what's the deal with collective2?

[dtrader98]

Some good comments and feedback here.
I was wondering how accurate the returns reported were. I was also wondering how close the trades were to actual tape execution. Seems like there is some discrepancy from the feedback thus far.


That being said, although we don't have a complete sampling distribution, it is quite useful to look at a sample of the available weekly data. Draw your own conclusions. But keep in mind, this is why I asked if the outliers continued to profit over the long run (i.e. 10 YRS) and how accurate the reporting was. Otherwise, the overall set of systems doesn't look too promising from what I can see. Don't know how long this will stay up, but this is something that I like to see.

Their website advertises to the effect that they contain over 7,000 systems from what I ascertained. Otherwise, the wording needs to change to reflect that.
It is far different from what the previous poster said. Experienced users, please add to the discussion.

 

[Pekelo]

Quote from dtrader98:

Their website advertises to the effect that they contain over 7,000 systems from what I ascertained. Otherwise, the wording needs to change to reflect that.
It is far different from what the previous poster said.

It is a little PR trick on C2's behalf.

It says they MONITOR 7000+ system, because it is true, if a system has been dead for 3 years, C2 still monitors it. 7000 sounds much better than:

We have 200 active profitable systems and 30 of them are actually worthy of subscription!

Now picking a system on C2 is pretty much like picking a stock or a hedgefund, past performances are not guarantees of, bla,bla,bla....

 

[marketsurfer]

our boy tim sykes remains #1 or #2 across the metrics on covestor. very nice!
http://www.covestor.com/rankings/portfolio?perf=risk&q=tgr&tf=si

 

[dtrader98]

Quote from Pekelo:

It is a little PR trick on C2's behalf.

It says they MONITOR 7000+ system, because it is true, if a system has been dead for 3 years, C2 still monitors it. 7000 sounds much better than:

We have 200 active profitable systems and 30 of them are actually worthy of subscription!

Now picking a system on C2 is pretty much like picking a stock or a hedgefund, past performances are not guarantees of, bla,bla,bla....

Therefore the winner/loser list I posted from above is potentially from the historical 7,000 actively monitored systems, no? Meaning that list in itself is irrelevant (in addition to the sample distribution I posted) if what you say is true (and I'm not saying it isn't, as I certainly didn't find 7,000 systems from looking around). Otherwise, how do you know if a system they are touting is even active or the trades were truly executed on some exchange vs. "hypothetical" trades?

I.e. The system we were looking at originally could just be a "hypothetical (paper-trade) system based on 3 yrs, vs. actual results printed on the tape?
If so, that is not very reassuring.

"On C2, you will find over 7,223 trading strategies. You can combine these strategies to create your own hedge fund. Here's how." Come to think of it, I don't see the adjective "active" anywhere.

 

[Trader KGB]

Quote from dtrader98:

And what's to stop some "friend of the company" from entering 1,000+ hypothetical systems, then picking a select group of survivors to display?

Shhhhhhh!

 

[Random.Capital]

Quote from marketsurfer:

our boy tim sykes remains #1 or #2 across the metrics on covestor.

Annualized returns of 3,605%.

Yes, that's a comma, not a decimal.

No, that's not credible.

 

[Pekelo]

Quote from dtrader98:

Therefore the winner/loser list I posted from above is potentially from the historical 7,000 actively monitored systems, no? Meaning that list in itself is irrelevant Otherwise, how do you know if a system they are touting is even active or the trades were truly executed on some exchange vs. "hypothetical" trades?

If so, that is not very reassuring.

1. They are the best stock gainers by % from the last 7 days.
2. Yes, they are irrelevant since you don't know how young (too short track record) they are. Still fun to look at them.
3. They only show active systems on this list because the passive ones are overtaken by the active ones sooner or later. But if a vendor kills a system today, it can still show up on the monthly gainers for a month...
4. If a system is autotraded, then the real, executed trades are shown on the system's page. Otherwise they are hypothetical.
5. As long as the papertrade is realisticly filled by the C2 engine (thus not just touch but trade through) it is pretty close to the real deal...

