USA open and so on. Why? Because the markets "live" and "move" in the first minutes / hours of those sub-sessions. I would look for analysis on previous day price action before UK open so that after UK opens the price will soar and bring in the profits.

2) Market maker brokers each have their own "daily candle open" times. Often they make it coincide with one of the big market trading hours but it's not a given. A relevant percentage of traders use such brokers.

3) Some trading packages let you define your own opening and closing times, in order to let you analyze markets at "human" hours and not i.e. at 3am. While this yields to less reliable analysis (since markets do not care about your hours or how you split YOUR candles), quite often it still works. In fact, candle bars have many neat properties (like in math you may get commutative property, permutation of terms etc). One of these properties lets you split a candle saying "something" into two saying the same "something" and vice versa. I.e. the "real" outer market could display a buy signal due to a pin bar, while your platform might split it into a two candles BUOB (another buy signal candle formation).
Vaerah Vahrokha
Vahrokh Consulting
#9 - 2011-12-27 19:14:31 UTC  |  Edited by: Vaerah Vahrokha
The tutorial has been greatly expanded on my web site, at this location. Portions will be added here as time allows.







Some markets properties

Markets are pushed by demand and offer and are driven by contrasting greed and fear.

Healthy markets always oscillate in an unpredictable zig zag motion. Markets are fractally self-similar, you may zoom in or zoom out and find the same kind of zig zag motions containing or being contained by other zig zags.

Price perpetually moves in an highly optimized manner, filling all the past and present available price levels and gaps (akin to how plants optimize light exposure by disposing their leaves and petals by Fibonacci sequences). This perennial motion is called “price discovery”.

Markets appear to move randomly (“random walk, brownian motion”) but that’s only true in the smallest time frames where micro-transactions between individuals form “noise”. The longer time frames are driven by demand and supply balance variations.

You may profit by knowing why / how / when demand and supply are changing (performing a so called fundamental analysis, i.e. reading patch notes about buffs or nerfs) or by reading the charts and understanding how and when demand and supply are changing (technical analysis). Fundamental analysis is potentially the most profitable kind of strategy and gives a time advantage in trading but requires a robust knowledge of many, often intermixing underlying factors including politics and rumors.



Trending and ranging markets

From about 70 to 80 percent of the time, markets find themselves in a state of quasi-stable equilibrium where prices don’t change much. Price stays confined and oscillates within a relatively small range. This price motion is called “range market” or RM.

When this balance is broken by any important enough event, price adjusts to the new equilibrium, often with some lag, overshooting the target and oscillating around it. The speed and smoothness of this process depend on many factors including available liquidity, market speed, big positions being closed speculations, manipulations. This price motion is called “trending market” or trending.

This in game graph shows some RM and trends.

The blue rectangles enclose RMs (ranging market zones) and are marked as “1″ on the chart. As you may see, there are many (I did not even mark all of them). They are delimited by an horizontal, plain rectangle and are also known as “static RMs” because of that.

RMs can also form in a diagonal direction. They are rarer and are called “dynamic RMs”. They are the red rotated rectangles marked as “2″.

Since markets are fractally self-similar, features like RMs may enclose other RMs. An example of dynamic RM inside a longer term static RM has been marked as “3″ on the chart.

The graph also shows some trends, the yellow thick lines marked as “4″. It’s easy to see how a market tends to range for most of the time and trends act as connectors between the various ranging zones.
Vaerah Vahrokha
Vahrokh Consulting
#10 - 2011-12-27 20:27:30 UTC
Identifying trends

Knowing when a market is trending and when it is ranging helps with trading. Traders want to trade in the same direction of the trend and never counter-trend (an old trader motto says: “the trend is your friend”). It is also possible to profitably trade inside RMs but it’s harder, expecially because in EvE it’s impossible to short sell. Therefore once identified a RM, it’s necessary to buy when price hits its bottom and sell when it hits its top.

Which elements make it possible to say “this is a RM” and which others “this is a trend”?

As said above, price moves in a zig zag motion, each zig zag is called “swing”. The placement of the next zig zags determines whether we are in a trend or RM. A zig zag looking like a nice ladder with ascending or descending swings is a trend. A random zig zag with same level or random swings is a RM.

This picture shows a descending trend or downtrend.

The blue segments are the price swinging down, the “HL” and “LL” labels mark the swings tops and bottoms. In particular, a donwtrend is identified by an orderly sequence of Higher Lows (HL) followed by Lower Lows (LL), like descending a ladder. The red dotted lines and arrows show the zig zags.

This next picture shows an ascending trend or uptrend.

This is exactly the most profitable setup in EvE Online. This time the blue segment (swings) climbs up forming Higher Lows (HL) and then Higher Highs (HH) like a climbing ladder. The zig zags show the same behavior. Uptrends are fairly rare and often caused by new patches altering the demand / supply equilibrium towards the demand side.



In the next section we will see both what are range markets, the dos and don'ts and most of all, EvE trading tips including both buy and hold and high frequency (0.01 ISK) trading ones.
Vaerah Vahrokha
Vahrokh Consulting
#11 - 2012-01-03 09:08:05 UTC
Taken from the updated tutorial on my web site at this location.


