Finders Keepers, Losers Weepers

“It’s not about how much money you make, but how much you keep, and how hard it works for you.” – Robert Kiyosaki

“The first rule is not to lose. The second rule is not to forget the first rule.” – Warren Buffett

Since I returned to C2 after a multi-year hiatus, I have followed in simulation and with real money a fair number of strategies. Some, including a couple of my own, have been steadfast in rewarding the “faith” (for lack of a better word) placed in them, carefully weighing the risk/reward proposition on each trade, letting profits run and cutting losses when appropriate to await the next opportunity. Others–including some of those popular here on C2–have seemed initially to behave in this manner, only to suddenly lose 10% or more of model account assets on a single trade, to my consternation and detriment. Still other strategies seem to reward their developers but punish their followers, who–due to market illiquidity of the assets traded–are unable to come close to matching the developer’s entries and exits.

I have just spent several hours reviewing C2 Member Portfolios, other C2 resources, and my own simulation and subscription history, and would like to give a shout out to the following developers:

First and foremost, to BlackBoulderTrading, developer of Leveraged ETF Trading and Index Futures Trading . The first of these shows up in almost every portfolio (including my own) which has benefited from its steady upward climb. The second has achieved nothing short of spectacular growth, enriching those with a bit more risk tolerance.

Systematic_Trader, of course, cannot go unmentioned. Mischmasch has a multi-year performance record that is unmatched, included a 300+% year in 2021. My choice of this developer’s several strategies, at least for the moment, is ES NQ Day Trades. Why? Because (1) I am more comfortable with the position size limit (1 contract max of each, long or short, per trade), which granularity allows me to better control the scaling (1x, 2x, etc.) of my position; and (2) the closing out of positions at end of day to limit overnight risk (although my experience suggests that–with respect to my own systems–the benefit of holding overnight on average far outweighs the risk).

TraderBen’s BattleFutures certainly deserves mention. Though a newer system (like “Index Futures Trading” above), the upward left-to-right sloping angle and regularity of this system is nothing short of remarkable, and I am glad to be among Ben’s first subscribers.

QuantEdgeTrader’s MNQ Edge Trading strategy has everything I look for in a trading strategy. Quoting the strategy’s description: “Depending on price action, a position is built progressively from 1 MNQ up to the max position size, only when the trade is profitable. Max size long 15 MNQ. Max size short 7 MNQ. Generally, No overnight risk. Rarely overnight exposure max size 2 MNQ. Strict risk management. Always a stop loss for each trade and a max drawdown for each position.” Put these attributes together with a knack for position entry timing and selection, and this strategy seems like a surefire winner (…yeah, I know, there ain’t no sucha thing…but this looks close!).

EdgebridgeCapital’s EDGE - T4, one of several offerings from this developer, is new to me but a strategy about which I’m cautiously optimistic. At this time I’m trading it 2x [edit: just upped to 4x] and have nothing bad to say thus far. Risk seems appropriate to model account size and profits look good; however, I’ll be watching it carefully and will cut and run the instant I discover my optimism to have been misplaced.

Finally, I think I’m justified in including my own two systems, Krakatau and Vesuvius, in this list. (It is, after all, my list…) Both strategies are similar: they enter only long positions in highly liquid ETFs and stocks; they do not trade futures, options or other derivatives; they establish profit targets and stop loss levels on each trade at the time of position entry, they attempt to risk no more than 1.5% of capital on any given trade, and–quite frankly–they are based on a reliably profitable algorithm yielding that oh-so-desirable steadily up-sloping left-to-right return profile. Oh…and like all of the above…they are either TOS [“Trades Own System”] certified or believed to have significant developer capital at risk.

Some might ask why, if my systems are so good, do I bother following those of other developers? Because I’m a risk-averse person, and figure that by spreading my assets across the best-of-the-best, I can achieve significantly above market returns while limiting draw-downs to levels with which I’m comfortable.

Here are a few statistics which I think justify my optimism:

So. Those are my thoughts. Let the stone throwing begin!