 

[kahai]

Quote from rolextrader:

C2 has an interesting concept but its design and implementation is seriously flawed in certain areas, particulary the administration of erroneous or inaccurate data it receives from third party trading platforms. When a system developer finds a problem with a trade, C2 will simply go to the information it received from the sender of the data to confirm that what C2 posted is in fact what it received. And that is where its responsibility begins and ends.

If the trade was transmitted inaccurately or way off base, C2 doesn't care as long as it can confirm the information on its servers. So, if your data provider for some reason sends a very bad report, that report will stay on C2 regardless how far from reality it may be. C2 assumes no responsibilty for these types of errors and has no interest in curing the situation regardless who is at fault for the mistake. There is no verification so to speak on C2 other than what a data provider sends to it. That's all C2 verifies--what it does or doesn't receive.

So what does this meant to the developer who is trying to market his system with a so-called "system verification" platform as C2 represent itself? Lets say, for example, he places a trade with his broker via NinjaTrader to BTO at X price and this is accurately reported to C2. Minutes later, he closes out the position at Y price ( a big profit), but Ninja has a data transmission problem or C2's server is on the whack and the STC order never gets to C2.

So, as far as C2 is concerned, the developer has an open order. The developer, upon learning that the order was not transmitted to C2, must now go to C2 order entry platform and close the trade regardless if the current price has no bearing to the actual price he was filled at from Ninja. So, if he takes a big loss on C2, it will be reflected on his system stats even though this trade is complete distortion of what transpired in reality and the system he is marketing is grossly inaccurate in its representation to those who view it on C2. In this theoretical example, a big actual profit will be reported on C2 as a big actual loss. This and this alone renders C2 useless in my opinion.

If C2 were a broker (which its not, although it has an auto-trade mechanism) this trade would be busted and the customer (developer) would get the proper fill. But that is not the case here and it opens a gaping hole in what both developers and subscribers need to make this business model have value.

The concept of C2 is good and may be valuable some time in the future if it can become a realtime, accurate reporter of live trades and work in tandem with it's data providers to insure the accuracy of the information it's presenting to the public regardless what party is at fault if it is inaccurate.

As a subscriber to several systems at C2 I can confirm what rollextrader says. In my case all the system failures I experienced were with systems, both stock and futures, I traded with OptionsXpress so I had to stop using OX. The futures systems I trade with Open E Cry and the stock systems I trade with Interactive Brokers via a third party API (TradeBullet) sofar have been working fine and the results are pretty much the same as reported by C2.

The bad eperience with C2's connectivity to OptionsXpress is even more disappointing to me since OX has an excellent auto-trading software for their roster of signal providers and, believe it or not, among all those quacks and frauds in this field I found two who have done an excellent job over the last two years. Don't ask me for their names, you will have to sift through all the data and performance reports of those signal providers and you will be amazed how much hype and misleading information you will find there. OptionsXpress is actually better in this respect than other brokers offering auto-trading because they only accept Registered Financial Advisors.

 

[BootyShorts]

I had a system on C2 once. Let me tell you one way to game it.

It was a scalping system where the trade would be held til near the end of the day. When you exit a trade near the EOD the reported price was usually way off the actual price. Who knows why. When the price wasn't in my favor I would report the price as incorrect (you just click on a button and tell him what actual price was) and it would be fixed. When it went for me, I obviously didn't say anything.

This would probably only work for intraday scalping systems where pennies matter. It all adds up and can add a lot of percents every month.

 

[fluttrader]

Quote from BootyShorts:

When the price wasn't in my favor I would report the price as incorrect (you just click on a button and tell him what actual price was) and it would be fixed.

Hard to believe that C2 will fix the price everytime you report it as incorrect