open All Channels
seplocked Market Discussions
blankseplocked Experiment #01: RL finance analysis applied to EvE
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: first : previous : 1 2 3 4 [5] 6 7 8 9 ... : last (11)

Author Topic

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.20 20:19:00 - [121]
 

Update

Robotics keep dropping as per previous analysis.

------------------------------------------------


As per previous reply to Akita T, Technetium is still not not giving clear signs about its future.
It rallied up at a best Gann 45 slope and subsequently is beginning going hyperbolic. This should lead to a trend exhaustion.

There is a "new" analysis I have not covered here yet, that might apply to this case and could be a reply to the requests of the poll above.

In RL trading, EMA crosses may indicate a trend retracement and further continuation up/downwards. I remind that EMA stands for "exponential moving average", that is an average of the previous N bars (days in EvE), where N is called "period".
Now, EvE comes with two SMAs with 5 (red) and 20 periods (green), both with "wrong" type and periods for an accurate study, but we'll do with them.


Let's look at this picture:


Picture linkage


http://i285.photobucket.com/albums/ll45/vahrokh/EvE/Finance/2010-09-20_Technetium_pullbacks.png



We'll skip the first spike and patches related effects. There is a very long consolidation (indicated as Flat on the graph) spanning almost two months.
Consolidations have several effects, one of which is to accumulate momentum that later on becomes instability and finally explodes in a trend.
This happens at the beginning of August, it's visually announced by the yellow dots (price) starting to climb and by the SMAs diverging. The faster period SMA in fact is calculated just on 5 periods and therefore it'll change quicker than the other.
The spread between the two may be used in analysis.
In fact, trends tend to retrace or "fall back" on themselves when they lose some steam. The retracements are small consolidations helping the trend propel up some more.
In RL finance a trend usually shows from 2 to 4 of these retracements before going "hyperbolical" and exausting itself for good (with a so called pin bar / hanging man / doji...) and starting to invert in the opposite direction.

After a while, at point 1) we witness a first retracement (also notice the volume). It's so hard that at first it looks like an early left shoulder. At the same time, that retracement is also a break out above the consolidation as we have discussed in previous posts.
The trend resumes its run upwards with a particular 45-ish angle that a theory attributed to Gann states it's one of the best to keep a steady trend progression.
After a while there is a second retracement at about 53k.
The trend resumes again, but now it's going more vertical. This means that despite the retracement the trend is not stabilizing enough.
This is expected, since as I said above, there should be but 1-2 more retracements before it runs out of steam. At that point the trend could hit a ceiling (a shoulder), see some sales and then resume up forming an head and then proceed to form the right shoulder and crash.
Otherwise it could just hit a ceiling, try passing it once or twice forming the so called "double / triple top" and then crash.

(Continued...)

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.20 20:41:00 - [122]
 

Edited by: Vaerah Vahrokha on 20/09/2010 20:42:01
The following picture shows a possible future for Technetium.
It's so large that it's shown as thumbnail. Click on the "linkage" link for the full size picture.


Full size picture linkage


http://i285.photobucket.com/albums/ll45/vahrokh/EvE/Finance/2010-09-20_Technetium-1.png



Now, this picture is NOT built with analysis, as of today there are no signs nor old data to support or deny it. It's an educated guess about possible price levels given the current pivot lines and the estimated permanently increased demand even after the speculation will be over.

In the next weeks there will be updates about how this trend unfolds.

Delagos Almondis
Posted - 2010.09.20 21:13:00 - [123]
 

As always, very interesting.

I noticed another pattern you talked about a few posts earlier, and wanted to give a different estimation based on it. Please note that I have no real clue about this and this is more of a question than a read estimation, but:
The traded volume has been dropping steadily the last few months, and its still dropping. In one post, you said that this was an indication that a trend was loosing its steam and might be dropping shortly (-> Robotics). What is different in this example that makes you think it will go up a fair bit before finally dropping?

My guess would be that the volume is dropping at a steady speed, but I wanted to hear your educated opinion about this. I'm trying to learn after all Very Happy

Keep up the good work (Are you doing this while ice mining? Very Happy)

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.20 21:30:00 - [124]
 

Originally by: Delagos Almondis
As always, very interesting.

