What you want to look at is what the system has now (not what it started with).

In your example, the system in question has $100K. So too does your account. So: if you want your account to trade at the same effective leverage as the system, then you would AutoTrade with 100% scaling.

That means that when the system buys 10 of something, you will buy 10 of something.

Matthew

Hi Matthew,

I think an important caveat has to be added.

There are systems which do not increase the position size as the account grows.

Take for example the excellent system RD30 Day Trading: Even though it presently shows Total System Equity of approximately $239,000 you do not need that amount to trade it the same way with the same risk, all you need is $100,000, the system's starting balance.

BTW, it also way understates the monthly performance figures for the same reason because C2 assumes positions are increased as the account grows. It also gives very misleading charts when comparing this system whith systems which do increase their position sizes.

I wish C2 would find a way to correct this.

Karl

Matthew,
I totally agree with Karl on this one. There are many reasons that a vendor would want to use a fixed portfolio. This allows system vendors to maintain a fixed portfolio of contracts traded and keep starting capitol small allowing subscribers too more easily allocate their funds if they wish to trade multiple systems at one time.

As Karl had previously stated a system vendor is unduly penalized for trading a fixed portfolio with the current rating system. Perhaps vendors trading fixed portfolio's should be able to declare upfront that they are trading a fixed portfolio system and that C2 should develop a rating system that reflects this.

Totally agree. I brought this up before to C2. My system uses a fixed portfolio so it remains the same for all subscribers and can be easily scaled. Of course in comparison to other systems using the max capital possible the performance between the systems is somewhat skewed. RD30 uses fixed trade amounts of 1 or 2. In my system (test mode) I use fixed $ amounts. I can easily make every trade $10k or more so fees and commissions have minimal impact on the trade but most subscribers will scale down to a much lower amount making the system stats look good but the subscriber account will not reflect the same results as the model.

If creating a separate model based on fixed trades is too complicated or time consuming. You know how C2 shows you the result of recent trades during the auto trade wizard as you scale the system. Can you simply add the historical stats for the system as well on the same page based on the scale selected??? This doesn't sound too complicated. It would give the subscribers a more realistic look at the system. Just my opinion.

Thanks, Julio

Dear Matthew,

Could we ask for your response to our concerns?

Thanks,
Karl

Hi, Guys -

Well, it's an interesting idea -- to somehow allow systems that use fixed trade sizes to be differentiated in some way from systems that grow their trade sizes with Model Account equity -- but to be honest it's probably not something we're likely to pursue in the near term.

Let me explain some of the issues that I see:

- If the proposed new feature is merely a cosmetic one (i.e. allow a vendor to put a graphical mark on his system page showing that he uses fixed portfolio sizes, or have this become a metric sortable in The Grid or System Finder) then there's not much real value to it. Any system on C2 can be made to use fixed trade sizing through the AutoTrading control panel (set a high scaling percentage and a low max unit size). Since every system can be AutoTraded as if it uses fixed trade sizes, why do we need one more user-interface dongle for those systems that make that choice on behalf of its subscribers?

- And if the proposed feature is more than just cosmetic -- that is, if it's somehow functional, too (example: a vendor can opt into a program that restricts his trade sizes to X units, no matter what, like Odysseus being tied to his mast) what real advantage is there of doing this? Several people have mentioned that there would be advantages to those systems in terms of performance reporting, and that currently, by not differentiating systems with fixed-sizing, we somehow "penalize" the performance of fixed-size systems. Well, to be frank, it's not C2's job to try to portray systems in the best light possible. That's not what we optimize. And the proposed feature opens a whole regulatory and compliance can of worms. It actually does matter to the NFA and other regulatory bodies whether performance data on C2 assumes reinvestment of profits (which is essentially the current C2 assumption). Would we now have to have two separate sets of disclaimers, two separate sets of warnings, on a system by system basis? To what benefit?

