Average advisor cost - page 8

New comment
 
hhohholl >>:

Внесу свои три копейки. Чтоб научиться писать корректные проги, надо потратить кучу времени (собственного) и приложить определенные усилия (при наличии способностей). И это никто не оплачивает. Это автоинвестиция в себя. Быть инвестором без дивидендов - смешно и унизительно. Выцарапать из заказчика внятный алгоритм поведения в рынке - архизадача! От этого зависят все последующие взаимоотношения или их отсутствие. Потом этот алгоритм надо вставить в алгоритм программы, с точностью до микрокопейки, понятно зачем. А вот это уже не механическая трансляция, это чистое творчество. Как говорят разные люди искусства - авторское! И последняя радость, нужно еще прогарантировать оплату своего труда. (Однажды меня прокинул заказчик с оплатой. С тех пор я не расслабляюсь.) Свое время и свой труд - ценю! А за написание дешевых советников из кодебазы не берусь, совесть, знаете ли.

I second that.

Custom software always costs one or two orders of magnitude more than standard software.

Therefore, if a toolkit is pre-written for sale, its price is 50-200 times lower than a custom one.

I.e. if there are similar MT4-solutions for $20 on the net, your "whim" will cost $100-2000.

And if anyone offers it for less - I think they are learning from the customer...

It's a waste of money.

 
The logic is obvious. If someone has assessed the market, delved into the problem. and implemented a quality project in the hope of self-sufficiency when 50-200 users are in demand - the price will take this into account.
For example, if a complex program was bought by 15 million people for $100 - the cost of a similar program is at least $2.2 billion;
It's just a saying. ;)
 

I am curious, are there any ready-made EAs for sale, with a small number of parameters, successfully passed forward-tests on different parts of the history, with "good" profit and low drawdown? How much would such an EA cost?

 

Forward testing, profitability and low drawdown requirements are different for each one.

What are your requirements, Oleg?

And how much would you estimate for an EA with these parameters (not a Pipsar)?

- max equity drawdown on the last 2-3 years history - no more than 5%,

- profitability - about 30-50% per year.

 
Vinin:

So how much does an advisor cost (although an advisor is not the same as an advisor)? You, gentlemen, make up your minds at the end of the day.


I came to the conclusion that the price of a software product will depend on three factors: the amount of work that will have to pull the programmer (read "the number of man-hours"), the average cost of living in the region where the programmer lives and the average cost of living in the region where the customer lives.

Living wage will be a conventional measure, the unit of measurement of the value of labor. For 10 quid in one region, you can buy as much as you can't buy in another. If the cost of living in the region of the programmer is high, and the customer's cost of living is low, then the price the programmer set for the customer will seem too high. And vice versa - if the programmer lives in a region where the cost of living is low, he's likely to either set relatively low prices, which suits the customer who lives in a region with a high cost of living, or he will "mimic" the programmer who lives in a region with the same high cost of living - then the prices are likely to seem reasonable to the customer.

For example, I live in Sevastopol. Webmoney now quote the exchange rate of hryvnia to the dollar = 7.48 - in one dollar 7.48 hryvnia. Let's put it this way. On average, it turns out that:

A loaf of bread here costs 3.3 hryvnias = $0.44

Cap sausage (from the cheapest) costs about 40 UAH = 5.35 USD per kilo.

A pale half-litre beer in a glass costs about the same as a loaf of bread.

Cigarettes Bond = 6,25 grn = 0,84 $

Lousy shoes can be bought for about 90 UAH = 12 quid.

Approximate prices are clear. And how much the same things cost in St. Petersburg, and in Moscow? And in Los Angeles?

I had a client who asked me to give him MQL4 lessons via ICQ and e-mail. He used to pay me $200 for a couple or three days of effort. He was from Russia himself. He went to Los Angeles and got a job there as a waiter. The man could well afford to pay that much for my work - it was money to him, of course, but for me here it was seen almost as a windfall. And he didn't care about that at all. :)

So we can conclude that the cost of an EA is only as much as the programmer and the customer agree on .

Why did I create this forum thread? I just needed to get a reference point and listen to the opinion of others. What is now clear on the pricing of my work, was not very clear then. Thanks to a conversation with the inhabitants of this forum, I clarified for myself what's what.

Many thanks to everyone who took part. And it didn't matter whether the information was positive or negative - both of them narrowed down the search. Once again, thank you all.

 
Mathemat:

Forward testing, profitability and low drawdown requirements are different for each one.

What are your requirements, Oleg?

And how much would you estimate for an EA with these parameters (not a Pipsar)?

- max equity drawdown on the last 2-3 years history - no more than 5%,

- profitability - about 30-50% per year.


A good question. The price of an oligarch's life. Having such an advisor means that money is made only by pushing buttons (with the pushing of buttons being done occasionally). Stability means that it is safe to bring in investors of any size. So a question like "consistently profitable" is a rather absurd question from the realm of fiction.
 
Mathemat:

For each one the requirements for successful forward testing, good profitability and low drawdown are different.

What are your requirements, Oleg?

And how much would you estimate for an EA with these parameters (not a Pipsar)?

- max equity drawdown on the last 2-3 years history - not more than 5%,

- Profitability - about 30-50% per year.

This is a rhetorical question, I hope I will be able to create something of value on my own (some ideas may be borrowed).

Some of my EAs withstand drawdown about 15-20% on 5 year history (200-300 trades) and turn 10 000$ into ~60-100 000$ during this period (backtesting), depending on currency and spreads. So far, the Forward testing has been performed only for one of them (the earliest one) and I have found out that it fits the history very well (exactly the fit, though there are not so many parameters) but the profit is not happy at all (though for the period of 3 months of trading it showed quite good profit and drawdowns are within expected ones, to my surprise after the Forward testing).

I'm trying to put ideas in the next EAs that have more logic than blind fitting - still in the process. This is just a comment on the requirements.

Here I thought that if I found some "proven" Expert Advisor, it would be interesting to study the reasons of its survivability and, perhaps, not to try to re-invent the wheel myself. It is unlikely that I would risk trading someone else's EAs in real life.

 
drknn:

So we can conclude: an EA costs as much as the programmer and the customer agree between them .
You are right, but it is not the price of the Expert Advisor, but the price of the programmer's services for writing it. This price does not depend on the product that we will finally get. If the output will be that hypothetical EA, the parameters of which were indicated by Mathemat, then the price of it should be measured at least in tens or hundreds of thousands of dollars.
 
goldtrader:
If the output will be a hypothetical EA, the parameters of which Mathemat identified, then the price of it should be measured at least in tens or hundreds of thousands of dollars.

According to your logic, for a loss-making EA, the programmer should pay the customer extra? :-))

 

You're full of shit!

I studied MQL just to implement different strategies in code and test it on historical data...

When people approached me with their strategy, if the idea was interesting (and not the same as mine) and I saw a potential profit given my small deposit, I did the expert for free. And if I did not like the idea, but the code was not complicated, then I took 10, 50 or 100 quid... This depended on my mood, financial situation and free time.

I don't consider myself an expert in programming, but I can do my best. Many of my Expert Advisors I created in days, while professionals would write them in dozens of minutes.

At the end a profitable Expert Advisor was created (Pips) .

And if a new idea comes up and I don't manage it, it's easier for me to clearly explain TOR to someone and pay some money and then I myself will check correctness of logic and correct it a bit... Only asked for help a couple of times.

I think many people have a similar story...

12345678910
New comment