I noticed another pattern you talked about a few posts earlier, and wanted to give a different estimation based on it. Please note that I have no real clue about this and this is more of a question than a read estimation, but:
The traded volume has been dropping steadily the last few months, and its still dropping. In one post, you said that this was an indication that a trend was loosing its steam and might be dropping shortly (-> Robotics). What is different in this example that makes you think it will go up a fair bit before finally dropping?

My guess would be that the volume is dropping at a steady speed, but I wanted to hear your educated opinion about this. I'm trying to learn after all Very Happy

Keep up the good work (Are you doing this while ice mining? Very Happy)


When a trend runs out of steam (be it at a top, at head and shoulders etc.) you should see a visible and brisk drop in volume (no one buys any longer except lemmings and slowpokes due to get burned). Volume just thinning out does not say what's going to happen next for sure, it could be just due to a temporary shortage, due to an holiday and so on.

Akita T
Caldari Navy Volunteer Task Force
Posted - 2010.09.21 00:03:00 - [125]
 

Originally by: Vaerah Vahrokha
Full size picture linkage
Now, this picture is NOT built with analysis, as of today there are no signs nor old data to support or deny it. It's an educated guess about possible price levels given the current pivot lines and the estimated permanently increased demand even after the speculation will be over.

The background analysis looks something like this...

It is noteworthy that moon mineral and advanced material Jita trade represents a majority of total game-wide market trade, while T2 item trade in Jita represents at most about half of the game-wide totals, if not less.
Before December 2010, the typical daily Jita trade of finished T2 items represented around 345k worth of Technetium, and the next day after the Dominion patch the exact same items would have represented around 670k worth of Technetium. A game-wide estimate of Technetium usage from just that is extremely unreliable, but could be anywhere from just 1 mil units/day to 1.5 mil units/day.
There's a CONFIRMED total of 230 Technetium moons game-wide (~552k/day production), and estimates of actual total count are also highly unreliable, ranging from 360 to 400, with the most likely number around 380 (so maximum possible extraction rates in case of no conflicts preventing moon mining somewhere between 864k and 960k per day, with the most likely amount 912k per day). Bottom line, the most likely current usage outstrips best possible extraction rate.

As a rough "actual demand" volume indicator, 1 unit of Nanotransistors contains 1/30 unit of Technetium, and 1 unit of Fullerides contains 1/60 unit of Technetium. No other advanced materials contain Technetium. Recent Technetium volumes hovered around 800k daily in The Forge, with Fullerides around 25 mil (so ~415k Tech) and Nanotransistors around 15 mil (so ~500k) for a total of ~910k Technetium equivalent traded into advanced materials. Technetium market trade volumes outside of Jita are negligible, and advanced material market trade outside Jita (at least in empire regions) is around 20% extra tops.
So if we were to follow just the advanced materials vehiculated in empire space, the Technetium usage rate lie somewhere in the vicinity of 1 mil units/day, but that's ignoring people that integrate all production or sell their advanced materials wholesale via contracts or direct trades. Then again you can say the same for finished T2 items or raw moon minerals, I guess.
The most likely scenario is that at best, demand is still above extraction rate.

There were up to three years worth of stockpiling (between one quarter and one half of potential extraction rates while prices were reasonably high to make mining it somewhat desirable, but unlikely to be fully mined, at least in the 9 months before last December) and several years before of much lower extraction rates and demand.
Since last December end-product-level usage has slightly more than doubled compared to the previous 2+ years (but also, rate of inclusion directly into end-level products as opposed to being sold "as is" has also increased).
The previously existing stockpiles could be anywhere between one quarter and three quarter depleted, with the most likely amount being somewhere around halfway depleted.

What is so far almost certain is that Technetium _IS_ the new production bottleneck (with the greatest gap between extraction rates and demand), and as such it WILL "suck out" most of the value out of the other moon minerals more and more as stockpiles deplete and only "natural extraction rates" keep the supply going.
Until to the vertical blue line and soon after, this looks very roughly like a possible price evolution, but past 4 months into the future, the projected graph is very unlikely to be even remotely accurate, IMHO.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 07:48:00 - [126]
 

