That said, if it's possible to do better than random guessing, then does this reflect the fact that the five charts are presumably hand-selected to be "interesting"?
My naive guess, and I'd be very curious to learn if this were wrong, is that something very close to the efficient market hypothesis is true; that, if it were possible to beat the monkeys on randomly chosen stocks on random dates, then someone would have figured this out already and deployed bots to capture whatever profits are available.
Thanks for flagging, summary is:
Charts are from a seeded random draw.
And the instinct is pretty spot on players state ~74% average confidence but hit ~54%.
Accuracy is basically flat across confidence brackets.
Up only strategy quietly beats the coin flipping monkeys (stocks generally go up and to the right) but players call down 43% of the time.
They're from a seeded random draw. Random S&P 500 stock, random 60-day window that's at least 12 months old, every player gets identical charts. Gemini writes the narrative vignette.
Your instinct is pretty spot on. 6,000+ calls in: players state ~74% average confidence but hit ~54%, and accuracy is basically flat across confidence brackets. Up only strategy quietly beats the coin flipping monkeys but players call down 43% of the time.
> Read the Tape gives players the same 5 S&P500 stock charts per day to predict. You select low, medium or high confidence and then call the chart UP or DOWN. It's a 1d chart which then resolves over 5 days. Alpha is scored against the Monkey Index, a basket of 11 random coin flips at low confidence which provides a tangible win/lose condition.
Your description here and on the website is not clear to me about what I am predicting. Am I predicting whether the stock is going to continue to go up the same day? Is the chart of one day and I am predicting whether it will go up or down the next day? In my opinion, you need to get the explanation of what I'm looking at and the directions of what I am predicting down to one clear sentence.
Yes I think you're right, it's unclear. I'm hopeful the first reveal explains the dynamic as the player goes.
Maybe something like- each chart shows 60 daily candles of a real S&P 500 stock, a past window with the ticker and dates hidden, you call if the stock closes higher or lower 5 days later?
- “up” and “down” was explained. If I bet “down” I’m betting my stock will underperform the monkey index? And if I am correct, I will make money (short the asset) or simply loose less than if I’d invested in monkey?
- loosing less money was green (win) when monkey is down even more.
Thanks for surfacing. Up or down is a call on the stock price alone, nothing to do with the monkeys. Down means you think its close is lower in 5 days. The monkeys make a call up or down on the stock price too, and then their pnl is subtracted from yours.
So you can lose money on the trade and still go green because the monkeys lost more, or make money and go red because a monkey made more. The idea is it's your edge over random. Hope that makes sense and thanks for raising, will strive to make it clearer in-game.
Edit- to clarify Up/Down are a plain long or short, no options and no leverage. Low, medium and high are position sizing.
I also was confused by this and the usage of the term bet for low/med/high risk. Are up and down supposed to model buying calls/puts? Or buying the stock itself? Wagering 8800 USD of 10000 on high to get 115 USD return was pointing at stocks (rather than options).
Thanks for posting. It's an interesting concept, but needs some work in my opinion.
First, that chart is super zoomed in. You can't zoom out or switch timeframes. You're not told how many bars you need to predict. You're not told the dates or what the stock is.
Not much you can see there other than a handful of candles, that's doesn't really tell you much.
If the stock and dates are a secret so people don't cheat, I don't see why. People can cheat on Wordle or Worldle, and they're still fun to play. But here you're basically just guessing, you have so little information it's basically a coin flip.
Love how you implement bet sizing, this is a signature of professional traders. How did you come up with the idea?
You mentioned stocks are "seeded random draw", but professionals don't trade random stocks - there is no edge there, price movements are mostly noise. You want stocks-in-play, stocks with heightened interest from investors on that day.
My suggestion is to manually pick interesting stocks each day and ask players what happens next. Don't hide names, so that players can asses market strength and group strength - 75% of success comes from these factors, and only 25% from individual stock selection.
Thanks. My original idea was some kind of 'Duolingo for TA'. Your suggestion would definitely be very relevant to a less gamey, more serious product. great thought.
Would love to use 'Duolingo for breakouts' for studying how stocks behaved just before they went on a 50%+ move. Successful traders spend thousands of hours studying these to build pattern recognition.
I ended up generating alpha so this is clearly a great game and I'm a genius investor (compared to random monkeys). When I lose it all tomorrow, my opinion may of course change.
Seriously though, this is a clever idea and I'm interested to see if I can consistently beat the monkeys. Time will tell! Thank you for sharing!
Thinking there are straightforward, simple algorithmic strategies to beat the market requires one to also believe that the billions of dollars behind large investments for some reason doesn’t care about using these same strategies to make money.