Identifying range markets

Range markets exist when there is no clearly identified direction for the swings / zig zags. There’s no “ladder” shape in the swings, they randomly go up and down often hitting an invisible ceiling and bottom.

This picture shows a typical RM.

It’s evident how the highs and lows sequence is completely irregular and shows no real direction.

Despite the irregularity, it’s possible to notice a peculiarity: RMs seem to stay confined between a top and a bottom limits.

Drawing those imaginary limits on the above example would lead to this picture.

Why does price tend to hit (and be stopped by) RMs tops and bottoms? The answer is very complex and it won’t be fully covered here. A simplified version will be given.

RMs are places where demand meets offer and traders place many orders around there. When price hits the top that the demand / offer equilibrium allow for, sellers orders (greatly aided by supply being stronger than demand) take the upper hand, sometimes with real “blocking orders” (big orders hard to fill). This behavior institutes what’s known as “resistance” (R), that is an imaginary line that repels price down, like a barrier. All the RMs, both static and dynamic have such resistance price level as their top.

The same happens for the opposite case, when price hits a RM floor. Buyers, greatly aided by demand being stronger than supply, take the upper hand, sometimes with massive blocking buy orders placed right below the floor. This forces the price to stay at or over an imaginary line called “support” (S).
Vaerah Vahrokha
Vahrokh Consulting
#12 - 2012-01-03 09:13:21 UTC
Last but not least, there are four very important rules about the support and resistance price levels:

1) The more price hits them, the stronger they become. Traders confidence on these levels “holding” against price rises over time and they place more and more orders mutually reinforcing themselves and their beliefs.

2) The more factors converge into forming a support or resistance, the stronger it tends to get. If a support or resistance happen to coincide with an easily remembered round number (aka RN, i.e. 2900 ISK pu) or, even more, big round number (aka BRN, i.e. 3000 ISK pu), those price level lines hold stronger. Rarely, there are further, exotic convergences: an horizontal support that also coincides with the extension of a diagonal previous support, or a support sitting at a so called Fibonacci number (humorous link). The more the factors involved in this so called “convergence” the stronger the price level holds.

3) Markets have a very strong memory. Once price explored a new support / resistance, that level will be reused later, sometimes after years. There are some old stocks that persistently hit levels they reached 30 years in the past. This memory effect is due to traders using graphs (self fulfilling prophecy) but also because that market demand / support equilibrium “resonates” at those levels by itself. In EvE this effect is reduced both because very few traders use graphs and markets liquidity is much lower than the real stocks markets, therefore patch and player induced variations may affect markets much more than they do in real markets. Each market has his own temper, some have strong memory others don’t.

4) When a breach of those levels occurs, it’s sudden and strong, like a dam bursting. And very curiously, when price gets back to them, they switch “role”. Supports that held price up, become resistances that will try avoid price getting back up and vice versa.

How the whole mechanism works is shown by this picture.

In the first section of the price graph, marked as “1″, there is a RM. The red line is the resistance level where sellers liquidity prevails, the green line is the support level where buyers liquidity prevails.

Something happens to the demand / offer equilibrium and price breaks thru the support. It falls to a lower level (next green support line below) that very often was already reached in the past due to the above mentioned memory effect. Then it moves up and will hit what was the previous support line and the roles inversion happens: the former green support line above now acts as resistance (thus it’s painted red now, marked as “2″ on the chart).

Price forms another RM until something happes again and the market crashes down to the level marked as “3″ on the graph.

This time price decides to trend up, as the climbing up zig zags show, until it seems to hit a diagonal, dynamic support line marked as “4″ on the chart and bounces up and down forming a dynamic RM. Exactly as for the static (horizontal) RMs seen above, price may breach its RM bounds and then use the extension of the resistance as new support (marked as “5″).

Not incidentally, the first price zig zag close to “5″ snaps exactly where in the past, at the same level, we had the support of that RM marked as “2″. The last zig zag all to the right also acts in a similar way (snaps to the same past RM’s resistance). This picture shows dashed lines to visually aid seeing the connections.
Gatan Hahran
Brukterer
#13 - 2012-01-03 09:54:02 UTC
Vaerah Vahrokha wrote:
Fibonacci number (humorous link).


Humorous Fibonacci picture for people who dont like walls of text!
You wanna add humorous fibonacci content for ALL reader groups.
Arcathra
Technodyne Ltd.
#14 - 2012-01-03 12:00:57 UTC
Thanks a lot!

Your writing gave me a lot better understanding of the graphs, I might even try start trading by myself. Smile
Vaerah Vahrokha
Vahrokh Consulting
#15 - 2012-01-03 12:02:59 UTC
Arcathra wrote:
Thanks a lot!

Your writing gave me a lot better understanding of the graphs, I might even try start trading by myself. Smile


Thank you!

In the next brief article I will post how you can use these market notions to make your trading safer, even for those who are only interested at the 0.01 ISK game.