2 Likes

No stones to throw here. I subscribe to the first 5 systems you mentioned and your analysis is spot-on. I’ve been watching the other 4 (including yours), but need a longer track record before investing.

I hope this doesn’t count as a stone. It is an interesting list.

Is the success survivorship bias, luck, skill…who knows. I set a reminder to check in 3 years.

I will say I don’t think sharpe, max dd, beta etc. have much value in predicting risk when applied to strategies on C2 which tend to have relatively short investing track records of 5 years or less etc. These data points I think are mostly only useful for large asset classes over long periods of time. The sharpe ratio of bonds versus stocks I think is meaningful. The sharpe ratio of two C2 strategies that have only been around for 5 years isn’t particularly significant and even less so the shorter the period. We have all seen strategies with great sharpe ratios explode with 80% drawdowns after being around for years.

1 Like

In general, I agree. However, as C2 subscribers our choices are pretty limited when restricting ourselves to systems with a track record longer than 5 years. I think the number of trades should be factored into the decision. Take Mischmasch for example – is this system just getting lucky after 1,142 trades? I’m betting not and that the system has a real edge.

Leveraged ETF meanwhile has only 82 completed trades (with many scaling trades in between). I’m more open to the idea that this system is lucky, but I’m willing to bet real money that it is not based on the trades I’ve seen and the daily updates the SM gives to the subscribers. So far my bet has paid off over the past 4 months. I’m not putting all of my chips into this system however, because I’ve seen many once-successful systems fail miserably as you mentioned.

3 Likes

One of the best C2 strategies (see what I did there?) is having a terrible, bad, no good week, retracing 24% of its admittedly stellar lifetime gains in just 4 days, thereby putting underwater everyone who’s bought in since September–some deeply so. Those who bought in before the spectacular August to November ballistic ramp, of course, are still sitting on handsome gains. To me, this apparently uncontrolled atmospheric reentry is indicative of inadequate risk management; to those less risk-averse than I it may be simply be part and parcel of what one should expect when attempting to shoot the moon in pursuit of life-changing gains. Perhaps this strategy will quickly regain orbital escape velocity; perhaps not. Were I a follower, I’d at the very least be inclined to trim my exposure until the answer becomes clear [JMO/YMMV].

Interesting. I had that one on my watch list. I was skeptical, since it is not TOS-certified. It does seem like the vast majority of blow-ups are not TOS, but of course TOS is not necessarily a predictor of future performance (as @InteractiveAssets has pointed out…most recently here: The Number 1 Strategy on Collective2 – How I Invest)

In the post you’ve referenced, @InteractiveAssets asserts that historical profit curves predating the apparent start of auto-trading by its developer should be [edit] regarded with skepticism. Without addressing the merits, or lack thereof, of this claim, I would like to state for the record that it is not always clear from C2’s pretty graphs when auto-trading actually began. For example, every trade recorded in both of my strategies from the moment of inception has in fact been mirrored from actual trades in my personal account. However, from late December until mid-January, I made my strategies “private” while I reworked and further improved upon their already Tradingview algorithms, refined the mechanics of the process by which the TV trades generated wend their webhooky way to my BrokerTransmit IB accounts, and (incidentally) attended to some pressing personal matters. By so doing, I intentionally precluded prospective subscribers from enrolling during that two-week period. Since making both strategies public again a few days ago, I’ve noticed that C2’s pretty graphs now make it appear as though I’ve just started auto-trading them (see below). This is one case in which I beg you please not to believe your lying eyes.


You are free, however, and in fact encouraged to ignore the flat two-week start on Vesuvius’s chart, which reflects time spent driving cross-country for 6 days to attend an out-of-state wedding and to visit seldom-seen relatives, which necessarily caused a start-up delay in my getting a new dedicated IB account set up, funded, webhooked to Tradingview (ask me how if interested), and connected via BrokerTransmit to C2.

Number of trades isn’t nothing but I would lump it in with the other metrics too. Not that great at predicting risk. Rewind just a few days and oneofthebest had over 600 trades and relatively small drawdowns. Fast forward, different story. I know 600 isn’t 1,000 but there are other examples out there of that.

