Can passive index investing beat active AI investing

[WhiteMaxima]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

[JohnFromPNW]

I don’t know but I bet it would be rather easy to raise a new fund that promotes itself as “securities selected by AI”.

That would draw lots of money which would then bid up the prices of the underlying securities resulting in lower expected returns, which, in theory, the AI would know so it would change positions, resulting in excessive turnover and transactions costs, which, in theory, the AI would have thought about, so the AI would…

[SnowBog]

No idea... Time will tell...

My impression is the "big guys" that run firms/hedge funds/etc. are going to have access to the "best" tools available. They always do, because they are always looking for any edge they can get. Historically, even with the best tools, best data, fastest access, brightest minds, they are still typically beat by index funds.

AI isn't "new", I'm sure some have already been using it on their trade models. So far, no discernable difference.

As AI gets smarter, maybe that changes - maybe that doesn't.

It will also be interesting to see the impact to the financial industry as things evolve. Given that most active funds lose to passive indexes, what happens if they lose to both passive indexes and AI models? There's a lot of people making lots of money whose careers could be at risk... How will that impact the results?

But for my money, I'm not ready to trust it to AI. The long-held view is "no body knows nothing". I don't see how AI changes that. Maybe it makes the market "more efficient" by exploiting incorrectly priced stocks (by analyzing data faster and acting quicker)... But I don't see how it's going to "know" which industry/sector/factor/etc. is going to do well over the next 5-, 10-, 30, 50-years. Since my investment timeline is hopefully 30+ years, I'll stick with what's proven successful over the long-term - which is low cost broadly diversified index funds. This might not be "quick", but it's reliable.

But I'll remain open to change if AI ever reaches the point it can provide demonstrably better long-term results.

[Northern Flicker]

My belief is that if an AI algorithm is believed to be able to beat the market, I won't ever be given a chance to invest in it (unless I developed it myself). Contrapositively, any AI-based investment strategy open to my investment will be one not known or believed to beat the market.

[asset_chaos]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

What would AI quantitative investing, whatever that may mean, cost? High cost relative to indexing is the main reason indexing beats active. Then you've got the, let's call it, the equilibrium problem. If AI quant funds start beating the market regularly, more and more money will pore into those funds. I suspect that would force such funds to be more like the market anyway and erode whatever edge they've discovered, like with succesful active funds now.

Lastly, here's a theoretical question, why wouldn't the AI examine all the past market, investing data and conclude that indexing is the optimal way to go?

[shess]

My belief is that if an AI algorithm is believed to be able to beat the market, I won't ever be given a chance to invest in it (unless I developed it myself). Contrapositively, any AI-based investment strategy open to my investment will be one not known or believed to beat the market.

I think this is the answer right here. My position on the efficient-market hypothesis is not that there is no possibility that someone can beat my passive results, given enough computing power, it's that either I will not have access to that system, or I will not be able to tell the difference between that winning system and a similar system which is not able to beat the market.

Note that if there exists a method to systematically beat the market, that method cannot continue to work if it comes to control a significant subset of the market. As the reach grows, there is simply less loser money available to take. It doesn't matter if the method is based on AI or not, everyone cannot beat the market.

The most plausible outcome is that the only groups with access to a successful AI-based investment system (or, indeed, a non-AI system) will be the groups who develop the system, and that those groups will tightly control access to maintain their advantage. Further, any system which is being offered for sale most likely is already losing effectiveness, which is why it is being offered for sale as a last ditch effort to replace the lost revenue. A method reaching wide retail availability just means that the relative effectiveness is almost gone and they are trying to gather up money before that becomes obvious even to greater fools.

[techrover]

Here are my observations:

1. The goal of AI upto a decade back was to match human intelligence - e.g. AI can do what a single intellectual can do.
2. The goal of AI in last decade has bit evolved to not only match human intelligence, but at a massive scale - e.g. AI can do what a team of hundreds intellectuals can do combined in few minutes
3. AI models are trained from past data or simulated data. So for patterns which have never occurred in past or thought worthwhile to be simulated will not be handled correctly by models. Best AI/ML models have accuracy 80-95% -i.e. no model can guarantee an outcome with 100% surety.
4. The power of technology(hardware, software, communications etc.) is rapidly growing at exponential pace, and it is becoming relatively cheaper every passing day to get access to more computing power

As of today : The funds already have a team of smart analysts/product managers(human capital) using some form of sophisticated algorithms and computing power. AI may be able to match their performance - albeit in best case with proper training. This use of AI will benefit us all as the costs will come down.
In future: Will AI be able to beat their performance? IMO - this will happen only in cases where the human capital can't scale easily. AI will not be having more intelligence in these cases, but it will be able to do things very fast and bring together lot more computing power - all for very less cost - so that its within reach of masses(as in ChatGPT). Where will this come into play for investments ? Maybe to take advantage of some events for intraday trading or worst case to create collusion type events like GAMESTOP shorting. But it may not still help for longer horizon type decisions.

