Oct

13

Young, more inexperienced specs would benefit from writing down all the observations, patterns, even subjective things they picked up and compare in the future (1wk, 1mo, etc), not just today but in particular this month. Instead of reading research paper no 3241.

William Huggins writes:

i recall a quarter century ago doc Steenbarger had done some private work for high figure chicago traders on their biases that he could only discuss with the list in the most general sense (ie, this not my work, just shouting back so its value is not lost to time). one of his "summary takeaways" for us that stuck, that i teach in my own classes on investments, is to write down -why- you think something is a good trade for the purpose of auditing your own thinking when it fails. if i recall correctly, the analogy he used was that of a professional sports team watching their own lost games to understand how to contain failure rather than seeking out the (supposed) source of success. the behavioral nugget on that one being the desire to fit our success to a heroic narrative often makes failure analysis more objective and hence productive. if you aim to improve, write down the reasons for your trades and then ruthlessly analyze failures in the aftermath.

Nils Poertner responds:

Wholeheartedly agree William! Victor always suggested to pass on "meals for life" and that would be one of them.

Asindu Drileba adds:

One of my favorite quotes is:

A good decision is still a good decision even if it turns out bad
& a bad decision is still a bad decision even if it turns out good.

A few months back I watched the Michael Jordan movie, Air (2023). A similar quote from the movie I liked:

Just because you're losing, does not mean you placed the wrong bet. - Sonny Vaccaro

Gary Boddicker offers:

Brings to mind the key concept from David Sklansky's classic, The Theory of Poker:

Any time you make a bet with the best of it, where the odds are in your favor, you have earned something whether you actually win or lose the bet. By the same token, when you make a bet with the worst of it, where the odds are not in your favor, you have lost something, whether you actually win or lose the bet.

Oct

12

For those who don’t read the WSJ, this article about Walmart’s automation status is interesting in terms of how long it takes to automate and the savings trade offs (bigger energy bills vs labor costs which may not be reduced though productivity may increase).

For Walmart, Replacing Humans With Robots Is a Multibillion-Dollar Struggle

Oct

11

Consistent with randomness.

Adam Grimes writes:

I agree that it looks consistent with sampling error. Did you run a permutation test? Would prefer that to assumptions of distributions.

"Hey Claude, I would like to write a Python program to take a time series of daily net changes to the S&P 500 and run a simulation in which the daily returns are randomly grouped into 5 bins. Calculate the average and standard deviation of each bin, and repeat the simulation 1000 times."

Steve Ellison offers:

Hey Claude, I would like to write a Python program to take a time series of daily net changes to the S&P 500 and run a simulation in which the daily returns are randomly grouped into 5 bins. Calculate the average and standard deviation of each bin, and repeat the simulation 1000 times.

Output: I believe the last item is the test for significance. Unless the difference between the best day and the worst day is at least 0.50%, the null hypothesis is not refuted.

Data: 188 daily changes
Overall mean = 0.000646 std = 0.008266
Simulations: 1,000 Bins: 5

Bins ranked by mean within each simulation:
avg mean 5th pct 95th pct avg std
Lowest -0.000888 -0.002053 0.000051 0.008245
#2 -0.000011 -0.000864 0.000659 0.008176
#3 0.000660 0.000013 0.001289 0.008179
#4 0.001320 0.000667 0.002063 0.008185
Highest 0.002149 0.001251 0.003346 0.008230

Spread (highest bin mean - lowest bin mean):
average = 0.003037 5th–95th pct = [ 0.001343, 0.005003]

Peter Ringel comments:

Significance - Schmicnificance. Nobody likes Lasagna Mondays, eh? (roll out the pillories)

I did the counting again for 2025,2026 on NDX & NQ EoD data and found: exactly the same lack of significance, as you guys. On intraday I found significance, but also found a bug in my script for the test from how the baseline slices are constructed. EoD and intraday should be nearly the same. None of this is vibe-coded.

The problem is, I already had a relation with Mondays before I did the counting. I shouldn’t even start talking about Thursdays. And I am thinking to start a relation with Wednesday. I really expect to find at least something weak in the stats. I tried to push it a little, e.g. via only looking at up-days. Still nothing.

This isn't over. Mondays are great (for now)

Jeffrey Hirsch writes:

<

Here’s my latest daily on the S&P from page 145 of the 2027 Almanac (click on table for full view). There’s more on pages 143-146 and other data and stats in the data bank section pages 122-183. We also run studies through our internal LLM. 2026 through 10/2 is below. (*Monday is first trading day of the week except when Monday is a holiday. Friday is last trading day of the week except when Friday is a holiday.)