span style= font-size:7pt i Edited by: Vaerah Vahrokha on 21/09/2010 07:53:06 /i /span br @Akita T br br Your fundamental analysis is near impeccable and shows how substantial research and effort went into gathering and assessing multiple information I recall you are collecting since many months i.e. the thread about moons . br br Now, notice how even if I b completely and utterly /b ignored anything T2 related, I d have drawn a plausible trend for the next 3 months in 5 minutes by simple usage of Technical Analysis. br This is quite identical once again! to RL finance, where powerhouses will hire an army of economists and fundamental analysts in order to get a clear picture about the possible future, while many who don t have the ability to do that will resort to more approximate but still viable other means. br br Likewise, your interest breadth as prominent actor in that T2 scene is larger by far than small traders like I could be, exactly like in the scenario right above. br Therefore you go at great lengths in order to get the details, while I just don t care and have the same standard approach for Technetium, isotopes, modules, T3, ships... br br Both of us are affected by what I ll state below though, and you will the one to suffer a larger loss. br br br BLOCKQUOTE font class=quote size=9px face= Verdana img src= /bitmaps/img/board_icons/icon_quote_message.gif border= 0 b Quote: /b hr height=1 noshade br Until to the vertical blue line and soon after, this looks very roughly like a possible price evolution, but past 4 months into the future, the projected graph is very unlikely to be even remotely accurate, IMHO. br hr height=1 noshade /font /BLOCKQUOTE br br This is totally possible and in fact I had to think for a while about whether to bother drawing beyond that line or not. br I decided to do so, in order to give a more visual extension of the head and shoulders and to show thow the further evolution would be tightly tied to previous pivot lines. br br The true evolution past December or January, as I posted some days ago, could be any of the following listed in this post: a href= http://www.eveonline.com/ingameboard.asp?a=topic&threadID=1375714&page=4#113 target= _blank Article /a br br br a href= http://i285.photobucket.com/albums/ll45/vahrokh/EvE/Finance/2010-09-09_HeadsAndShoulders.png target= _blank http://i285.photobucket.com/albums/ll45/vahrokh/EvE/Finance/2010-09-09_HeadsAndShoulders.png /a Picture linkage

http://i285.photobucket.com/albums/ll45/vahrokh/EvE/Finance/2010-09-09_HeadsAndShoulders.png


where scenario 5) would be a possible one.

(Continued...)

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 07:54:00 - [127]
 

Edited by: Vaerah Vahrokha on 21/09/2010 07:58:15
You seem to suggest a straight and uninterrupted trend, possibly without an exhaustion pattern formation but it's not very plausible.
I know there are other threads discussing about "drunken man walk" and similar boutades but markets are not random. They are a sequence of non random fractal human expectations, that get partially "randomized" by external events and conflicting projections.
In case of a strong trend (like we have now at Technetium), these random bits only bump the strong line up, to see a really random-alike motion we have to be in a "flat" market. And even then, flat markets or consolidations are just the "brewing up" phase before the next strong trend.

This blurb means: the current trend retraced twice already and is going too vertical to arrive to 120k pu with no incidents. It's vastly more probable to have one or two dips - even large ones - before getting to those heights.
In fact, if you keep what you said true (i.e. price will be like I drawn till December), the left shoulder (by the fractal nature of markets) would be exactly one of those large dips even if there will never be a descending motion after the head's top.


What I have to underline again is this: being precise, for a Technical Analyst does not matter. Deleverage and cost averaging will usually do well enough for a non major actor to achieve his profit, with the added benefit of having to pour in little research and most of all with the big benefit of being agile.

This becomes expecially useful if what you say for long term will happen and Technetium will become a very narrow bottleneck.
This could cause what for the fundamental analyst hard work is disaster: CCP's economist decided the bottleneck is too narrow and decides for a patch where less Technetium is needed in T2 products (or more is just generated). Then it's pain time.
The technical analyst instead, will just adjust 3-4 lines and will move on.

Notice how both kinds of analysis are useful to someone so there's no real predominance of one on the other, I am just listing how both have good and bad sides.
In my case I extremely appreciate the "slack" approach of TA, as I have little time for the game and much less for elaborate fundamentals.