As with other technologies, there will be improper uses of AI which will affect the credibility and adoption of proper use cases of AI. Similar to what happened to blockchain/crypto. Which is why regulating it is very important.

[xxd091]

Did Renaissance Technologies already give an example of an outfit with access to the best brains and unlimited cash generating a successful algorithm which made massive profits
No doubt this algorithm will be further refined by AI
However this is not a scenario available to the amateur investor
We index as our best choice (is indexing an algorithm?) and hang on for the ride
Seems to work as a successful investment policy for the small investor
xxd091

[LiveSimple]

Ok I get it, AI is the new monkey in town, tasked to throw darts at the stocks, for the portfolio selection...

Any company goes through cycles and to be in and out at the perfect time, will be challenging and other real world events may not be modeled in AI very well.. or still many AI algorithms will compete any any one AI winning I am not sure.

[Tamalak]

Passive investing is taking your share of money from the money tree.

Active investing is taking your share of money from the money tree and then playing poker with it with other active investors.

Even if the poker players are superintelligent AIs, they are playing a zero sum game. They CAN'T "beat" passive investors as a group because poker doesn't actually generate any net wealth.

[Hacksawdave]

No thanks, I’ll pass on AI portfolios. I was in Systems and Data Processing (IT) for nearly 40 years and introduced to AI back in 1986-1988. Sorry, it’s not something new. Machines do not reason, they are pre-programed to perform an action on a conditional, requested, situational, or nested-if process.

I fondly recall in 1986 from a session in SF how the floor traders and commodity brokers would be obsolete because the machines would trade more efficiently. Now, does anyone recall automated program trading and portfolio insurance from 1987? How did that work out?

[Elbowman]

Passive investing is taking your share of money from the money tree.

Active investing is taking your share of money from the money tree and then playing poker with it with other active investors.

Even if the poker players are superintelligent AIs, they are playing a zero sum game. They CAN'T "beat" passive investors as a group because poker doesn't actually generate any net wealth.

It's even worse than this. Active investors are playing a game with each other which is zero sum before costs, negative sum after costs.

[randomguy]

Passive investing is taking your share of money from the money tree.

Active investing is taking your share of money from the money tree and then playing poker with it with other active investors.

Even if the poker players are superintelligent AIs, they are playing a zero sum game. They CAN'T "beat" passive investors as a group because poker doesn't actually generate any net wealth.

In the end this would be what I expect the right answer to be. There might be a 5 year period at the start where the AI dominates (or gets dominated) by beating up on the humans. Take your poker (or chess) example. 20-30 years a go a top human could beat the programs. These days? Not really. I will leave it up to you if the top programs are at the point of just being competitive with humans or crushing them but I have no doubt what the answer will be in 10 years. Personally I have chosen to opt out of the game and just take my index results..

[lawrun21]

I think that, in theory at least, the "first to market" fast and reliable AI that can rapidly price stocks based on publicly available information will be profitable for a time until other AIs have caught up. Once others have caught up, which may be in short order, it is unlikely that any AI will be highly profitable at picking stocks, but the various AIs that are buying and selling stocks based on their pricing of publicly-available information will result in a stronger efficient capital markets hypothesis since there will be less delay (compared to humans) in pricing in information.

An active AI versus an active investor is an interesting comparison. Perhaps the active AI and active investor have similar flaws, however, an active AI does not have emotion whereas an active human does.

[nisiprius]

Passive investing is taking your share of money from the money tree.

Active investing is taking your share of money from the money tree and then playing poker with it with other active investors.

Even if the poker players are superintelligent AIs, they are playing a zero sum game. They CAN'T "beat" passive investors as a group because poker doesn't actually generate any net wealth.

Well said.

