Hi,
It would be appreciated if someone can help me to make sense of the C2 scores. I know my system is too young and it needs to prove itself, so its score is rightfully low right now.
I had no complaint until I saw another young system: http://collective2.com/cgi-perl/system69201846
It has a C2 score 989, whopping 650+ points more than mine! Based on the statistics, I would not consider it is so much better than mine at this point. I know it is 10 days older, but I would not think my system could jump this much in next 10 days (my score has been stuck around 340 even though the equity curve has been making steadily ramp-up).
Any insights? Thanks a lot. //TQT
P.S.: Sergey, sorry for pointing to your system here. Your system may be a viable one. I just would like to find out why Matthew likes yours so much more.
[LINKSYSTEM_69417480]
Hello,
Very briefly: We give a lot of weight – I venture to say, too much weight – to longevity and to dollar-volume of autotraders. Our feeling with the first is that long track records matter; our feeling with the second is that money talks, and real customers using a system mean more than hypotheticals.
Both systems are new but one of them shows the paid subscribers icon/award.
That said, I agree this often gives too much value to bad systems. This is why we are in the process of re-evaluating the metrics that go into the c2 score, and will hopefully roll out a new algorithm soon.
Alen
help@collective2.com
Alen,
Thank you very much for your reply. I understand the C2 scoring scheme better now. In this case, I think that you are ok with the longevity weighting, but you definitely gave too much weight to the dollar-volume of autotraders. If I have to guess, a reasonable score for the other system may be around 600 (It’s a young system after all!).
I see many broken C2 systems that have a score around 990+. This kind of laughable score greatly undermines the C2 score’s credibility. That being said, I would like you guys to know that I was pleasantly surprised when I came back to the C2 site about two months ago. Congratulations on all the improvements in recent years!
Thanks again. //TQT
[LINKSYSTEM_69417480]
Hi,
Matthew and Alen are in better position to answer your questions. But, it seems to me that the C2 score is for the vendors as a whole. I believe this is another remaining C2 flaw.
For example, if a vendor finally has a viable system after ironing out the bugs (of course, he or she would have had quite a few ugly test systems on records), he will get penalized with a low score. On the other hand, if a vendor with a high score (he or she already has a successful system on C2) is going to try a new strategy, the uninformed readers may get misled by the high C2 score in trusting this unproven new system.
Anyway, the current C2 scoring scheme gives me the following impressions:
1. All young systems are presumed guilty until proved innocent. (Fine with me)
2. All mature systems with subscribers are presumed innocent long after proved guilty (even after they have come to a sudden death with a crash). (I have a big problem with this one)
Whatever the new algorithms Alen was talking about, I hope all the problems mentioned above can be addressed.
Matthew/Alen, any comments?
//TQT
[LINKSYSTEM_69417480]
Hello,
The C2 score is based on overall SYSTEM DEVELOPER for all systems created. It is not based on individual systems (expired or otherwise).
Alen
info@collective2.com
Apples to oranges, my friend.
1 trade versus a 94% win percentage out of 50 and you expect to carry as high a score?
No. Won’t ever happen until you have at least 10 trades, and, even then, I doubt your win percentage will be as high. You have an artificially high APD stat because your first trade had no drawdown.
The 692 system is better because there is more statistical evidence than your 1 trade only system.
Thanks for your inputs. But, I believe Alen was 100% right on the root causes for the low score here, and your comments are irrelevant.
I would like to challenge you to have my system on your watchlist, and come back with new comments once I have 10 trades on record. I am confident that most people who are looking for superb risk adjusted return will be pleasantly surprised.
//TQT
[LINKSYSTEM_69417480]
Sorry, my friend. I should have said:
"I would like to INVITE you to have my system on your watchlist, and come back with new comments once I have 10 trades on record. I am confident that most people who are looking for superb risk adjusted return will be pleasantly surprised." //TQT
[LINKSYSTEM_69417480]
Sadly, we just saw another âheroâ system coming to a sudden death with a spectacular crash. Please see here: http://www.collective2.com/cgi-perl/system57938439
Like many other similar systems we saw before, this kind of âheroâ systems were usually highly leveraged without proper risk management.
Matthew/Alen, the problem I have is to see it is still has a C2 score of 989. This was exactly what I said the systems with subscribers were presumed innocent even after they proved guilty. Your attention to this issue is appreciated. //TQT
[LINKSYSTEM_69417480]
A two cents worth on the thread discussing the C2 ratings…
I think longevity is a very valid thing to look at regarding ratings, especially including perhaps the total number of trades. (generally the more trades, the higher the confidence level IMO)…
However using auto-trading volume as a big factor in the ratings, and if the C2 rating is given for all systems a developer puts out, I question whether those are the best approaches.
Regarding auto-trading volume, isn’t that a chicken-and-egg thing? Perhaps auto-trading volume is low in part because the C2 rating is low…?
And if I have a great system that’s been out for a long time, and therefore I have a high C2 rating, but if I as a developer get a wacky, risky idea, and I put out that system, I don’t see why that system should benefit from my other time-tested and sound system.