49473
Jita Trade and Research Institute
Posted - 2010.09.21 08:08:00 - [128]
 

Originally by: Vaerah Vahrokha
You seem to suggest a straight and uninterrupted trend, possibly without an exhaustion pattern formation but it's not possible.
I know there are other threads discussing about "drunken man walk" and similar boutades but markets are not random. They are a sequence of non random fractal human expectations, that get partially "randomized" by external events and conflicting projections.
In case of a strong trend (like we have now at Technetium), these random bits only bump the strong line up, to see a really random-alike motion we have to be in a "flat" market. And even then, flat markets or consolidations are just the "brewing up" phase before the next strong trend.

This blurb means: the current trend retraced twice already and is going too vertical to arrive to 120k pu with no incidents. It's vastly more probable to have one or two dips - even large ones - before getting to those heights.
In fact, if you keep what you said true (i.e. price will be like I drawn till December), the left shoulder (by the fractal nature of markets) would be exactly one of those large dips even if there will never be a descending motion after the head's top.


What I have to underline again is this: being precise, for a Technical Analyst does not matter. Deleverage and cost averaging will usually do well enough for a non major actor to achieve his profit, with the added benefit of having to pour in little research and most of all with the big benefit of being agile.

This becomes expecially useful if what you say for long term will happen and Technetium will become a very narrow bottleneck.
This could cause what for the fundamental analyst hard work is disaster: CCP's economist decided the bottleneck is too narrow and decides for a patch where less Technetium is needed in T2 products (or more is just generated). Then it's pain time.
The technical analyst instead, will just adjust 3-4 lines and will move on.

Notice how both kinds of analysis are useful to someone so there's no real predominance of one on the other, I am just listing how both have good and bad sides.
In my case I extremely appreciate the "slack" approach of TA, as I have little time for the game and much less for elaborate fundamentals.



Just so you are aware my thread about " "drunken man walk" and similar boutades" is not actually intended to demonstrate that markets are random, people might wish to interpret it that way however.



Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 08:26:00 - [129]
 

Originally by: 49473
Just so you are aware my thread about " "drunken man walk" and similar boutades" is not actually intended to demonstrate that markets are random, people might wish to interpret it that way however.





I am aware, otherwise instead of

Quote:

I know there are other threads discussing about "drunken man walk" and similar boutades



I'd have replaced "discussing" with "preaching about"...

49473
Jita Trade and Research Institute
Posted - 2010.09.21 08:57:00 - [130]
 

Originally by: Vaerah Vahrokha
Originally by: 49473
Just so you are aware my thread about " "drunken man walk" and similar boutades" is not actually intended to demonstrate that markets are random, people might wish to interpret it that way however.





I am aware, otherwise instead of

Quote:

I know there are other threads discussing about "drunken man walk" and similar boutades



I'd have replaced "discussing" with "preaching about"...


While it's a divergence from this thread; I'm not quite sure whether you think I am "preaching" or if you are referring to someone else. Please clarify.

As for technetium, I don't very much like it and wouldn't hold too much of it, but there's some profit to made still.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 09:45:00 - [131]
 

Quote:

While it's a divergence from this thread; I'm not quite sure whether you think I am "preaching" or if you are referring to someone else. Please clarify.



Yes, I'd relly prefer to keep this thread "clean".
And yes I am sorry my English is quite bad so it can be confusing. I am referring that in the other thread it's being discussed whether market is pure random or not, with you defending the not true random nature of it vs others who do.
I also believe market is not random, otherwise trading would be pointless.
There are techniques to trade seemingly random consolidation markets but they are barely worthwhile as they bring in small wins followed by a big loss.

If this is still unclear I'd prefer in game mail.


Quote:

As for technetium, I don't very much like it and wouldn't hold too much of it, but there's some profit to made still.



For a "naked trader" (ie little or no indicators except maybe SMAs and fast MACD), Sept 7th would have been the last day for a confirmed and "safe" purchase.
I don't know about one using indicators.

It's possible that another buy signal will come at next retracement but it'd be a typical 3rd retracement buy, that is with a possible increased risk and quite reduced profit target.

clixoras
Posted - 2010.09.21 10:43:00 - [132]
 

An example of RL market manipulation by the FED:
http://www.zerohedge.com/article/fed-injects-record-5-billion-stock-market-todays-pomo


Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 12:46:00 - [133]
 


49473
Jita Trade and Research Institute
Posted - 2010.09.21 20:26:00 - [134]
 

While my own was no better, I find it very difficult to make a comment on the more ambiguous ones without the actual price data; as that's the way I do my analysis.

