Volatility reduced SPY

(If someone has a more accurate name for the strategy, please suggest it!)


I hope this section of the forum is the appropriate place to post this.

This is an incredibly simple strategy you can easily replicate yourself. It has been described by many blog posts, books and so on. The full strategy description has not been approved yet on the c2 strategy page, but here it is:

This is a long only algorithm that only invests in SPY. Its leverage ranges from 0 to 1.

This algorithm rebalances its leverage (from a minimum of 0 to a maximum of 1), weekly, according to these criteria, that you can use to replicate it yourself if you are so inclined:

Leverage starts at 0.
If SMA200 > SMA50, increase leverage by 0.65.
If SMA50 > SMA15, increase leverage by 0.25.
If the return over the last 10 trading days is positive, increase leverage by 0.1.

The hope of this strategy is to decrease drawdowns and volatility when buying and holding SPY, giving up return in exchange. A backtest of this would, for example, lose less money in 2008, but make less money in the bull market we’ve been in the last decade.

As the description explains, it’s easy to replicate yourself by just checking every week those parameters and rebalancing yourself. If you wish to subscribe to the strategy instead, the price is set to $5. It would be free if the subscription to c2 also was :slight_smile:

I’ll leave it to more experienced quants and traders to come up with more sophisticated strategies :smiley:

Have fun trading!

Hello @BoringTrader,
If the strategy (which I have not read about anywhere… post a blog reference perhaps?) truly yields an improvement in risk/reward (e.g., Sharpe ratio?), then perhaps it will be more interesting to the C2 investor crowd if you run it with a 2x or 3x leveraged index fund instead of SPY. If it can actually beat SPY with lower drawdowns, then it may attract interest here. But I’ve yet to see a long-lived strategy here that has long-term subscribers and lower returns than SPY. A $5 price is not the ticket to getting subscribers here at C2, but risk-adjusted returns are. Good luck!

Hi @v1Trader,

I see that you’re an experienced trade leader on c2. Mind if I ask you a couple questions? For starters, would c2 investors be uninterested in the strategy if it does not beat buy and hold (even if it does have lower volatility and drawdowns in its track record)? Regarding the $5 price, it reflects how simple the strategy is, not an optimized “marketing” decision. What would your suggestion be in that regard?

Here are some datapoints from the backtests with a couple of different parameters to give you an idea of the risk/reward. Again, the main improvement is drawdown. Sharpe is actually lower than buy and hold, haha! As you can see, they perform worse in bull markets than buy and hold.

I’ll put both backtests including 2008 and backtests including only the bull market since.

Backtests running from 2006-11-30 to 2019-04-01:
  Using SSO (2x leveraged SPY):
    Max drawdown: -30.31%
    Sharpe: 0.63
    Total return: 216.37%
    CAGR: 9.77%
  Using SPY:
    Max drawdown: -15.53%
    Sharpe: 0.71
    Total return: 116.22%
    CAGR: 6.43%
  Buy and hold:
    Max drawdown: -54.54%
    Sharpe: 0.96
    Total return: 158.96%
    CAGR: 8.01%

Backtests running from 2009-09-01 to 2019-04-01:
  Using SSO (2x leveraged SPY):
    Max drawdown: -29.96%
    Sharpe: 0.83
    Total return: 216.34%
    CAGR: 12.76%
  Using SPY:
    Max drawdown: -15.48%
    Sharpe: 0.88
    Total return: 96.98%
    CAGR: 7.32%
  Buy and hold:
    Max drawdown: -19.66%
    Sharpe: 1.19
    Total return: 231.64%
    CAGR: 13.31%

here’s a blog post on a similar strategy: https://mutualfundobserver.com/discuss/discussion/5580/flack-s-sma-method

the common thread among all of these strategies is that they try to avoid drawdowns simply by getting out when there is some SMA crossing.

There are many variations. This particular variation has simple parameters that I chose and ran with to avoid data snooping problems (I’ve seen articles optimizing the SMA periods over the same time range they use to backtest - yikes!).

IMO on C2, most people like to see higher annual returns than max drawdown.
That is, 20% CAGR w/ 15% max DD, etc. or better. None of these systems fit that.

Lord knows this is darn near impossible in the real world, but you do get some folks that do it here for a period of time.

Just my thoughts.

Thanks for the context! It’s worth noting that these “systems” are objectively worse than any real quantitative algorithm. Best case, if it goes according to backtest, you get lower drawdowns, with ~1 beta, ~0 alpha, and lower CAGR than just buying and holding.

Understandable that no one would be interested in running it… I’ll just use it to play around with the c2 platform :slight_smile:

Ditto, MAR needs to be above 1, Sharpe above 1, recovery from drawdown should be fairly quick (nobody likes to stay down for long, and this is hard without leverage of some sort). @BoringTrader - glad to hear you’re realistic about it. C2 is a hard place. Welcome.

My IRA Truck strategy has had a $zero subscription fee for many months, but it wasn’t until I beat the December market drop (long VXX + SQQQ) and then bought into the current run up and got my MAR above 1 that subscribers started joining. I’ll build my track record a bit more before I start charging a fee.

1 Like