Conversely, if I’ve had a poorly performing system out there, and therefore a low C2 rating, if I get a great idea for a new system, and back-test it well, and put it out and it’s doing very well, I don’t see why I’m being punished for the previous system which was a poor one…
I believe every system should be evaluated on it’s own merits (rather than mingling the rating with other systems for that developer), and as long as the system’s stats are very good it should get a good rating regardless of it’s auto-trading volume.
Instead of placing a value on auto-trading volume, I think the thing Collective2 should consider to make sure that a system would actually trade profitably in the real-world (and therefore in auto-trading) would be to take into account the real world fact that it takes 3 DAYS FOR A TRADE TO SETTLE.
I see all kinds of systems out there on Collective2 that are committing 100% of their capital to a 3x ETF going long the market as an example, then 3 hours later they commit 100% of the system’s captial to a 3x ETF going short the market, then early next day they go 100% in some other ETF or stock.
In the real world, YOU ARE NOT ALLOWED TO DO THAT…! (I wish you could, but you can’t)
After selling that first ETF you’re out of the game for a full 3 days, as you wait for that trade to settle and you’re cash to free up for trading again.
So if you wanted to make sure that a system would really be a good money-maker in the real world, and that it’s stats are honest (instead of allowing people to puff-up their stats by doing the kind of rapid in-out trading I just described), the Collective2 code could be changed to do exactly what the brokers are required to do by law, that is, to force a 3 day settlement period.
So if I put all my system holdings in a stock/etf and then sell it, if I try then to immediately place another trade the Collective2 system would say “sorry, you must wait 3 more days until the previous trade settles before having equity to make this trade”, or something similar.
If Collective2 made that change, then it would in fact mean that the statistics for the various systems were REALISTIC, and that therefore the statistics, whether good or bad, are valid in the real world.
I think that would be much more meaningful than allowing users to sign-up for one of these rapid in-out systems, and then finding out in doing auto-trading that they are simply not allowed by their broker to make the trades that the fictional collective2 system is doing.
Why allow something to be done on the Collective2 system, that we know for a fact we know we’re not allowed to do in real trading, by Federal law ??
Aren’t we trying to develop and market systems for successful real world trading, rather than fantasy football type of trading?
Just my thoughts and opinions…
As a newbie single system vendor, it is fine to move in and out because c2 assumes there is margin. Not all accounts are non-marginable and c2 doesn’t make that assumption. For cash accounts it would work that way, but c2 is based on margin trading priveleges. The reg T margin also available at IB for IRA retirement accounts does allow you to do a buy one day and a sell the next and a buy the next. This is also acceptable.
Maybe more making some systems to be non-marginable as an option is what you’re suggesting, but it’s perfectly reasonable to do what you’re talking about if you have an account on margin. You are misleading investors by stating that you can’t do this, because you can. There’s no rules against it. If you didn’t have margin, and didn’t even have reg T margin like at Interactive Brokers that would be the case but any account with margin can do that and it will only take one overnight settlement for you to do that with Reg T margin limited to 1 100% trade allocation per day.
[LINKSYSTEM_68858255]
I stand corrected Financial Scientist, and I apologize to all for making a blanket statement regarding the limitations of cash accounts, that apply less stringently to margin accounts.
Thanks for correcting the errors in my previous post.
All of my trading capital is in IRA accounts, so I’m painfully familiar with the 3 day settlement requirement, as on a regular basis I’ve often been prevented from making a trade, and I wasn’t aware of ‘Reg T’ margin account having greater latitude in that area.
1. I would say however that there are systems that buy 100% of the trading account capital at 9 am, then close that position at 11 am and immediately go 100% long some other position, and maybe do the same again at 2 pm, and I gather you’re saying that the transaction done at 2 pm would not be ‘doable’ even in a ‘Reg T’ margin account? (trading power not freed up until the next morning)…
If so, would you agree it would be reasonable for Collective2 to enforce the same rules, and therefore allow only one 100% allocation trade per day?
I think Collective2 should be trying to create as much of a ‘real world’ environment as possible, and therefore IMO should implement the limitations that exist even in a margin account, regarding buying power.
No reason to make the Collective2 trading environment enable trading that can’t be done in real-life trading.
2. Also, it seem to me anyway that if a strategy is regularly following a trading pattern that is not really practical in a cash account, maybe there should be an asterisk on the system, flagging the strategy as a strategy that it is “strategy best used in margin accounts”, or something similar?
This might actually be a positive for the strategy as it would prevent a user with a cash account from getting frustrated or angry by their inability to be able to take many of the trading signals, because they are waiting for the 3 day settlement period for trade to settle…
They may endup asking for a refund of their subscription fee, send angry e-mails to the developer, and/or may get disillusioned with that developer and maybe Collective2 in general.