However I'd urge others to have a try however.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 21:00:00 - [135]
 

Edited by: Vaerah Vahrokha on 21/09/2010 21:01:38
Originally by: 49473
While my own was no better, I find it very difficult to make a comment on the more ambiguous ones without the actual price data; as that's the way I do my analysis.

However I'd urge others to have a try however.


It's because you are used to numerical / statistical analysis, which this thread is not about. This thread is about:

- getting to the point at-a-glance or just more than that

- seeing if it's somehow related to the RL counterpart of this, as per the references at the beginning of the thread.


Basically, the graphs are purposedly imprecise in the finest details so that you can excersise on where it counts: structure. If you get what market structure "likes" to have, then you should immediately see which graph(s) is / are realistic and what ones are not, with the exception of one which I expect no one will guess.

Edit: no one except you, which should be the most qualified to spot it.

Delagos Almondis
Posted - 2010.09.21 22:01:00 - [136]
 

Edited by: Delagos Almondis on 21/09/2010 22:02:56
Number 1:
Number one looks to me like a drawn-out head-and-shoulders-graph, where the second sholder has a lower peak than the first one, which means, if I remember correctly, that we can assume that the trend will be downwards. Smart traders would have sold at the head, and if they like it a bit more risky, bought at the lowest point between head and right shoulder to sell at the peak of the right shoulder (in an ideal world).

Number 2:
This could be another head-and-Shoulders model, but it also has parts of the W-Model (with the bottom point at august). I personally wouldnt invest, but that may be because I am in desperate need of money and cant afford too risky stuff. If you assume this is a Head-and-Shoulders, you could justify investing though, because the right shoulder has a higher peak than the left one (again, only judging from the graph, without knowing what commodity it is)

Numer 3:
We have an early head-and-shoulders in a small scale, afterwards we have an upward trend which is testing the price line of 24k ISK, it seems. In the last week or so, it seems to have broken it, and is now approaching the 25k ISK BRN (Big round number). Again, if you have spare money, you could invest, I would wait until the price drops a bit though, and try to buy stuff when it is low and about to rise to 25k again.

Number 4:
I dont think that it is possible to create such a graph in a healthy market. I can see no patterns and would leave this market alone if I was a "pattern trader".

Number 5:
This graph seems a bit more realistic, but I would not invest in such a strange behaving market. The only pattern I could make out is a head and shoulders between mid-june and august, and it indicates a downward trend, if I read it correctly. Again, no market for me.

Thanks for doing this quiz-like thingy, it was really interesting, and I'm interested in other opinions and the answers of our expert VV Very Happy

Delagos

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.21 22:06:00 - [137]
 

Delagos, I am not asking for you to recognize patterns (but congrats for doing that!), but to tell what of those graph (one or more) is the one (or more) that is realistic and why, or which of those graphs (one or more) is unlikely to be a real market and has been generated a real random.

The excercise is to debunk the idea that markets are easily simulated by randomly generated graphs with no real sense or structure.

Delagos Almondis
Posted - 2010.09.21 22:10:00 - [138]
 

Quote:
Which one(s) are realistic and could be probable markets?

I fail at reading comprehension, it seems. I thought I had read "profitable" markets. Very Happy
Hmm. At the moment, i can only tell that graph 4 looks not really real to me. About the others, I will have to think and write tomorrow, I am falling asleep infront of my PC right now (could explain my reading failure Laughing)

Misha M'Liena
Amarr
21st Eridani Lighthorse
Posted - 2010.09.21 22:26:00 - [139]
 

Edited by: Misha M''Liena on 22/09/2010 00:39:46
Oh boy.

4.. i'll hazard its the one not real

I deleted most as i misread it. But i think the first graph is the real one. But i am not well versed in economics to say what type it is.

I am probaly way off the mark. But i have learned from these and Akita and the lady with numbers as her nameEmbarassed


edit...i screwed up sorry.

Marshiro
Posted - 2010.09.21 23:43:00 - [140]
 

#4: Zomg two delta function in fourier space with higher frequency noise? Laughing I missed my old signal and systems classes. Laughing Obviously totally absurd to see on the market and I'd buy the lottery (as well as smack ccp) if I see it on the chart.