That being said mismatch is one I have considered before myself and they have done phenomenal. But I am very cautious about such high leverage use. In the roughly 2 decades of C2 there have been zero strategies to have daily average leverage of 7 and survive for 1,500 days or more with a good annualized return/max drawdown. I’m not saying they can’t do it. I’m just saying the statistics are not super. But if that fits someone’s risk tolerance I say go for it. I just haven’t been able to convince myself to pull the trigger on it yet.

I do also like that it looks like they have lowered their leverage overtime. So I would imagine they know the risks they were taking and are comfortable deleveraging overtime.

Agreed that high leverage is usually a recipe for disaster. However, just looking at average or max leverage alone does not tell the whole story. One of my favorite C2-reported statistics is this one:

Avg(MAE) / Avg(PL) - Winning trades

In summary, it measures how much drawdown a trade is expected to incur on average in order to achieve a profit. Mischmasch has a very low value of 0.136 for this metric. The system cranks up the leverage when the odds are in its favor, but cuts the losses very early if the trade goes south. This also explains why it only wins 31% of the time, which is typical for trend-following systems such as this. There is also very little overnight exposure, which is a big plus for me.

The systems that should be avoided are ones which the leverage is dialed up during losing trades to try to turn them into winners. If you see systems with sharp equity drops followed by sharp recoveries, this is an obvious sign. The above metric is usually high as well for these systems.

2 Likes

Prospective risk and associated returns (or lack thereof) cannot, as observed by @InteractiveAssets, always be determined by examining the statistical record to date. To illustrate this point, I have prepared the table below:

My cursory way-too-early-in-the-morning study of this table suggests the following:

(1) Sharpe and Sortino Ratios Matter!

(2) if you want to swing for the fences as “safely” as possible, look for a high Calmar ratio in combination with high Sharpe and Sortino ratios, with a low “Hold & Hope” figure offering a beneficial bonus (“Hello, BattleFutures and, to a lesser degree, EDGE - T4!”);

(3) if you’re more risk-averse and looking to grind out singles and doubles, seek out strategies having a low “Hold & Hope” figure in combination with high Sharpe and Sortino ratios (Howdy, ES NQ Day Trades , Krakatau, and Vesuvius!);

(4) “Avg(MAE) / Avg(PL) - Winning trades” is of limited value in risk assessment, but the data does suggest that a figure greater than one, in concert with relatively lower Sharpe and Sortino ratios, may be indicative of an unfavorable risk/reward condition.

My own risk/reward profile is reflected in the subset of strategies I’ve selected from C2’s available universe and the way in which I’ve apportioned my assets among them. I’m looking to field a balanced team, one that can get some runners on base to be driven home by a few long balls over the Green Monster. I think the current roster, with a just-now-added double-shot of “Absolute Safe Returns” is looking pretty good for the coming season.


C2 Graph showing a $40,790 increase in hypothetical portfolio value
with minimal drawdowns during the 3 Nov 2023 to 26 Jan 2024 period.

1 Like

p.s. Gotta add that I ***LOVE LOVE LOVE *** how the GTC stop-loss orders for “Absolute Safe Returns” show up directly in my IB TWS screen!

1 Like

Some of these mentioned strategies are too new (only 3-6 months) - we will see if they can stand the bear market …

2 Likes

Hence the need for constant vigilance.

Is this something unique they do or do all orders put in by strategy leaders show up on followers TWS screens? Thanks, I never knew about this.

My understanding is that if the signals are automatically generated from either BrokerTransmit or PlatformTransmit that the filled orders are simply sent as market orders. If the SM issues the trades manually then they can issue stop loss orders that end up on the subscribers’ side as well. I’m not positive, since I’m a C2 Investor and not a Manager.

2 Likes

I can confirm this at least in regards to BrokerTransmit versus C2 Web Trader. The stop orders are only viewable/placed in their accounts with the Web Trader. With brokertransmit subscribers never see the stop orders as stops. They just show up as mkt orders when the stop gets triggered.

2 Likes