Summary

In this episode, Afshin Mashrefi emphasizes approaching financial markets as an engineer rather than as a Wall Street traditionalist: define behavior, measure it rigorously across regimes, and only then ship tools that help humans decide. He explains how TradeWave’s AI infrastructure ingests long‑run market data, normalizes it, and uses pattern‑detection and seasonality models to surface high‑probability periods and setups, while still keeping outputs deterministic and reproducible so traders can audit why a given pattern appears. Throughout the dialogue, both he and Anastassia stress that “humans still set the pattern, AI helps size the odds”: algorithms can rank historical tendencies and flag anomalies, but risk, position sizing, and the decision to act remain squarely with the human trader and leader.

Afshin Moshrefi’s thinking around the 100‑Year Pattern begins with a simple question: are widely repeated “seasonal” stories about markets backed by real data, or are they just trader folklore? To answer it, he spent years building his own data engine, running nearly a century of S&P 500 history—close to a trillion data points—through systematic seasonality tests and election‑cycle analysis, eventually isolating a long‑horizon pattern that explains a disproportionate share of index returns. In his framing, the 100‑Year Pattern is less a magic formula and more a probabilistic map: a recurring October‑to‑September window within the four‑year presidential cycle that historically outperformed simple buy‑and‑hold by a large factor, plus a family of related seasonal setups traders can use as scaffolding for decisions.

Afshin Moshrefi is a data scientist, inventor, and entrepreneur focused on applying AI and pattern recognition to financial markets. He is the founder and CTO of TradeWave AI. This platform turns long‑term market seasonality and trillions of historical data points into tradable “patterns” for options and equities traders. Trained as an engineer (M.S. in Electrical Engineering with a focus on statistical physics), he has spent decades building AI‑driven products across telecom, healthcare, IoT, and trading. He holds multiple patents in proactive user interfaces, media recommendations, and crowdsourced problem-solving.

He is also the author of The 100‑Year Pattern, an Amazon #1 new release that argues long‑horizon market seasonality can be quantified and used as a probabilistic “map” for traders, with AI helping to surface and size those recurring patterns.

Key Takeaways:

Investors can achieve significant success by applying the seasonality hypothesis to their trading.

A solid scientific foundation is key before deploying an AI and automation platform.

Chapters:

1:22 Afshin's path into AI-powered trading

3:46 Physics as a statistically driven discipline; AI roots in statistics

5:21 Randomness of technical financial analysis vs fundamental analysis of companies' health, and why excellent outcomes can't exist without fundamental analysis

9:27 What constitutes the 100-year pattern?

12:12 Discovery of the pattern

14:40 TradeWave as a research platform

18:17 AI vs. data science

20:18 History of working with seasonality

24:05 Scientific concept came first, building the platform happened second

27:38 Overreliance on AI in trading and trading automation: pitfalls and issues

30:14 AI agents as a team of capable workers

Hyperlinks:

Afshin Moshrefi LinkedIn

Twitter/X

Website TradeWave

56: Signals and Sensemaking: Human Edge in AI-Powered Trading with Afshin Moshrefi