#3: TIME GOES BACKWARDS Laughing On the first rise before April renders the graph absurd.

I don't know what I'm suppose to think, since they are obviously not 5 day/20day moving averages. (due to slope) They could be considered connected dots of median, but not having seen that sort of graph in eve proper, I'm unsure what it is suppose to look like, especially in markets out of jita, where there is no "natural" progression of prices due to trader 0.01 behavior.

#5: This feels too strange as the slope of changes. Most strange is a large place of perfect flatness, which implies zero volume which is extremely unlikely given it is at the middle price. If this is a persistent low volume item (in terms of buyer/seller) then the persistent smooth angles else where is suspect.

----------
This leaves only #2 and #1 as possible, with #2 feeling more possible due to increase volatility over "smoothness" of something you spend 3 minutes in MS paint with a mouse.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.22 07:41:00 - [141]
 

I see some very good answers started tickling in.

I'd love to see other replies before I post the solution(s) after today's downtime.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.22 14:02:00 - [142]
 

Solutions:

The correct answers are 2 and *surprise* 4 is halfway correct and wrong at the same time.


But let's talk about why the others are not first.

Despite the apparent randomness, human made markets have structure and purpose. Humans tend to "slack" and use round numbers when trading and they also tend to condition themselves in self fulfilling prophecies.
I.e. if "everyone knows that retracements happen at Fibonacci numbers" then people will setup buy and sell orders at Fibonacci numbers.

Moreover there are certain almost forced patterns: supply dumping huge amounts of stuff causing self-similar dips, demand getting easily over-hyped causing self similar rallies, getting exhausted as prices rise and thus causing patterns of pull backs and retracements.

Now, the pictures were purposedly schematic to show the juice of the concepts.


Wrong solutions:


Picture 1

There is no realistic human behavior behind this graph.

Price has been purposedly drawn to skip pivot lines, round numbers. There's no realistic retracements, no believable channels + break outs. There's a sort of head and shoulders too bad it's totally asymmetric and the left shoulder is barely formed and too round. There is a too much round top around May 10th and a too much clear cut flat top at the end of the graph.
Market could not live with squared shapes as it means zero activity is being made yet price keeps fixed for almost half month.


Picture 3

Once again, price steers too much clear off pivot lines and round numbers, humans don't do that.
Before the quite irregular head and shoulders there is an huge rally up with no "fight", no retracement (not even a sensible bump) for 15 days.
The price right after the head and shoulders drops with a slow slope that does not resemble panic selling at all, it passes thru 20k (4 '0' digits BRN, therefore very important) without a flinch.
Once at the bottom, there's no visible distribution and the next "valley" does not look like a 'W' at all. The subsequent rallies have too squared and regular tops, it just won't happen. The last dip all to the right is a smooth semi-circle, very rare formation in a market.


Picture 5

This is an example about how a realistic market is not just a market with proper patterns and formations, but also with specific proportions.
In fact the leftmost side is realistic enough but then at May there are visibly unreal huge spikes. In a real market it'd be quite rare to see them, such a strength would quickly result in a powerful up or downtrend.
Consolidations (those clumps of waves) tend to be of quite limited height (measured upper peaks to lower peaks).
Around Jun 1st there's a way too straight horizontal line,
At the same time, notice how once again prices seem to skip round numbers and pivot lines. Expecially noticeable at the right of the graph is to see price happily hopping around 15k with no "battle", no retracements, no bumps, nothing.
After Aug 1st there is a sort of double top but it's way too round in its features.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.22 14:47:00 - [143]
 

Now the "right and wrong" solution:


Picture 4

It's wrong because it's too magnified to exist on a daily graph spanning a year.
It's right because it's an actual copy of a RL market snapshot taken in the second half of Aug 2010 on the CME Futures Market.

It's half right and half wrong because despite being on a RL market, it's a particular kind of trading bot an institution switches on in very low volume days.
Therefore the market is "real" but it's computer generated and it shows. The particular algorythm of this bot is made to maximize whipsawing (one of the worst market conditions where losing is almost sure) and minimize predictability in order to screw off the small "human" traders on the 5 minutes timeframe.
Botting in RL markets is allowed and the institutional bots (unlike the end user jokes sold as "expert advisor" for $97) are extremely effective at taking money off those who don't know to avoid them.
Some softwares (I don't think I can tell commercial names) can even detect the institutional bot type and name in order to help fighting (or avoiding) them.