“They have different goals”, I hear. Literally nobody’s investment strategy involves passing up above-average returns for low risk.
There are plenty of known effects that are relatively simple to automate
Holding overnight risk yields superior risk-adjusted returns. Buying end of month and dumping few days into new month. There are dozens of such effects
These have mechanical reasons for their outperformance - overnight risk has to do with how borrow interest rates for equity markets are calculated and firms unwilling to hold unhedged exposure overnight. Longing EOM has to do with people getting their paycheck EOM and auto buying index funds, and so on and so forth.
Systematic trading isn't magical, it's identifying these kinds of (often simple) effects and building a portfolio of them.
As an individual investor you actually have a huge advantage over large institutions in that your portfolio is nimble and easy to get out of.
As a relatively simple exercise - consider a hypothetical portfolio that's simply long SPY for the year. Could you identify _one_ day in which you'd rather be flat? The answer is probably yes, and the reason you can do this (and not a billion $ AUM fund) is that rotating in and out of positions is cheap for you.
When news about Iran hits the tape, who do you think can exit their positions faster? Joe Schmoe with $30k in his brokerage account, or Citadel with a $100bn position?
You cannot in the same breath argue that there are simple market strategies that offer superior risk-adjusted returns, and then claim that big money won’t go after it because they can’t get out.
There are more actors in the market than just Citadel. For starters there are day traders at virtually every wealth level and with every level of risk tolerance. Someone—or more accurately, many someones—with more money, better information, and better market access is going to claim that free alpha until there’s none left. The notion that there’s so much free, obvious, and reliable market edge simply lying there for the taking by any shmuck with $100 to put into Robinhood is farcical and flies in the face of quite literally all empirical data we’ve collected on the performance of active market participants.
Every single time this kind of discussion happens on the Internet, people inevitably rush to the comments to say that they know the magic strategy that beats the passive indexes. And yet every single study to date has shown that active market participants perform worse than chance and there is zero correlation between those who beat the indexes one year and those who do so the next.
> As an individual investor you actually have a huge advantage over large institutions in that your portfolio is nimble and easy to get out of.
This is a joke. By the time you or I can act on market information, those same big players have known for a comparative eon; more than long enough to change the price of those investments enough to remove any edge.
If a big enough proportion of the market is also following the same technical analysis strategies as you, it will predict the market. There doesn't have to be anything actually correct about the analysis.
In reality I don't think a big enough proportion of investors are reading tea leaves for this to be true.
The config is currently if you're on desktop, you get text copied to the clipboard. On mobile, you get text + an image to share. Sharing points are 'Challenge a friend' and 'Share result' on the Performance Review, or you can share your career card from the Career Record.
I added the image to the mobile shares as early users were screenshotting for x eg https://x.com/investingidiocy/status/2075471303809659212
It’s a fun game and it throws funny shade for getting the answers wrong, but who would have guessed and sized these answers right based on this sparse data without any additional context? There’s barely enough data in the charts to do TA properly.
Well the map is not the territory. Just because there’s been a big gap down on a chart that wasn’t the chart doing that, the chart displays that as some kind of causal event like an earnings release or something the company has done to make the market react. So I don’t know what kind of Quant can look at that sparse TA data knowing nothing else about the company and make those calls accurately or know how to size the positions. One random money in a crowd will randomly do better than trying to sensibly do TA with this data.
Exactly — chart-only TA is tough without fundamentals context. For Korean markets this is even more pronounced since retail flow dominates. I've been working on fundamental data for KOSPI/KOSDAQ if you ever want to add an Asian market mode.
Volume is on there but not shown until toggled on. Yeah making it blind on company and broader macro/market movements is what produces some of the the jeopardy, and outperforming the random monkeys turns out to be a bit tougher than some players might be expecting.
This is fascinating.
That said, if it's possible to do better than random guessing, then does this reflect the fact that the five charts are presumably hand-selected to be "interesting"?
My naive guess, and I'd be very curious to learn if this were wrong, is that something very close to the efficient market hypothesis is true; that, if it were possible to beat the monkeys on randomly chosen stocks on random dates, then someone would have figured this out already and deployed bots to capture whatever profits are available.
> That said, if it's possible to do better than random guessing
Yes - by always picking Up.
Random stock on a random day has 53-55% chance of closing higher. Over 5 days, you will be right ~60%
1996–2016: 53.3% of days were positive. 2016–2021: 54.9% of days saw gains.
The baseline is the market average, not 0.
The interesting answer by the author, will_asouka, has been marked 'dead' for some inscrutable reason.