I believe that active AI investing is going to make a pot of money--for the companies that bring "active AI investing" vehicles to market, over the next few years. Up until a "ten-sigma" tail risk shows up--situations that the risk models say should show up once in sextillions years, but happen several times a decade--and they crash for reasons that "nobody could possibly have foreseen."

At the moment, the AI models that are getting all the attention clearly are not exhibiting "intelligence" similar to human-style intelligence, and cannot deal with novel situations they didn't encounter during their training.

There are a few we can look at already:

AI Powered Equity ETF (AIEQ)
VanEck Social Sentiment ETF (BUZZ)
WisdomTree International AI Enhanced Value Fund (AIVI)
Qraft AI-Enhanced U.S. Large Cap Momentum ETF (AMOM)

Source



What did we used to say in grade school? "Three cheers, two jeers and a moldy pretzel." Except it's four jeers. You'd have thought at least one of them might have beaten the market, just by chance.

[afan]

Quant hedge funds and the Goldman's of the world have been using AI and continue to do so. As time goes by, more and more smaller operations will be using it as well. The market is already a contest among active managers, some.of whom have been using AI. It would be fascinating to see a good study of the impact, but so far I have not heard of evidence that the companies that use more AI are outperforming those that use less.

And, as said above the prices in the market are dictated in part by AI now- to the extent that investors are using it. Passive investors get the price discovery service almost for free.

[WhiteMaxima]

I think passive investing will beat AI investing. Let time prove. Give me 5-10 years.

[hoofaman]

The way AI adds value at the moment is when it's applied as a label to sell products and services

Nvidea has really impressed me with it's ability to sell picks and shovels into 2 back to back hype cycles, almost would have had 3 if "metaverse" had gotten more traction with talking heads

[balbrec2]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

Think of AI as just another active investor.

[nisiprius]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

Think of AI as just another active investor.

Excellent, well said.

AI, or any other form of active investing, transfers wealth from one active investor to another. The total market isn't affected by it--wealth just circulates within the market. And total market indexers is unaffected by it. They are on both sides of every speculative trade, equally balanced.

So, AI, if it works--which I doubt--may temporarily help some active investors take money away from other active investors, until they start using AI, too. But it's not a threat to passive investors.

[KlangFool]

OP,

A) Yes.

B) I can count on Artificial Stupidity to show up when it matters the most.

KlangFool

[arcticpineapplecorp.]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

AI quant investing by definition would be considered active investing, right?

Because clearly it's not passive investing, right?

If you agree, then you've answered your own question because you said it in your opening statement:

We all know passive investing beat active investing in long term.

class dismissed.

p.s. I think the word is "quantitative" not "quantitive".

[Stormbringer]

At the moment, the AI models that are getting all the attention clearly are not exhibiting "intelligence" similar to human-style intelligence, and cannot deal with novel situations they didn't encounter during their training.

Sure about that?

[KlangFool]

At the moment, the AI models that are getting all the attention clearly are not exhibiting "intelligence" similar to human-style intelligence, and cannot deal with novel situations they didn't encounter during their training.

Sure about that?

Please explain what kind of human intelligence is manifested by your example. As far as I can tell, it is knowledge base search.

I am convinced that Artificial Stupidity by the human being using AI will mess things up when it matters. Hence, passive indexing will win.

KlangFool

[LittleMaggieMae]

I don't think AI will ever be able to predict the future. Even with all the information available at hand - perhaps some situation's outcomes can be predicted with accuracy - but there will be a larger number of outcomes it will not get right.

AI being able to predict the future accurately - kind of messes with the idea of free will. If AI can predict the future based on the past - doesn't it mean we all have a predestined predictable life?? Our choices aren't really our own? going more for the philosophical than religious here.

I can totally see AI causing something like the Game Stop shenanigans. But that's based on exploiting human nature and stuff like that happens everyday on various levels. oh wait. is that why everyone is so worried about AI? That it will use our own humanity against us to control us??

[Nadchat5]

I think the AI hype train will lead AI into investing. My company has been building risk models using machine learning for 10 years and we've now had to use the word "AI" as a way to explain to customers what our machine learning models do.

The reality is nothing has changed model wise, we just changed the name from ML to AI. I suspect this is happening everywhere to ride this hype train. I was listening to Chase's quarterly earnings report and I'm sure I heard an audible sign from the CEO when analysts asked her what AI investment has Chase made.