Finally, the right solution:


Picture 2

I purposedly detailed it to give out an "help". Real markets are fractal and battles are fought and scars litter them.
Notice the small or big retracements in the middle of rallies or dips, Real markets (EvE included) may have very deep and steep spikes and pin bars, expecially during manipulations (in RL institutions or big players manipulate and dump exactly like in EvE).
Notice the "barbs" at the tops, the "breather" pauses in the middle of rallies, the GREAT (for a trader) book example - also presented for Gallente Charters in previous articles here - around May 1st. Big selloff or manipulation with subsequent W bottom (aka double dip, the kind of scare thing the news fling in these days about world economy).
Notice how the head and shoulders neatly hovers a pivot line and is followed by a typical retracement at around Jul 26th.
Notice how price tends to "do stuff" around pivot lines aka price action. In fact traders tend to setup their orders around there.
Notice how trends may be seen, even long term ones (diagonal lines).

For the most hard core technical analyst, I am going to link and show a picture that shows the many relations that may be extracted off Picture 2:


Picture linkage


The blue lines show the sensible amount of relations that may be found with little effort (I am sure there are others beyond what I found).
The white lines show where price action happened at round numbers or pivot lines. Notice how around Jun 1st there is a classic break out centered around the 20k BRN.

Claire Voyant
Posted - 2010.09.22 17:44:00 - [144]
 

Mine ice much?

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.22 18:54:00 - [145]
 

Originally by: Claire Voyant
Mine ice much?


1) It's what I can do most of the day while I work.

2) It's one empty image taken one day filled with different graphs.

Inbrainsane
Posted - 2010.09.25 02:22:00 - [146]
 

Originally by: Vaerah Vahrokha
*graphs*


Your Graphs have so many horizontal lines. But you forgot, that eve online is ruled by VERTICAL lines. And in case you still wonder: The vertical lines mark the patchdays for expansions. They are the points where everything changes, always. So stop makind predictions beyond patchdays based on chart analysis, TIA.

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.25 10:07:00 - [147]
 

Edited by: Vaerah Vahrokha on 25/09/2010 10:08:40
Edited by: Vaerah Vahrokha on 25/09/2010 10:08:16
Originally by: Inbrainsane
Originally by: Vaerah Vahrokha
*graphs*


Your Graphs have so many horizontal lines. But you forgot, that eve online is ruled by VERTICAL lines. And in case you still wonder: The vertical lines mark the patchdays for expansions. They are the points where everything changes, always. So stop makind predictions beyond patchdays based on chart analysis, TIA.


You overlook 3 fundamental factors:

1) Patches are now so far behind while the scope / take profits of the ensuing speculations are blossoming right now of since some weeks ago. I don't know the other investors but I am focused on future profits not on past analysis. Therefore this is a thread focused on similarities between RL and EvE Technical Analysis, not on fundamentals, there are valiant and competent Fundamental Analysts posting in these days and are much more informed than I could be.

2) Vertical lines are also in RL and are created by market makers (the most powerful one in EvE being CCP and their patches) and those are beyond the scope of T.A. TA and price action study are "passive", expecially TA, they come in after facts happened. I refer you to 1) for people who study and anticipate patch effects. I don't even have the money to invest and profit like they do, it's out of my radar.

3) Vertical lines are untradable when they form. No market could go perfectly vertical, market lags behind events, even just a bit of inertia due to human nature. This is akin to "news trading", I don't do that nor arrogate myself the ability to do so with "my" tools. EvE markets don't even have the liquidity for sustained, huge spikes, in fact EvE markets are smooth like silk, I wish RL ones were so easy to look at. It's like looking at decades ago markets here, when some simple trend following strategy was all needed to become rich.


Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.25 21:39:00 - [148]
 

Illiquid / choppy / niche ranging markets


You have probably seen them a lot of times, they are everywhere.

There are markets that just "don't make sense". They don't show a dominating trend since at least the last patch.
They may be specialist items with a tiny market.
They may be items only a small niche is willing / able to trade.
They may be "were good" items that a patch made less lucrative.
Finally, they might just be located in a low volume region.

Welcome to the illiquid / choppy / niche ranging markets.