Thanks for flagging, summary is: Charts are from a seeded random draw. And the instinct is pretty spot on players state ~74% average confidence but hit ~54%. Accuracy is basically flat across confidence brackets. Up only strategy quietly beats the coin flipping monkeys (stocks generally go up and to the right) but players call down 43% of the time.
They're from a seeded random draw. Random S&P 500 stock, random 60-day window that's at least 12 months old, every player gets identical charts. Gemini writes the narrative vignette.
Your instinct is pretty spot on. 6,000+ calls in: players state ~74% average confidence but hit ~54%, and accuracy is basically flat across confidence brackets. Up only strategy quietly beats the coin flipping monkeys but players call down 43% of the time.
> Read the Tape gives players the same 5 S&P500 stock charts per day to predict. You select low, medium or high confidence and then call the chart UP or DOWN. It's a 1d chart which then resolves over 5 days. Alpha is scored against the Monkey Index, a basket of 11 random coin flips at low confidence which provides a tangible win/lose condition.
Your description here and on the website is not clear to me about what I am predicting. Am I predicting whether the stock is going to continue to go up the same day? Is the chart of one day and I am predicting whether it will go up or down the next day? In my opinion, you need to get the explanation of what I'm looking at and the directions of what I am predicting down to one clear sentence.
Yes I think you're right, it's unclear. I'm hopeful the first reveal explains the dynamic as the player goes. Maybe something like- each chart shows 60 daily candles of a real S&P 500 stock, a past window with the ticker and dates hidden, you call if the stock closes higher or lower 5 days later?
It would find it clearer if:
- “up” and “down” was explained. If I bet “down” I’m betting my stock will underperform the monkey index? And if I am correct, I will make money (short the asset) or simply loose less than if I’d invested in monkey?
- loosing less money was green (win) when monkey is down even more.
Thanks for surfacing. Up or down is a call on the stock price alone, nothing to do with the monkeys. Down means you think its close is lower in 5 days. The monkeys make a call up or down on the stock price too, and then their pnl is subtracted from yours.
So you can lose money on the trade and still go green because the monkeys lost more, or make money and go red because a monkey made more. The idea is it's your edge over random. Hope that makes sense and thanks for raising, will strive to make it clearer in-game.
Edit- to clarify Up/Down are a plain long or short, no options and no leverage. Low, medium and high are position sizing.
I also was confused by this and the usage of the term bet for low/med/high risk. Are up and down supposed to model buying calls/puts? Or buying the stock itself? Wagering 8800 USD of 10000 on high to get 115 USD return was pointing at stocks (rather than options).
Thanks for posting. It's an interesting concept, but needs some work in my opinion.
First, that chart is super zoomed in. You can't zoom out or switch timeframes. You're not told how many bars you need to predict. You're not told the dates or what the stock is.
Not much you can see there other than a handful of candles, that's doesn't really tell you much.
If the stock and dates are a secret so people don't cheat, I don't see why. People can cheat on Wordle or Worldle, and they're still fun to play. But here you're basically just guessing, you have so little information it's basically a coin flip.
Love how you implement bet sizing, this is a signature of professional traders. How did you come up with the idea?
You mentioned stocks are "seeded random draw", but professionals don't trade random stocks - there is no edge there, price movements are mostly noise. You want stocks-in-play, stocks with heightened interest from investors on that day.
My suggestion is to manually pick interesting stocks each day and ask players what happens next. Don't hide names, so that players can asses market strength and group strength - 75% of success comes from these factors, and only 25% from individual stock selection.
Thanks. My original idea was some kind of 'Duolingo for TA'. Your suggestion would definitely be very relevant to a less gamey, more serious product. great thought.
Would love to use 'Duolingo for breakouts' for studying how stocks behaved just before they went on a 50%+ move. Successful traders spend thousands of hours studying these to build pattern recognition.
I ended up generating alpha so this is clearly a great game and I'm a genius investor (compared to random monkeys). When I lose it all tomorrow, my opinion may of course change.
Seriously though, this is a clever idea and I'm interested to see if I can consistently beat the monkeys. Time will tell! Thank you for sharing!
From a very non altruistic point of view part of me wants to encourage this
Oh boy, I know more than a few guys who will get a kick out of this, hah
Appreciate you spreading the word, thanks
To me the description was very clear, the game is fun to play. Would love to see the "prediction" of a LLM in the stats after playing
Nice idea, maybe when there's enough traction for weighted scoreboards I could enter some models
Technical analysis is astrology for boys.
