Building algorithmic trading models. So far results continue to be good with every model outperforming the market on both absolute and risk-adjusted basis since going live.
This is an unleveraged, apples to apples comparison. These are not high frequency trading models. Most of them only change signal once every 2-4 weeks on average. During long signals, the models are simply long the S&P 500 and during short signals, they go to cash.
One of the pros of this macro swing-trading/hedging style is high tax efficiency, by holding a core ETF long position that never gets sold and then selling S&P 500 futures (ES or MES) of equal value to the ETFs against the long position. This way your account will accumulate unrealized capital gains indefinitely and you'll only pay tax on the net result of successful hedging. The cherry on top is that the S&P 500 futures are section 1256 contracts that are taxed at 60% long term / 40% short term capital gains rates regardless of the duration they are held.
The models use a variety of indicators, many of them custom built. Most important are various VIX metrics (absolute level, VIX futures curve shape/slope, divergences against S&P 500 price, etc), trend-following TA metrics (MACD, EMV, etc), mean-reversion TA metrics (Bollinger Bands, CMO, etc), macroeconomic (unemployment, housing starts, leading composite), and monetary policy (yield curve inversion, equity risk premium, dot plot, etc). They've been backtested very cautiously to avoid overfitting to the best of my ability.
I've been curious about doing algotrading for both the data engineering aspect and the quant. Do you have suggestions about books or others sources to get inspiration from ?
Is this a one man venture or do you have a group discussing edges ?
For inspiration, I highly recommend "The Man Who Solved the Market" about James Simons and Renaissance Technologies. Some of Ernie Chan's books are great for learning about the basics, but ultimately finding an edge is the most difficult part. Books can teach you some of the best practices for researching edges, how to avoid common pitfalls in backtesting, etc, but no book will ever lay out the details of any strategy that contains alpha of course.
Grizzly Bulls is currently a one man (and wife) venture :)
I've looked before for good online communities but never found one, I'd be interested if anyone has a source. The best people work for hedge funds so wont disclose anything and the individuals out there mostly are clueless or lucky, I suspect its too hard to find an edge without the resources of a large firm.
I expect the long term CAGR for the top models to be in the 20-40% annual range. That's certainly high enough to get wealthy over a couple decades, or sooner if you are already starting with 8 figures, but it's not overnight Roaring Kitty style fast money. Grizzly Bulls' growing revenue helps even out my overall income, and I could definitely see it growing to $10M+ ARR over the long term, very significant even with a 9 figure net worth.
The models are not HFT. Swing-trading the most liquid instrument in the world (ES futures) has extremely high strategy capacity, well into the billions or perhaps 10s of billions, so selling signals does not (currently) in any way negatively impact my own returns.
The alternative would be to start a hedge fund, but that's an expensive and highly regulated endeavor that appeals to a different audience.
could this have any legal risks, e.g. your clients will sue you for manipulations and causing losses to them? Is this a valid business at all from SEC standpoint?
We have the standard disclaimers in the TOS. Essentially, Grizzly Bulls is not a financial advisor, offers no financial advice, and only sells access to signals generated by proprietary models, not the underlying source code for the models. How subscribers choose to use those signals is entirely at their own discretion. There are hundreds of similar businesses out there, and really signals are no different than buy/sell ratings published by more mainstream sites like Seeking Alpha or Morningstar.
All that said, subscribers have generally been happy with Grizzly Bulls' service as evidenced by our low churn rate, especially for the higher tiers.
How would I even go about starting investing using this? Let's say I have a trading212 account or similar? Where do you even start? Do you have a "how to get started page" assuming someone knows little about investing.
https://grizzlybulls.com/how-it-works is the best page I have explaining the basics, but I probably need to be more accommodating to complete beginners. Grizzly Bulls is intended to be a great complement to the buy and hold passive indexing strategy that most people use.
The easiest way to use Grizzly Bulls is to hold VOO in any brokerage account, sell it when the model generates a sell signal, and then rebuy it when the model generates the next buy signal. A slightly more advanced but more tax efficient approach would be to open a margin account with futures trading permissions and sell S&P 500 Futures (ES or MES) of equal value to your VOO during sell signals, then repurchase the contracts you sold during the next buy signal. With this method, I've found you can usually reduce your overall tax burden to less than 15% and you'll only owe taxes on the net result of your futures trading.
Sorry for the total newbie question here -- I'm familiar with options and have traded them a little bit (though that was a long time ago). I've never traded futures before. With your "more advanced" approach, can you help me wrap my head around what the possible outcomes are of buying/selling the futures contracts? What is the impact to me if I, for example, hold $100k of VOO in my brokerage account, sell futures amounting to $100k total on a sell signal, and then I'm "called" (sorry, don't know what the correct term here is) on my futures? Am I wrong in thinking that I'd be required to cough up $100k or my VOO shares?
Good question - you won't be "called" or anything like that in this scenario as you are effectively market neutral. If VOO goes up, your ES/MES futures value will go down accordingly and your account's net liquidation value will remain unchanged and well above your maintenance margin figure.
