San Francisco Frontiers - SkySail Strategies Outperforms Wall Street's 30-Year Risk Standard With Proprietary AI Inference Model

NYSE - LSE

BCC	-0.25%	75.25	$
CMSC	-0.15%	20.42	$
RIO	-2.38%	97.64	$
RBGPF	2.81%	69.99	$
GSK	3.84%	50.05	$
JRI	-0.42%	11.96	$
NGG	-2.53%	74.96	$
BCE	0.34%	23.47	$
BTI	3.58%	57.29	$
CMSD	-0.4%	20.24	$
RELX	5.38%	35.72	$
VOD	0.74%	17.53	$
AZN	2.2%	163.78	$
BP	-0.37%	45.93	$
RYCEF	-2.63%	19.04	$

SkySail Strategies Outperforms Wall Street's 30-Year Risk Standard With Proprietary AI Inference Model

ECONOMY 15.09.2026

SkySail Strategies Outperforms Wall Street's 30-Year Risk Standard With Proprietary AI Inference Model / Photo: © Financewire ( 226626 )

State of the Art by SkySail, the Rubix VM RPT model predicts how far markets will move with less error than the models institutions have relied on since 1993.

Text size:

SkySail Strategies, a quantitative investment firm specializing in applied quantitative research and algorithm development, today announced that its Rubix VM forecasting system predicts market risk with 20 to 40 percent less error than the industry-standard models used since 1993. For trading firms, funds and allocators, it addresses a three-decade problem: the number that governs position size, stops and risk limits has been least reliable exactly when it matters most.

Rubix VM is built around Rubix RPT, a custom AI inference model SkySail developed for one purpose: predicting how far a market can travel in the coming hours. The models it outperforms trace to research that earned the 2003 Nobel Prize in Economics and have anchored institutional risk management since 1993. Thirty years of alternatives have rarely improved on them.

"The industry has measured risk the same way since 1993 because nobody has been able to reliably beat it, we have" said Brian Devens, Managing Director of SkySail Strategies. "Rubix predicts how far a market will travel more accurately than the models the biggest institutions run, on any market with a price history."

In SkySail's testing, Rubix VM outperformed the 1993 standard that still governs institutional risk management, the realized-volatility models that hedge funds and quantitative desks regard as the state of the art, and the range estimates common on trading floors. The gap was widest where it matters most. On the most volatile days in the study, Rubix predicted the size of the market's move with 98 percent accuracy. The realized-volatility models used by quantitative hedge funds were 73 percent accurate. The GARCH models that have set Wall Street's risk limits since 1993 were 49 percent accurate: Rubix was twice as accurate, with one-twentieth the error. See more details at www.SkySailStrategies.com/press/rubix-vm

"That model held for thirty years because it earned it. Beating it was never going to be easy, and that is why it was worth doing," Devens said. "Everything in investing rests on one number: how much can this move against me. Make that number more accurate and every risk decision built on it gets better. Managing risk is what SkySail does best, and Rubix is the sharpest edge we have ever had in doing it."

For trading firms and desks

Proprietary trading firms, options desks and systematic funds depend on a market's expected range to size every position and place every stop. A more accurate prediction means capital neither wasted on moves that never come nor exposed to the ones that do, and over time, stronger risk-adjusted returns.

For allocators and risk managers

For family offices, funds and institutional risk teams, Rubix VM delivers a risk number that holds its accuracy when the market changes character, where conventional models fail. It is the same advantage SkySail runs its own book on, and it has drawn allocator interest as the firm extends the system into new markets.

"It's an exciting time for us. Rubix has been remarkably effective, and as we deploy it across more markets we're seeing real traction from family offices and funds," Devens said. "They understand what a more accurate risk number is worth."

Rubix RPT reflects an approach to AI without close counterpart in financial markets. Rather than adapt a general-purpose model to market data, SkySail built its own inference technology around a single question, how far a market can move. Devens, a systems engineer specializing in probabilistic and stochastic algorithms, founded SkySail on the conviction that this is where superior risk management begins.

Institutions can learn more and contact SkySail at www.SkySailStrategies.com

About SkySail Strategies

SkySail Strategies is a quantitative investment firm specializing in applied quantitative research and algorithm development, with risk management at the center of its practice. The firm develops proprietary forecasting and trading systems, including Rubix VM and its Rubix RPT AI inference model, for global futures and financial markets.