Thinking there are straightforward, simple algorithmic strategies to beat the market requires one to also believe that the billions of dollars behind large investments for some reason doesn’t care about using these same strategies to make money.
“They have different goals”, I hear. Literally nobody’s investment strategy involves passing up above-average returns for low risk.
There are plenty of known effects that are relatively simple to automate
Holding overnight risk yields superior risk-adjusted returns. Buying end of month and dumping few days into new month. There are dozens of such effects
These have mechanical reasons for their outperformance - overnight risk has to do with how borrow interest rates for equity markets are calculated and firms unwilling to hold unhedged exposure overnight. Longing EOM has to do with people getting their paycheck EOM and auto buying index funds, and so on and so forth.
Systematic trading isn't magical, it's identifying these kinds of (often simple) effects and building a portfolio of them.
As an individual investor you actually have a huge advantage over large institutions in that your portfolio is nimble and easy to get out of.
As a relatively simple exercise - consider a hypothetical portfolio that's simply long SPY for the year. Could you identify _one_ day in which you'd rather be flat? The answer is probably yes, and the reason you can do this (and not a billion $ AUM fund) is that rotating in and out of positions is cheap for you.
When news about Iran hits the tape, who do you think can exit their positions faster? Joe Schmoe with $30k in his brokerage account, or Citadel with a $100bn position?
You cannot in the same breath argue that there are simple market strategies that offer superior risk-adjusted returns, and then claim that big money won’t go after it because they can’t get out.
There are more actors in the market than just Citadel. For starters there are day traders at virtually every wealth level and with every level of risk tolerance. Someone—or more accurately, many someones—with more money, better information, and better market access is going to claim that free alpha until there’s none left. The notion that there’s so much free, obvious, and reliable market edge simply lying there for the taking by any shmuck with $100 to put into Robinhood is farcical and flies in the face of quite literally all empirical data we’ve collected on the performance of active market participants.
Every single time this kind of discussion happens on the Internet, people inevitably rush to the comments to say that they know the magic strategy that beats the passive indexes. And yet every single study to date has shown that active market participants perform worse than chance and there is zero correlation between those who beat the indexes one year and those who do so the next.
> As an individual investor you actually have a huge advantage over large institutions in that your portfolio is nimble and easy to get out of.
This is a joke. By the time you or I can act on market information, those same big players have known for a comparative eon; more than long enough to change the price of those investments enough to remove any edge.
If a big enough proportion of the market is also following the same technical analysis strategies as you, it will predict the market. There doesn't have to be anything actually correct about the analysis.
In reality I don't think a big enough proportion of investors are reading tea leaves for this to be true.
I call it money-astrology.
Needs a way to copy and paste text to share with friends. I feel like this is what made wordle viral
The config is currently if you're on desktop, you get text copied to the clipboard. On mobile, you get text + an image to share. Sharing points are 'Challenge a friend' and 'Share result' on the Performance Review, or you can share your career card from the Career Record. I added the image to the mobile shares as early users were screenshotting for x eg https://x.com/investingidiocy/status/2075471303809659212
It’s a fun game and it throws funny shade for getting the answers wrong, but who would have guessed and sized these answers right based on this sparse data without any additional context? There’s barely enough data in the charts to do TA properly.
If I was going to add 3 more indicators, what should they be? And/or more history?
Well the map is not the territory. Just because there’s been a big gap down on a chart that wasn’t the chart doing that, the chart displays that as some kind of causal event like an earnings release or something the company has done to make the market react. So I don’t know what kind of Quant can look at that sparse TA data knowing nothing else about the company and make those calls accurately or know how to size the positions. One random money in a crowd will randomly do better than trying to sensibly do TA with this data.
Maybe Volume would be useful though.
Exactly — chart-only TA is tough without fundamentals context. For Korean markets this is even more pronounced since retail flow dominates. I've been working on fundamental data for KOSPI/KOSDAQ if you ever want to add an Asian market mode.
Volume is on there but not shown until toggled on. Yeah making it blind on company and broader macro/market movements is what produces some of the the jeopardy, and outperforming the random monkeys turns out to be a bit tougher than some players might be expecting.
200 day, ATR, ATR extension from the 50
shipped
I 100% think you are going to harvest these decisions to train an ML model to do the inverse.
you should probably rename it since reading the tape is the opposite of charts :p
This is impressive and a tremendous concept.
I did 5/5, looking for quant jobs now
you should include volume
It's under the chart, toggled off by default. 3rd comment so very obviously needs to be made more prominent, on it.
would love to see Volume on the chart
On there as a toggle off by default, under the chart. You're the second to not notice it so design feedback duly noted
[dead]