Now, the first question that comes up is: can these markets be profitable? Are they worth it?

The answer is ... depends.

If you are an high volume / very large turnover trader they are not really where you will find your El Dorado.

But if you happen to have some spare millions sitting unused or you are a beginner or have little time to play and are generally a good candidate for the kind of trading known as "slow selling" you may find there some great opportunities.
These markets, in fact, tend to cycle around some price levels and therefore placing 90 days orders makes it easy to have them being hit sooner or later.
Low risk of "missing" the trend bus: just park orders at "known" levels and the market will sort them out by itself.

The only time when it might be advised to babysit these orders would be when price gets to your order level: the "market residents" will probably set their orders 0.01 ISK below yours and if volume is very low this could lead to your order being missed by the "price wave".


How to spot such a market?

It's fairly easy. Look for items that don't seem to "evolve" in price over a decent amount of time.
Their price should be "boxed" inside a range and horizontal pivot lines should be easy to spot. Price will switch around those lines.


Example:

Let's consider something easy. A "former really good" item, Multifrequency M. In early 2009 it used to sell in decent volume and with some 20%+ value over production price.
Today, after several patches affecting it negatively (minerals, Minmatar buff and much more) this item has lost much of its charme but it may still be used as exercise, expecially by new traders who cannot play with big wallets.

Let's see a 6 months chart about it, taken at Amarr.


Picture linkage


It's possible to spot the key elements: barring few outliers, the price seems to be boxed, price levels are very evident and are also close to RNs (round numbers). Pivot lines may be easily drawn.

Now, with a little of Paint Magic this graph could be analyzed. No complicate math or other stuff needed.

A possible result could be the one shown next:


Picture linkage


What does this modified chart tell us? First of all there is the price range, which I colored blue. A more elaborate analysis could have used Bollinger Bands or even a Keltner / Donchian Channel but that's not really a must.
Inside that box it's possible to see several white horizontal pivot lines. Since there's no short selling in game, less RL strategies than usual may be applied here.
So the aim would be to buy at the low side of the range, i.e. the 21k - 22k lines, wait, "auto-sell" at above 25k after some weeks when price switches to the upper levels.
I said "switches" because despite price being a continuum on the daily scale, it's easy to see how at long term price seems to jump from RN to RN, a telltale sign of a market with a restricted number of "always the same heads" bigger traders / builders (playing MMs here).

Irae Ragwan
Posted - 2010.09.25 22:30:00 - [149]
 

Slightly off-topic, though I have enjoyed the free education:

Related to threadnaughts like this and those posted by other market heavyweights such as Akita T, how much do you wager they influence the markets? I've been casually trading on my alt for years now and kept abreast of things in this very forum. Most every trader I know is at least a lurker here if not a fan of someone. My supposition is, with a large portion of the traders (at minimum those who trade in meaningful volumes) tuned into this forum on a regular basis: do you patently expect to generate market flux with your posts or is that something you see as an unavoidable side-effect of market discussion with real, rather than hypothetical, units?

Vaerah Vahrokha
Minmatar
Vahrokh Consulting
Posted - 2010.09.25 23:12:00 - [150]
 

Originally by: Irae Ragwan
Slightly off-topic, though I have enjoyed the free education:

Related to threadnaughts like this and those posted by other market heavyweights such as Akita T, how much do you wager they influence the markets? I've been casually trading on my alt for years now and kept abreast of things in this very forum. Most every trader I know is at least a lurker here if not a fan of someone. My supposition is, with a large portion of the traders (at minimum those who trade in meaningful volumes) tuned into this forum on a regular basis: do you patently expect to generate market flux with your posts or is that something you see as an unavoidable side-effect of market discussion with real, rather than hypothetical, units?


There are several noticeable differences. Akita T is an older player, performs a lot of fundamental analysis, has the cash to be a market maker and so on. Plus is playing the markets the "old, proven way".
I don't have a reason to get "rich" in EvE as I am a newer player who came too late in this MMO and with my limited available game time I don't have any hope to ever be relevant in "heavy weight" things. But I have done some RL trading and like to do innovative things in MMOs so I am implementing these techniques in EvE.


Pages: first : previous : 1 2 3 4 [5] 6 7 8 9 ... : last (11)

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only