The only way to really drop below your maintenance margin is if you are either leveraged long (i.e. more than 100% long) or short (i.e. less than 0% long), and the market moves significantly against you. In that scenario, your broker will automatically start liquidating some of your positions.
That makes sense from the perspective of my brokerage/margin account, thanks! I guess I was also curious about the futures themselves; since they’re a contract to buy/sell just like options, would I ever be required to take an action, assuming I’m holding the futures contract at expiration?
Oh no, you are never required to take an action with equity index futures as they are cash settled every quarter. So whether you have an open long or short position at expiration time, it will automatically disappear from your account with your balance left exactly as it should based on the settlement price.
However, this does mean that you'd need to open an equivalent position in the next quarter's contract to maintain your hedge, if one was open, at expiration time which is regular trading hours opening time on the third Friday of expiration month.
Ofc, but as far as I understand algotrading, you don’t need to move a market for other bots to abuse your scheme eventually. Works on real non-toy money since 2022 and works on your napkin simulation since 2022 is a huge difference.
Depending on the market, that threshold can be crazy low. Many smaller brokers make most of their profit by playing against their "customers". Small players can't use the big brokers, and even if they become big players, this just attracts other big players to use their bots against them.
It's a dark forest out there.
Source: I knew several people starting out with a few hundred thousand, "doing really well", and then soon as they crossed some magic number like one million dollars they suddenly started getting front-run on every trade until they gave up. Two of them spent years trying to "fix" their algorithm until they finally figured out what was happening to them.
This basically confirms my concern. It’s known that algo doesn’t live long on sensible amounts. It’s a never-ending churn between bots who don’t even take you personally. It’s as simple as the N+1 guy learning (ml or human ideas) on the whole market that N is a part of and stealing profits from all n <= N by design.
If OP managed to run it live since 2022, I’d be very surprised and glad for them. If not, condolences for the time lost on napkin profitability.
I use Interactive Brokers for automated trade execution and as data source for real-time ES and VIX data. Data for the other indicators comes from a wide variety of sources. One of my favorites is https://fred.stlouisfed.org/.
Not yet, but I plan to write a revised edition incorporating some of the feedback I've gotten in the near future. I'll include a digital version with it.
Yes, in fact non-bull regimes are where they earn most of their relative outperformance. During buy signals, it's impossible for the models to outperform as they are long the S&P 500. During sell signals, they are in cash with the hopes of rebuying lower (doesn't always work out as they are of course imperfect). When the market is rising with low volatility, there aren't as many opportunities for outperformance.
Yes, but that's why the preferred method of implementation is using the S&P 500 futures (ES and MES) as the hedging tool during sell signals. With this method, you hold your preferred ETFs/stocks of choice forever and continue to accumulate unrealized gains indefinitely. Then on sell signals, you sell ES and MES of equivalent value to your long holdings to effectively go market neutral.
At the end of each year, you'll only owe taxes on the net result of your hedging with futures, and futures are section 1256 contracts so they are taxed as 60% long term gains / 40% short term gains regardless of holding period. In practice, I've found that this usually works out to an effective capital gains tax of less than 15% of annual profits. If a strategy returns a gross 30%, then the after-tax return would be about 25.5%.
Also, if you implement in a retirement account which many of our members do, capital gains are irrelevant.
You have a point, but it still makes sense to report before tax performance. Before tax performance depends on model only, but after tax performance depends on model and the user. this website couldn't report that even if it wanted.
Since launching https://grizzlybulls.com in January 2022:
Model | Return | Max drawdown
-------------------
S&P 500 (benchmark) | 21.51% | -27.56%
VIX TA Macro MP Extreme | 64.21% | -16.48%
VIX TA Macro Advanced| 59.13% | -19.12%
VIX TA Advanced | 35.20% | -22.96%
VIX Advanced | 33.39% | -23.93%
VIX Basic | 24.29% | -24.23%
TA - Mean Reversion | 22.30% | -19.92%
TA - Trend | 27.07% | -24.98%
This is an unleveraged, apples to apples comparison. These are not high frequency trading models. Most of them only change signal once every 2-4 weeks on average. During long signals, the models are simply long the S&P 500 and during short signals, they go to cash.
One of the pros of this macro swing-trading/hedging style is high tax efficiency, by holding a core ETF long position that never gets sold and then selling S&P 500 futures (ES or MES) of equal value to the ETFs against the long position. This way your account will accumulate unrealized capital gains indefinitely and you'll only pay tax on the net result of successful hedging. The cherry on top is that the S&P 500 futures are section 1256 contracts that are taxed at 60% long term / 40% short term capital gains rates regardless of the duration they are held.
The models use a variety of indicators, many of them custom built. Most important are various VIX metrics (absolute level, VIX futures curve shape/slope, divergences against S&P 500 price, etc), trend-following TA metrics (MACD, EMV, etc), mean-reversion TA metrics (Bollinger Bands, CMO, etc), macroeconomic (unemployment, housing starts, leading composite), and monetary policy (yield curve inversion, equity risk premium, dot plot, etc). They've been backtested very cautiously to avoid overfitting to the best of my ability.