S&P DAILY PERFORMANCE EACH YEAR SINCE 1952

To determine if market trend alters performance of different days of the week, we separated 24 bear years—1953, ’56, ’57, ’60, ’62, ’66, ’69, ’70, ’73, ’74, ’77, ’78, ’81, ’84, ’87, ’90, ’94, 2000, ’01, ’02, ’08, ’11, ’15, ’22, and ’24 from 50 bull market years. While Tuesdays and Thursdays did not vary much between bull and bear years, Mondays and Fridays were sharply affected. There was a swing of 11.1 percentage points in Monday’s and 9.8 in Friday’s performance.

For 2026 we get:

Adam Grimes writes:

I have a set of tools that I developed over many years that start with an event study and move potentially interesting ideas through a simulated trading platform, with options for out-of-sample and forward testing. (None of this originally vibe coded, and the methodology a blend of academic + practical).

I fed it the discussion on weekdays, which was a pretty easy question. I've always been suspicious of seasonal influences because I rarely see them rise above noise, but thought this was worth considering. Attached is a fairly comprehensive result. (I would be especially wary of methodologies that categorize days in retrospect. For example, categorizing a day as "in a bull or bear year" based on a year-end lookback introduces fatal bias.) Results in the pdf:

SP500_weekday_returns.pdf

Larry Williams responds:

Great work but change the exit strategy to make it tradeable. A make the hold a little longer.

Adam Grimes wonders:

Maybe I'm missing something here, but the test shows no edge beyond the market's natural drift (which any weekday entry gets). Won't a longer hold just capture more of that drift? The historical hold that worked was Monday's close to Friday, which was about +0.18% per week over the average day 1953-1977, dropping to +0.02 since 1980.

I would think the key question here is historical stability. Some holds will show an edge over some period of time, but the real question is whether those persist into following periods. Happy to look at this any way you'd find helpful. I can easily run every entry day against every exit day (and even into following weeks if you want longer holds). But I'm reading these results as a conclusive evidence that the old Monday effect is gone since about 1980 and that no weekday stands out from chance. Any random day entry and exit will probably show a profit (due to the drift), but the question I'm judging against is whether any chosen hold makes money above that random hold.

Oct

7

Speaking to Legends: #9 Victor Niederhoffer – The Market Speculator
July 01, 2020

Victor Niederhoffer is one of the most famous investors and a true pioneer. After rubbing shoulders with a few of the future Nobel Laureates at the University of Chicago, he started a trading firm and made $20 million from the starting capital of $50k within 6 months. In the 90s, he was the right hand man of George Soros and with a 34% rate of return over 20 years, in 1996, he was recognised as the world’s best trader. However, next year he was on the wrong side of the trade and lost all of the capital. In this episode, we discuss the emergence of quant investing, how he started trading for Soros, what are some of the lessons he learnt while being an active market participant for 60 years straight and why he think that Jesse Livermore was a horrible trader.

Oct

6

A quote from myself from the 2007 Spec Party on Vic's Central Park Rose Garden talk which I remember vividly. "Chairman espoused his theory of the markets as the mechanism that mediates between government and the the populace…that causes not only the obvious capital allocation but buffers the power of government to protect the needs of the populace through price and capital demands. It is a novel theory which I have never heard discussed anywhere else. The market itself is the mechanism that effects large scale social and political change through changes in incentives, attitudes and demand, and is more than merely shuffling money. Understanding the markets as a social and political driver can lead to long term trading strategies and keep the trader on the right course."

As Bonds hit multidecade lows, I ask (rhetorically) who is more powerful, the market or the government? What does it portend?

James Gavrity writes:

That question raises more questions than it answers. I could answer with a long macro analysis, but the answer to these two sub-questions will give us the answer to the original.

I ask myself: which one is more fearful of the other? And which one tries to exert control and influence over the other? The one that's more fearful and tries to exert the most control is typically the least powerful.

James Carville famously hung a sign in the Little Rock campaign war room during the 1992 election that read: "The economy, stupid."

Who tries to control the price of fuel and eggs? Who takes credit for a booming market? In my view, the market is the most powerful force, whether you live in a democracy or not. Quoting James Carville, "The economy, stupid"—this is what has the power to change governments of all types.

Jeff Watson is unambivalent:

Markets are bigger than governments, hands down.

Larry Williams comments:

Great point. When everyone is a millionaire thanks to the stock market Bernie Sanders pitch falls on deaf ears—just one example of this theory.

William Huggins writes:

while its clear that markets are quite powerful, its important to remember that the sorts of markets we are talking about (capital markets) can only really thrive when there are enforcement mechanisms and dispute resolution - both of which are often supplied by the "force monopolists" of the state. this pattern extends into the bronze age (see: Assur)

imo, states and markets have a symbiotic relationship. either could exist in isolation but would be much poorer and weaker than they are today. in that context, i'm inclined to consider whether one of the two symbiants might try to "devour" the other for its own survival - it would certainly spell the end of a beautiful marriage.