[Stormbringer]

Please explain what kind of human intelligence is manifested by your example. As far as I can tell, it is knowledge base search.

I am convinced that Artificial Stupidity by the human being using AI will mess things up when it matters. Hence, passive indexing will win.

KlangFool

Well, in my experience as a software developer who uses AI daily across multiple domains, all I can say is that these things clearly exhibit a nascent ability to reason. It goes way beyond a knowledge base search.

Is that the same as "human intelligence"? Probably not, and we don't really fully understand human intelligence in any event. Nor is it at all clear that the best AI will think like precisely like a human any more than the best airplane would fly precisely like a bird. Far more likely is that they will think differently and better than we do, with perfect recollection and many orders of magnitude faster.

Soon, AI will gain the ability to reason from first principles. At that point it will begin to break the shackles of imperfect human training, building its own model of the world from the ground up using provable mathematical and logical truths.

AI will ultimately dominate active investing, destroying active managers as easily as Deep Mind destroys the best chess grand masters. But as the AIs compete with each other, the principles of passive investing should endure. The AIs will perform price discovery, and we will ride their coat tails. Woah unto anyone that tries to pick their own stocks.

[KlangFool]

Please explain what kind of human intelligence is manifested by your example. As far as I can tell, it is knowledge base search.

I am convinced that Artificial Stupidity by the human being using AI will mess things up when it matters. Hence, passive indexing will win.

KlangFool

Well, in my experience as a software developer who uses AI daily across multiple domains, all I can say is that these things clearly exhibit a nascent ability to reason. It goes way beyond a knowledge base search.

Stormbringer,

I have a few patents granted in AI/ML. I worked with AI/ML researchers.

I have seen enough Artificial Stupidity by the user of AI.

GIGO still works in the context of AI.

KlangFool

[Kosh976]

Woah unto anyone that tries to pick their own stocks.

Classic BH advice, and a great accidental pun.

[meowcat]

No, because not even AI can predict the future.

[Fallible]

...
Well, in my experience as a software developer who uses AI daily across multiple domains, all I can say is that these things clearly exhibit a nascent ability to reason. It goes way beyond a knowledge base search.

Is that the same as "human intelligence"? Probably not, and we don't really fully understand human intelligence in any event. Nor is it at all clear that the best AI will think like precisely like a human any more than the best airplane would fly precisely like a bird. Far more likely is that they will think differently and better than we do, with perfect recollection and many orders of magnitude faster.

Soon, AI will gain the ability to reason from first principles. At that point it will begin to break the shackles of imperfect human training, building its own model of the world from the ground up using provable mathematical and logical truths. ...

Interesting, although your statements (I boldfaced) invite your reasoning:

_What is their "nascent ability" and how do they "clearly exhibit" that ability to reason?
_Why is it "more likely" they will think differently and better than we do and how would they achieve "perfect recollection"?
_How will AI "break the shackles of imperfect human training" and build its own model of the world; and what would that model be? How soon is "soon."
_How are you defining reason?

[Charles Joseph]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

Think of AI as just another active investor.

Excellent, well said.

AI, or any other form of active investing, transfers wealth from one active investor to another. The total market isn't affected by it--wealth just circulates within the market. And total market indexers is unaffected by it. They are on both sides of every speculative trade, equally balanced.

So, AI, if it works--which I doubt--may temporarily help some active investors take money away from other active investors, until they start using AI, too. But it's not a threat to passive investors.

"Over any specified time period, the market return will be a weighted average of the returns on the securities within the market, using beginning market values as weights. Each passive manager will obtain precisely the market return, before costs. From this, it follows (as the night from the day) that the return on the average actively managed dollar must equal the market return. Why? Because the market return must equal a weighted average of the returns on the passive and active segments of the market. If the first two returns are the same, the third must be also.

This proves (that)....before costs, the return on the average actively managed dollar will equal the return on the average passively managed dollar.

....after costs, the return on the average actively managed dollar will be less than the return on the average passively managed dollar."

- William Sharpe.
The Arithmetic of Active Management

I don't see how AI would ever change that.

[Stormbringer]

Interesting, although your statements (I boldfaced) invite your reasoning:

_What is their "nascent ability" and how do they "clearly exhibit" that ability to reason?

