Rescale Metrics and Drawdown

Hi,
we understand the rescale metrics staying mostley the same and that annual profit goes down a bit since the subscription cost is accounted on a lower capital base historically.

The problem is we want to rescale at least twice a year down to starting capital around 100K but then the Drawdown gets higher also. As we see in these two pics. This means we will get a very high drawdown after a couple of years that is not true or fair.

The annual profit we can take but not the Drawdown in a longer term. Do you think this can be resolved in the future @MatthewKlein ??

In other case we must stop the rescaling after it reaches some point perhaps around 25% Drawdown.

in the context of a new message that I just saw about model account sizing limit of USD125000, I would like to ask about the concern that is addressed here in the thread. @MatthewKlein , how are you going to tackle this? else leaders who are here long term will constantly see max drawdown increasing with each rescaling down. I have systems that already are probably out of the subscriber radar due to the max drawdown, and I do not want to increase the max dd further just by scaling down

For reference in case some of you have not seen the news
Collective2

To add, i am totally in favor of your decision (to systems where this can be applied) as long as rescaling does not increase my drawdowns. Also there may be systems (i have one for sure which trades 5 subsystems with 1 nq contract each) which does not fit to rescaling due to the system characterestics

2 Likes

I’m not sure there’s anything to tackle. If a strategy’s Model Account size decreases, then fixed costs such as monthly subscription price will have a slightly larger effect on a percentage basis. That’s just math, and it can’t really be changed.

Forgive me for being blunt when I say this: It’s not C2’s mission to put a gloss on stats so that they appear in the best possible light.

I mean, sure, we could potentially allow strategies to rescale upward to absurdly high numbers. Why not allow Model Account sizes of ten trillion dollars? That would make fixed costs such as monthly subscription fees appear de minimis. It would improve a strategy’s drawdown and return stats, certainly.

Why don’t we allow that? Well, because no customer at C2 has a ten-trillion dollar brokerage account. Presenting stats in that context is not helpful.

Similarly, most C2 customers don’t have $500,000 or $1,000,000 brokerage accounts. So it’s counterproductive to present most Model Accounts in that magnitude. (The folks that do have a $500K or $1M brokerage account are generally sophisticated enough to be able to scale their trading upward, if needed.)

C2’s goal is to present strategies in a way that increases trust in the platform, and that allows subscribers to make sensible decisions. This benefits everyone. It benefits subscribers (who can make more informed trading decisions about how much capital actually needs to be allocated). It benefits C2 (less risk of novices overtrading). And, yes, it benefits strategy developers, too, who hopefully will see increased numbers of subscribers.

Once you accept the basic premise – that it is better for subscribers when Model Account sizes stay in a narrow, comfortable, consistent range – then it necessarily follows that this may mean sometimes some %-based stats may change a little when strategies rescale in order to meet this requirement.

But the stats are still accurate mathematically. Before a strategy rescales, we report percentage-based stats on a base of X; after rescaling, we report percentage-based stats on a base of Y. It was the right thing to do to use X at first, and then later it became more useful to use Y. The numbers are still calculated correctly in each case.

Let me close this message by asking strategy managers here on C2 to try not to fixate on whether your Max Drawdown number is 16.8% or 17.0%. Remember that everyone will be operating under the same constraints – everyone’s %-based stats will change a little as they rescale. What matters is: do subscribers feel comfortable choosing a strategy? I believe this effort to bring Model Account sizes into a consistent narrow range will increase subscriber comfort and will help everyone here on C2. Strategy developers: please try to be patient and keep an open mind, and let’s make C2 work for the subscribers, who ultimately are the reason we are here.

@MatthewKlein everything you wrote make sense except that we should just “take it” as long as the DD decreases “Just a little bit” thats totally wrong. Our strategy Falang will need to rescale several times a year and it would not take a long time to reach 25% DD for example even if the real was 16% at start. That is NOT ok. If i had had a smaller starting account i would have a smaller subscription cost. Why is it so hard (technically or whatever) to scale down the subscription cost in same fashion and same % the account goes down. We will lift at least the Falang strategy away from C2 unless this is solved. may be only a few percent difference at start but not long term. I bet you can change the subscription to rescale the same % as the account and that is the most fair solution.

1 Like

I see your point. OK, I’ll figure out a way to make subscription costs make sense in the context of continual rescales. (But we have a little bit of time to figure it out.)

5 Likes

great to hear, thanks @MatthewKlein and @c2125358467 to making it clear. I think the topic was also addressed here C2 Star Mathematically Impossible on the Long Term - #4 by MatthewKlein, but discussion is from profit perspective, but the same applies to drawdown as well.

You don’t need to rescale down if you don’t want new subscribers in that system , just saying.

of course we want to rescale so we can have new subs, but not by sacrifice the DD metric.