13 Comments
User's avatar
Erich Thalacker's avatar

Data looks legit from my limited knowledge of it. Max makes a compelling case in my book.

The ability to have such a tight R value and a 2 week lead to act on the data seems powerful. Would be a good to keep an eye on this trend. Personally, I wouldn’t bet the house on it cause you never know when the trend will break. Also though I am skeptical of current financial markets. I believe there will be another sharp drop to come before we go up firmly again. Just my opinion based on what data I have seen.

Expand full comment
Mark Helfman's avatar

Cool. Do you know how to generate a chart that tracks the data?

Expand full comment
Erich Thalacker's avatar

I guess I should confirm, when you mentioned "data" you were talking about the thread from which this was originally birthed? The data sources I see are all very spread out and my feelings are general sentiment from authors/newsletters (including you!) that I read and trust to do solid research.

Expand full comment
Erich Thalacker's avatar

sure. So you could just bring this Fed reserve data into a spreadsheet (https://fred.stlouisfed.org/data/WALCL.txt) (note that units are millions).

Then subtract the TGA from here (https://fsapps.fiscal.treasury.gov/dts/files/22092600.txt) (find for "TGA" its at the top) (units of millions). Will have to get this like weekly (rather than it being given over course of dates like other 2 datasources), but not a huge deal imo)

Then subtract those numbers from the Reverse Repo here (https://fred.stlouisfed.org/data/RRPONTSYD.txt) (note that units are billions here)

Then it's just a simple chart over time! I bet the author of that thread has a better chart than I can make (cause he has S&P data on it, though I bet that would be easy enough to pull in as well). Easy enough to do with Google sheets :). I could share with you, but I can't commit to keeping it up to date at this moment.

These type of signals are ones that I would classify as "they work until they don't". But, great to add to an analysis of overall market trends!

Expand full comment
Erich Thalacker's avatar

Did some more research on this. Didn’t find the correlation in the way I did. Comparing percentage change of each did not work either... I’m likely just doing something incorrect

Expand full comment
Mark Helfman's avatar

It took me a while to get to this. I'm lost! Maybe you're doing everything correct and Max's analysis was wrong?

Expand full comment
Erich Thalacker's avatar

Could be. Hard to say. I was trying to use excel. Had a friend with some special software. He said he couldn't get it fit correlation either. Neither him nor I do this regularly though. I haven't spent any time on this though since my last comment.

Expand full comment
Mark Helfman's avatar

Understood! Thank you for bringing it up and trying.

Expand full comment
Mark Helfman's avatar

Ok, good. Thank you! I'll give it a shot 🤞

Expand full comment
AKB's avatar

Lot of chatter on this theory. Arthur’s take somewhat entertaining: https://blog.bitmex.com/teach-me-daddy/

Expand full comment
Mark Helfman's avatar

Seems like they're saying the same thing, no?

Expand full comment
Amole's avatar

Not qualified to comment on Max's theory (Lyn Alden, or Cullen Roche, perhaps?) but he has an interesting set of followers.

Expand full comment
Mark Helfman's avatar

Oh? What do you mean by "interesting?"

Expand full comment