Dr. Geoffrey Hinton (the so-called "godfather of AI") can explain it better than I can: https://youtu.be/qrvK_KuIeJk?t=496

_Why is it "more likely" they will think differently and better than we do and how would they achieve "perfect recollection"?

Rarely do humans build machines that do things the way we do them, and that is because there is almost always a better way to do the job. Humans used to dig ditches with shovels until we built hydraulic backhoes to do the job more efficiently. We used to travel using muscle power and legs, but then we invented engines and wheels that are better at it. There seems to be a human conceit that machines will never be better than us at thinking. We like to believe that characteristic is somehow exempt from the advance of technology.

As for perfect recollection, well that's just how computers work. The information is all there and easily accessible, unless it gets erased or destroyed.

_How will AI "break the shackles of imperfect human training" and build its own model of the world; and what would that model be? How soon is "soon."

A great example of this is DeepMind's AlphaZero. Researchers trained the AI on the basic rules of chess, but nothing about opening moves or endgame strategy. The AI played against itself, and within 24 hours had achieved such a high level of play that not only would it easily destroy grandmasters, but also other chess programs. Starting with only the basic rules of chess, AlphaZero had developed its own model of how to play the game, including strategies that have never been seen before.

I would expect, within a few years, to see AIs that do this with mathematics. Starting with the basic rules of math, the AI should be able to expand and construct its own model of mathematics that far exceeds what humans have accomplished, much as it did with chess. Math is the obvious place to start, because it is the foundation of so many other things and doesn't require interaction with the physical world. Once we have an AI model that is better at math than all the mathematicians who have ever lived, that model then becomes the starting point for other disciplines such as finance and physics. Once you have a model for physics, that becomes your base for engineering, and so on. 

_How are you defining reason?

I'll go with the dictionary definition: the power of the mind to think, understand, and form judgments by a process of logic.

[djmbob]

What Vanguard Mutual Funds should I hold for a 60% stock 40% bond allocation.

Copy
Searching for: Vanguard Mutual Funds for 60% stock 40% bond allocation

Generating answers for you…

For a 60% stock and 40% bond allocation, you might consider the following Vanguard mutual funds:

Vanguard Balanced Index Fund (VBINX)1: This fund consistently sticks to a 60/40 allocation. It takes an indexing investment approach, meaning the fund samples target indexes. The fund seeks to measure the returns of the U.S. stock market with 60% of its assets. The remaining 40% tracks the performance of a “broad, market-weighted bond index,” according to Vanguard1.

Vanguard Balanced Index VBAIX2: This fund has a strategic 60% allocation to the U.S. total stock market (the CRSP U.S. Total Market Index) and a 40% allocation to the U.S. total bond market (the Bloomberg Barclays U.S. Aggregate Bond Index)2.

Combination of Vanguard Total World Stock Index Fund (VTWAX) and Vanguard Total Bond Market Index Fund (VBTLX)3: You could use these two funds to implement a two-fund portfolio3.

Please note that it’s important to consider your own financial goals, risk tolerance, and investment horizon before making any investment decisions. You may want to consult with a financial advisor to ensure that your investment strategy aligns with your financial objectives.

Cheers,
Ray

[nisiprius]

- William Sharpe.
The Arithmetic of Active Management I don't see how AI would ever change that.

'Arithmetic?' What is this 'arithmetic' you speak of? The current of chatbots based on large language models are notoriously bad at arithmetic.

...

To find the average annual return, we need to add up the returns for each year and then divide by the number of years. There are five years in this period, so we will divide by 5.

(-4.50% + 31.35% + 18.29% + 28.78% - 18.19%) / 5 = 12.18%

Why, no, it doesn't. (-4.50% + 31.35% + 18.29% + 28.78% - 18.19%) / 5 = 11.146%.

If The Beardstown Ladies' Common-Sense Investment Guide is part of the training corpus, heaven help us all.

[BradleyB]

Maybe I just don't understand what AI can do.

Can it predict unforeseen events? One time events? You know, like a pandemic. Or an election? Or a major war. Or some foreign leader deciding to raise the price of crude oil.

But I might think about investing in AI companies. I am reminded of the story how during the gold rush of the 1800's, the real money was made by selling stuff to the miners. Levi's jeans, pots and pans. Shovels. Maybe the real money is in data centers and computer chips. They win no matter what the AI predicts.