However, if the strategy was flagged as a ‘best for margin accounts’ strategy, hopefully that could be avoided, or at least as a developer, I can honestly reply to that angry e-mail if it comes, saying, 'as it’s noted on the site, the strategy is best when used with a margin account"…
I think this might make Collective2 more usable and tradeable for users…
Strategies that are usable in both cash and margin accounts could be flagged in the same way with verbage such as “suitable for both cash and margin brokerage accounts”, or something similar.
There are many people that have a large portion of their stock holdings in IRA or other retirement accounts (like me), and it seems that Collective2 should try to match people with trading systems that are well-suited for that cash account situation, while allowing others with margin accounts to find systems that take advantage of the greater flexibility there…
As both a system provider and, at least for a few months in 2011, a system subscriber, I have a couple of suggestions:
The best way to confirm a system is "real world" tradable is to see if it has a TOS Badge. That one thing confirms, without any question, the system is tradable enough that the provider is willing to put his own money at risk, on his own trading system, in a real brokerage account.
If a system has a TOS Badge, then just send the system provider an e-mail asking what kind of account they are using. Any good system provider will be very happy to answer questions about what ideal account type and ideal equity should be used to trade their system (without getting into any specific recommendations, which might violate regulations).
Overly emphasizing TOS systems does not account for remarkable statistical analyses that you’d pass up simply because you didn’t understand why the vendor is looking for passive income in a hypothetical account.
I’m looking to develop a C2 system, and as long as we can do this, we have capital tied up elsewhere and more productively utilized than the 20+ contract system we’d be assigning to our hypothetical allocations after doing optimizations on all the contracts.
Once you’ve executed your first trade it is up to C2 to do it. You won’t find TOS systems trading that many contracts. You may find the occassional vendor doing trades on 3-4 specific markets, but just because he’s TOS the exclusion of hypothetical modelling will cost too much than simply following a robust system meant for the vendor to collect a passive income.
You got to start somewhere, and having a system provider trade their own system is a pretty good start.
I recall someone posting a comment that they think C2 should implement some sort of graduated subscription fee, based on the number of shares or contracts a subscriber trades, and that other trading system web sites do this. The little detail that was left out was that, in order to post your system on those web sites you have no choice but to trade your own system and have your trades verified by the web site. At least on C2, system providers have a choice.
You can buy and sell in the same day a reg t margin IRA like a cash account up until you’ve done it 3 times in 1 day. You could do a buy and a sell, but not a buy and a sell and a buy in the same day but could do buys and sells for three consecutive days and the fourth time you’d get flagged.
The reg T margin of an IRA allows the account to be used like a cash account, so you can buy but you’re not supposed to sell until a day later and then it’s not considered a day trade. Without reg t margin you can do a buy and a sell in the same day or sell the next day but you can’t get 1 overnight period without reg T margin.
Interactive Brokers knows most traders have the patience for cash account trading, but sometimes get a little ancy and day trade, but until they do that three times as long as they have reg t margin, not full margin or day trading accounts, they can engage in any transaction buy one day and do anything you want the succeeding periods until you sell, then you wait 1 day or one overnight period and you can get back in on reg t margin which is essentially a cash account that may also have margin priveleges but only for the purposes of lending. It is not for daytrading.
I think you’re more concerned with defining account types to have margin or not to have margin, and most vendors cannot do this but it’s certainly a plus for people that know they can be traded in IRA’s as long as they wait at least one day between transactions. They may do up to 3 day trades, get a warning, then penalized if they do those transactions the same day on the fourth trade. The issue I think you’re having is whether C2 should differentiate those systems and I think they should, just as they should also charge per unit per month pricing for the benefit of vendors.
Latest high probability event for me was a nice $200,000 profit in one day.
[LINKSYSTEM_70692142]
Guess the OP got his c2 rating bumped up for his single QQQ trade. It’s still uncomparable to the system he had posted.
Thank you for your interest in QQQ Swings UP. Itâs our policy not to comment on the systemâs action. However, I would suggest you to take a close look at the system description. The system did exactly what it was supposed to do. Please do not rush to judgment. You will find out soon that there are quite some trades within the current single position (The position has been actively managed).
The current C2 score is meaningless. I will contribute my 2 cents on how to improve the scoring system in a separate post, since you guys have hijacked the topic somewhat here (Thanks a lot, Guys! -:)).
By the way, no need to gloat on your single trade. Trading is a marathon, not a sprint.
//TQT
[LINKSYSTEM_69417480]
I decided to “kill” an old test system y’day - it just had a couple of trades in it and was more of an experiment in getting to know the site… Well - my C2 score has plummeted from high 800’s y’day to 293 today!! A bit harsh - I would say!!
>Latest high probability event for me was a nice $200,000 profit in one day.
Fin, I took a look at PTQQS 2.0.B The Aftermath Part II.
Looks like you took one huge position on @NQ. This was the only trade since the system was started 1 week ago.
Looks like you could’ve killed the system and left no trace of its existence if the trade did not work out because it has made less than 5 trades. I am not saying that was your intention, but sure smells fishy. I have a nose for such things.