- The proposed feature (at least, as I envision it) would immediately break every reporting and analysis software program on C2. How would it work exactly? To be meaningful in any way, then when a profit was made,wouldn't that profit "disappear" from the Model Account? Okay, well then, time to rewrite every single line of code related to reporting, charting, database queries. And - put aside the fact that it would require rewriting everything, and possible breaking it -- how would the methodology of "disappearing cash" affect AutoTraders? Perhaps there's a way to do it... but, you know, we're at the point where AutoTrading has become really stable, and rolling out new features always introduces new edge conditions that are hard to anticipate and that sometimes have ugly effects.

In summary, all software development risks need to be weighed against potential benefits, and here I see tons of risk: regulatory, trading, broken software, tons of new work ... and not much benefit.

This is the reason why I have been silent in this discussion to date - because it's not currently a project that I think we're likely to tackle. And I often fear that, once I weigh in with the "official C2 point of view," I will perhaps shut down a valuable discussion. I prefer not to do that, because sometimes, after a long enough time, good ideas do appear, even if their first iterations did not seem to me to have been so.

But you did ask for my opinion, and so here it is -- the unvarnished truth.

But please -- don't let this message stop you from posting more ideas about this or other features on C2. Please keep the ideas coming. I am always really delighted and flattered when C2 Members spend their own valuable time to write about proposed new features or to criticize old ones (well, I'm only human, and so I like the first better than the second; but I still value the second). It means you care about C2 and want to make it better. How could I desire anything else?

Matthew

This is an excellent and candid answer, thanks. It would just be good to know which systems scale up trades as they grow and which do not (without sifting through trade by trade with my calculator). Maybe some encouragement to have system developers clearly indicate in their descriptions which path they will be on. No new buttons or functionality required. I'm always surprised how little information most developers provide about their systems, and I always seem to have to send PMs to find out enough info. TY

Thank you Matthew for a well explained inside look to the idea. I must ask about my other idea. I'm not sure if you read it or your explanation included the idea.

You know how C2 shows you the result of recent trades during the auto trade wizard as you scale the system. Can you simply add the historical stats for the system as well on the same page based on the scale selected??? This doesn't sound too complicated. It would give the subscribers a more realistic look at the system. Just my opinion.

I think this idea has great benefits and doesn't sound like it would require a lot of changes. I just don't like the fact that a lot of systems look really good on the surface but once you start trading is a different story and having the stats change based on the trading scale, would help a lot in choosing a winning system to the scale you can afford to trade. I really think you should consider this one, I know is helpful because you already do it on the auto-trade wizard.

Right now the stats are based on the account model assuming most people trade at the same scale as the model. Do they?? I know I don't, specially the successful $100k models. As soon as I download the CSV and copy the data to a my performance spreadsheet, the results are so much different and very unsuccessful at anything less than the 1:1 scale of the system.

Thanks,

Julio

Hi Matthew,

Thank you for responding to the issue.

I did not expect you do revamp the whole software system. The way I see it, the bigest problem I have is when I compare charts of systems which do not reinvest profits to those who do and the first thing I do when I am interested in a system is to compare its chart to my benchmark systems' charts.

Take for example two successful systems which have a reasonable long trading history, system A which reinvests profits and system B which does not. When you compare both on the chart screen it gives a very misleading picture by severely penalizing system B.

Would it not be great if one could click on a second chart button for system B which would bring up a purely hypothetical chart for B which assumes that B also reinvests its profits. Then you would have, at least visually, a valid comparison and as you know a picture is worth a thousand - is it now 'numbers' or 'words'? :)

Karl

I agree. Just different ways to present the data (historical and charts) to potential subscribers. One as you select a different scale and another without re-investing profits. Two great ideas that would benefit subscribers and present a fair picture among all C2 systems!!

Thanks,

Julio

Whether a system "reinvests" profits (which is better termed as Kelly sizing) is part of the system's inherent design. It therefore has nothing to do with C2. They can't possibly amend a system to show another version that uses fractional Kelly sizing or no Kelly sizing because position sizing is itself a major part of a strategy's design!

I do agree, however, that it would be very useful for the system owner to be clear about their intentions in this regard. A simple check box would suffice which can be added in the box that currently shows "Instruments Traded", "Strategy", "System Start Date" and "System Developer".

Suggestions:

"Uses Kelly sizing" <Yes/No> or
"Position sizing is a function of Starting Equity or Current Equity" <Check one>

Discontinued, please see new system