James Gavrity responds:

IMO. As a trader and businessman, I am no expert here; this is just my perspective from where I sit. The market is the boss in that relationship. The Market is the King Maker And Wealth Creator.
The government is the tail attempting to wag the dog, The theory at hand the way I see it doesn't view the market as a quote board; it's the ultimate check on the politicians. The government pushes fiat and mandates, but the market reflects what people are actually doing with their money. When D.C. overspends, overregulates, or debases the dollar, the market acts as a shock absorber. It reprices risk and pulls the leash on the government before ideological policies can completely wipe out the public's economic foundation.

Price discovery doesn't lie. If they print money to fund a political agenda without creating real value, the market violently reprices hard assets and demands a premium to hold Treasuries. It’s not just shuffling cash around; it’s capital voting against political arrogance. It forces a behavioral shift by making it too expensive for the government to keep running its game, which ultimately protects the public from total ruin.

To your question: the market wins, hands down. Central banks can manipulate and suppress rates like the QE we've dealt with for over a decade but that's just building a dam. The market is the water. The government might write the rules, give away free stuff and tax our Income, but the market enforces the economic physics. Eventually, the dam breaks and gravity takes over.

With the long bond hitting multi-decade lows, the market is effectively ripping up the sovereign blank check. If you look at this through the macro cycles, it points to a few massive structural shifts:

The Bond Vigilantes Are Back: The market is stepping in to enforce the fiscal discipline the politicians refuse to legislate. As debt service eats the national budget alive, the government is going to be backed into a corner—austerity, higher taxes, or financial repression—regardless of who is in office. The market is forcing their hand.

Flight to Hard Assets: The era of zero-interest-rate easy money is dead. The mechanism is actively rotating capital out of paper promises and fiat debt, and moving it into commodities, precious metals, and real production. It’s a classic defensive posture against debasement.

Social and Political Friction: Cheap debt has numbed the public to government incompetence for a long time. Turn off that spigot, and the real cost of these policies hits home. When the public actually feels that economic pain, it drives massive shifts in societal attitudes and intense political turnover.

When you watch the tape through this lens, it means looking past the daily noise and short-term setups to see the tectonic plates shifting. When the bond market moves like this, it’s not just another trade. Understanding the market as a social and political driver keeps us on the right side of the trend long-term.

Carder Dimitroff offers:

I generally agree that markets, rather than governments, can be kingmakers. But not all markets are equally independent of the state. Sovereigns establish and enforce the legal order within which markets function, and they control access to territory, airspace, waters, natural resources, strategic infrastructure, and military procurement. Some markets would scarcely exist without that framework; others operate only by license and under close state supervision. When a government judges that national security or another core interest is at stake, firms in those markets must adapt, exit, or risk losing access to the market itself. In that sense, the king may not choose every winner—but it can still decide who remains in the game.

James Gavrity again:

Agreed my thoughts were focused on the 10,000-foot view. Yes they can pick a choose winners and losers. When a government judges that national security or another core interest is at stake, firms in those markets must adapt, exit, or risk losing access to the market itself. In that sense, the government may not choose every winner—but it can still decide who remains in the game. Like Intel and others, from the bigger, broader picture, there is a sequence of order to this. The hubris of the state always brings it to a Marie Antoinette moment: "Let them eat cake." Ultimately, the markets will bring the government and those consensus elites exactly that fate, regardless of how many kings they make along the way. That quote was not her actual words. "Let them eat cake." was a political myth popularized to fit a narrative. Historically her actual last words were an apology to her executioner after accidentally stepping on his foot on the scaffold: "Pardon me, sir. I did not do it on purpose." When the steak and lobster are gone, a new king will be crowned.

Oct

3

Remember trend days? Recently trade has been, I wouldn't say choppy, but with a lot of reversals. The legs are better than trend days in the past, point wise at least. Which led me to study trend days over the years. See how they cluster, and there have been none this year, at least how I defined them. Interesting to say the least. What does it mean? Algos? …that's my theory.

Adam Grimes asks:

What definition are you using?

Zubin Al Genubi answers:

Pick one, any one will do. The moral of the story is the same. Hint: These were up days only.

Adam Grimes responds:

I think the definition matters quite a bit. I would typically work with a definition like:
- opens near the low, closes near the high
- range > X * ATR(20) (starting with values of something like 3 for X)

Key concept is a larger-than-average day with open and close at opposite ends of the day. The reason I asked is I count far more than you do so I wondered if you were looking at "truly exceptional" days. I'll take a look at distribution by year later today.

I don't see any value in thinking in points, either. Saying the legs are longer point-wise when the price level is so much higher doesn't say anything at all, right?

