What we're working on in the New Year

Trading with Creation/Redemption Data Corona Del Mar, CA   Howdy !   As we entered 2023, I got my hands on creation/redemption ETF data.   Non-price data is the key to big, long-lasting edges.   I know what you're thinkin': What the heck is that?   Authorized participants like market makers and financial institutions can exchange their stocks for ETFs -- usually in blocks of 50,000 shares.   That makes it a process that only the Big Boys are participating.   They are usually seen as the "Smart Money."   If it wasn't for this process, then your favorite ETFs would be trading out of whack like the Grayscale Bitcoin fund (GBTC).   Anyhoo...our data provider gave us 10 years of data to play with.   Here's the initial result: Well, the in-sample looks pretty good with a 57% annual growth rate.   Buuutttt...a big 'ol J Lo buttt...   The all important out of sample validation isn't lookin' so hot.   As I look closer at the data, there's only 225 trades.   It's just not enough to make this strategy statistically significant.   In other words, I'm not tradin' it with any money.   So do I throw it out like I did AAII sentiment, and index put/call data?   Naw...I went to the data provider and asked "pretty please with sugar on top" to see if we can 20+ years.   I'll let you know the results soon.   You might be wondering...why do we need more weird data?   Using TAP data to build strategies with The Boss SuperAi has been killing it.   The answer is simple: It'll build more non-correlated strategies.   The key to extremely smooth trading results is to trade a bunch of non-correlated strategies.   That way if one zigs, the others zag.   I wasted a handful of years trying to build the perfect trading strategy.   It wasn't until I joined one of the original Turtle Traders in the early 2000's that I finally "got it."   That's why I'm a firm believer that even if you fell for the sh!t coin pump n dumps or the NFT nonsense, there's always hope that you can see the light and become a practitioner of evidence-based Tested Analysis.   Heck, even Jim Simons didn't fully convert until he was 50.   More from me soon... Trade smart,   Dan "Prince of Proof" Murphy                                                 Portfolio Boss, Inc. 260 Newport Center Dr, Suite 100 Newport Beach, CA 92660   About us: Created by our founder, Dan Murphy, Portfolio Boss 1.0 was released in 2014 as an alternative to mainstream back testing software. Although it was slow and clunky at first, it quickly grew because it offered scientific testing on large baskets of stocks and ETFs, including a ranking engine, and used a simple drag and drop strategy builder. No programming required. Also helping matters was the release of the popular report, The Relaxed Investor, with over 500,000 downloads and counting. Inside its pages laid the foundation for rotational investing, proven since 1926. Portfolio Boss quickly morphed into a multi-strategy trading machine with completely new concepts such as strategy stacking, spigoting, and meta-strategic trading. After years of development, Portfolio Boss morphed once again into the first automated A.I strategy builder with massive parallel cloud computing available to the public. With over 3,500 computer cores, "The Boss" was born. Not only can it invent trading strategies using human-made indicators, it also programs itself using C code. Currently, The Boss has been running for over 1900 years of compute time. Starting in 2022, The Boss was updated to include uncommon data such as fund flows and True Asset Pricing to improve strategies by a factor of 18x to 36x over indexing. What's next? Intraday trading and automation for starters. Our goal is to provide hedge fund level A.I tools for everyday kitchen counter traders.   Don't want to stay in the loop with Dan? We'll be sad to see you go, but you can unsub to no longer receive emails.   Government required disclaimer: The results listed herein are based on hypothetical trades. Plainly speaking, these trades were not actually executed. Hypothetical or simulated performance results have certain inherent limitations. Unlike an actual performance record, simulated results do not represent actual trading. Also, since the trades have not actually been executed, the results may have under (or over) compensated for the impact, if any, of certain market factors such as lack of liquidity. You may have done better or worse than the results portrayed.