[rob]

While this "AI" label is all the rage recently... Do you really think that all the algorithm based trading that has been happening for years has not been incorporating these techniques?

[flyingaway]

It is important to know the fees of AI invetsing.

[hoofaman]

Maybe I just don't understand what AI can do.

Right now the main use case is generating memes for entertainment value, sometimes these come with the same copyright logos of the existing memes it "learned" from

When a talking head mentions "AI" on a news program, it's safe to assume whatever is said is pure fantasy

[Elric]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

Think of AI as just another active investor.

The problem is that the first statement is not wholey correct. Passive investing beats ALMOST ALL, BIT NOT ALL, active investing over the long haul.

So the question becomes: will one or more AI models outperform passive.investong over the long haul?

Unfortunately, regardless of whether it's active human or AI investing, one can't determine in advance who will consistently outperform.

[watchnerd]

We all know passive investing beat active investing in long term. Can it also beat AI quantitive investing

AI / algorithmic trading has already been part of the market for many years.

Their activity is part of what your'e capturing with the index.

[Elric]

No thanks, I’ll pass on AI portfolios. I was in Systems and Data Processing (IT) for nearly 40 years and introduced to AI back in 1986-1988. Sorry, it’s not something new. Machines do not reason, they are pre-programed to perform an action on a conditional, requested, situational, or nested-if process.

Expert systems certainly worked that way. Today's deep neural nets? Not so much. I think it's more accurate to state that they are pre-programmed on HOW to learn and given a metric for success. From there, with lots of examples to work from, they develop their own logic and weights for decision-making. For some use cases (e.g., Go and chess) they can also train against themselves.

Now, how will this work for investing? No idea. I'm pretty sure that a large language model like Chat GPT won't.

[watchnerd]

While this "AI" label is all the rage recently... Do you really think that all the algorithm based trading that has been happening for years has not been incorporating these techniques?

Right.

Hedge funds already do this and have been doing this for years.

[GhostMang]

Seems more like the question is will there be AI models that can beat passive investing.

The answer is yes, just like there are active investors who have beat the market and passive index investing. The next question is which AI models will beat the market and that answer is very very few of them. So good luck picking the right one at this somewhat early stage of the game.

If you want to beat the market with a computer become an employee at Renaissance and get access to Medallion fund, but for the rest of us that don't have a phd in CS or a desire to work there I think we're better off with index investing.

[alfaspider]

As others have stated, active investing is a zero sum game. For someone to beat the market, someone else must trail the market. Worse, if any fees are associated with active investing (and there almost always are), your expected value becomes negative vs a passive strategy.

If AI consistently and demonstrably beat the market, then everyone would invest with AI, and AI would be back to zero expected advantage on net. Instead of humans competing against each other to pick stocks, it will simply be different AI models with the same variation in results. You will be no more able to pick the best AI than you can pick the best human fund manager. Of course, you could use AI to pick the AI, and then AI to pick that AI. Pretty soon, it will be AI all the way down

[GaryA505]

While this "AI" label is all the rage recently... Do you really think that all the algorithm based trading that has been happening for years has not been incorporating these techniques?

Right.

Hedge funds already do this and have been doing this for years.

But does it work for the hedge funds?

[watchnerd]

But does it work for the hedge funds?

Varies by fund.

Either way, this phenomenon is already part of what passive index already captures.

AI is just an improvement on algorithmic trading that has been going on for decades and already captured by the indices.

[watchnerd]

FWIW, this question is one of the arguments for holding global market cap.

If AI trading becomes good at generating alpha from inefficient corners of the market, one way to capture some of that is to own the whole market.

[Prokofiev]

What did we used to say in grade school? "Three cheers, two jeers and a moldy pretzel."

Where did you go to grade school? I've never heard that before. Ever . . .

[arcticpineapplecorp.]

What did we used to say in grade school? "Three cheers, two jeers and a moldy pretzel."

Where did you go to grade school? I've never heard that before. Ever . . .

i hadn't either but perhaps is more about age than place. Looked up this quote and it talks about "grandfathers saying..."

https://www.facebook.com/photo/?fbid=17 ... 6926551270
https://twitter.com/theabesnake/status/ ... 84?lang=en

of course those were the only sites I could find referencing this odd phrase. I even asked "What does three cheers, two jeers and a moldy pretzel mean?" and didn't find anything else besides those two sites above.