I don't replicate your numbers at all. These are using open and close in the bottom/top X of the range. One table for day's range > 1.5 * ATR, the second for > 2 * ATR. I can't imagine the usefulness of extremely restrictive criteria for trend days, which is why I asked how you were defining them. At any rate, ES futures do not bear out your claim. What am I missing?

[Click to view tables.]

Bullish days: high-low range > 1.5 * ATR(20) table.
Bullish days: high-low range > 2 * ATR(20) table.

And why only look at bullish days? The thesis collapses further when we consider both directions (which I would argue we should do).

Bearish days: high-low range > 1.5 * ATR(20) table
Bearish days: high-low range > 2 * ATR(20) table

Is my math off? (Seriously asking, working with a relatively new and somewhat unvetted system here.) Or what am I missing?

Zubin Al Genubi writes:

Adam, My definition was pretty tight so your math is not off, my def is not too good. I subsequently did down days, which were larger than up days, and fewer of them. I am trying to find conditions that precede trend days. The only thing I am seeing is other trend days, like those 3 big up days 2 months ago ( which my system did not catch) Still playing with it as well.

Adam Grimes offers:

Interesting timing. I'm redoing some old work on the same subject:

A new look at measuring trend days
Catching the wave: ideas for finding trend days
Finding trend days in index futures

Ralph Vince offers:

Isn't it vital as well that the open be nearly the same as the low or the high!

Zubin Al Genubi responds:

There are a lot of interesting ways to define it, and I like your percent of new bar highs one. The range constriction break out is a good precursor. I've found there's not that many trend days so it doesn't seem like something to count on. I have a few more definitions I will be looking at: range efficiency |C-O|/(H-L), Kaufman efficiency ratio: |C-O|/\sum|\Delta C|, Choppiness-type ratio: total path divided by daily range, e.g. \sum|\Delta C|/(H-L). I just used retracement for the post.

I found an average of 28 trading days between trend days, but with wide dispersion over the years. Some years more, some less. That's what led me to the study, my observation of few trend days, lots of Z days this year, with long too straight un-choppy, overly efficient legs, caused by (in my theory) by bots.

Looking at your low vol triggers (for example NR5,NR7) over that last 5 years, I'm finding it persists to the next day. Not what I would have expected. Also, I found 0 instances of 3-consecutive-decreasing inside days (Russian Doll set up) over the last 10yrs!

Adam Grimes writes:

Yes, that no longer surprises me (failure of low vol triggers). Technical analysis literature and marketing has a lot to say about compression as a trigger. I don't think it's any kind of trigger. What's the best predictor of tomorrow's volatility? In models, the answer is today's volatility, by a wide margin. (Consider ARCH and GARCH models.) So low vol tends to lead to further low vol. (Also the observation one of our list members misses when he's concerned about low VIX = impending market crash.)

But if and when you DO get a move out of low vol it tends to be more directional and probably leads to a higher probability of trend days. One issue with all of this work is that good trend days are sparse enough you can't really generate reliable stats.

Zubin Al Genubi observes:

There were only 7 bottom tick/top tick day in the last 15 years and no perfect down days. So the definition needs to be wider. I used 10/90.

Adam Grimes responds:

Yes it has to be something like this. There's also a trend day subset where one part of the day might be a powerful and large trend and the other a range or a trend in the opposite direction. Intraday data can obviously untangle this, but one of the limitations of using daily data is that these types of days will be missed in the count.

Oct

1

AI Discovers DC

There is something strikingly familiar about the sudden enthusiasm of leading artificial intelligence companies for government oversight. The technology may be revolutionary, but the political economy surrounding it is not.

For more than a century, a pattern has recurred: an industry emerges with vigorous competition and rapid innovation. Successful firms establish advantages in capital, expertise, distribution, or market share, and then…some of those incumbents subsequently discover the virtues of regulation.

AI and the ‘Dangers of Electric Lighting’
Thomas Edison warned that alternating current was an unacceptable threat.

Letting Big AI Write the Rules

It seems inevitable that the artificial intelligence frontrunners will get the regulation they desire. That being essentially a fait accompli - unfortunately - the next important question is who will actually write and enforce the rules.

David Lilienfeld comments:

Speculation, but rates have been moving up. Hence, raising capital is more expensive. If they stir up enough concern about safety and everyone has to slow down, they can slow down on their spend and hope to raise money when rates are lower and capital costs less. Seems pretty rational to me.

Back in the 1900-1910 period, the embryonic pharmaceutical industry had to deal with a bunch of epidemics of tetanus and the like because of a lack of care in handling the products. The public stopped buying the products and the industry screamed for some government intervention to regulate the industry with the hope the public would regain its trust. It didn't happen overnight, but eventually, the public did regain that trust and there's been regulation of the industry ever since. Companies like people often act rationally. Sometimes they don't, but often they do.

Archives

Resources & Links

Search