- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
| Favorite team: | LSU  |
| Location: | Hoist the black flag, slit throats |
| Biography: | |
| Interests: | |
| Occupation: | |
| Number of Posts: | 774 |
| Registered on: | 6/15/2015 |
| Online Status: | Not Online |
Recent Posts
Message
quote:Then it's actually a hybrid test of some sort. I don't know if it has an actual name, but it's basically a robustness test and not directly a risk test. It can indirectly examine risk based on sequencing and/or the possibility of unfavorable generated regimes/events. I don't think it can create any really novel simulated markets with unknown and rare risks (what a true black swan is rather than how it is used in common nomenclature). To do that, they'd need to use something like a jump diffusion model to create their simulated markets. That said, the white swan is already dangerous enough to show the downside risk.
Jason extracted a price-behavior profile from historical data and fed it into a "restricted randomizer that generates daily changes within this profile to simulate many markets."
Overfitting is still an issue, in my opinion. I've no interest in modeling the full systems, but I have analyzed the basic premise of the systems: buying after a quarter failed to meet a certain threshold. To do that, I gathered monthly SPX data from investing.com from February 1970 through December 2025, created quarterly data from it, and examined the results for the quarter after the previous quarter failed to meet the 3% threshold.
2010-2025: The quarter after a signal to increase position size returned an average of 4.2486%; out of 26 signals in 64 quarters, the next quarter exceeded the 3% threshold 18 times, resulting in typically quick trimming (37 trimming quarters in total). The average return when considering all quarters in this period was 3.1558%.
1990-2009: The quarter after a signal to increase position size returned an average of 1.6216%; out of 45 signals in 80 quarters, the next quarter exceeded the 3% threshold 22 times (36 trimming quarters in total). The average return when considering all quarters in this period was 1.7803%.
1970-1989: The quarter after a signal to increase position size returned an average of 0.7323%; out of 39 signals in 79 quarters, the next quarter exceeded the 3% threshold 18 times (40 trimming quarters in total). The average return when considering all quarters in this period was 2.1498%.
If you changed the target to 2%, the results are 4.5097%, 0.844%, and 0.967%, respectively. At 1.5%, they are 4.567%, 1.028%, and 0.967%, respectively. 4% isn't statistically significant from 3%. 5% is 3.7489%, 1.7767%, and 1.1564%.
If you just bought after a negative quarter, your average returns would have been 4.3981% in 2010-2025, 1.016% in 1990-2009, and 1.3287% in 1970-1989.
If you instead bought after a positive quarter, your average returns would have been 2.7754% in 2010-2025, 2.1918% in 1990-2009, and 3.1007% in 1970-1989.
I know this is nowhere close to the 3Sig system, and these don't reflect its actual historical returns for those periods. But I think, despite those limitations, you can understand that this indicates that the system relies on a signal that heavily depends on the current market regime and does not fare nearly as well outside of it. That's an overfitting risk that could mean lower returns to go with the drawdown risks already mentioned should the market regime change.
Honestly, I don't want your capital or the capital of anyone else on this board, I don't want to manage it, and I don't want to extract subscription fees or anything like that from anyone. I don't have a dog in this fight and no incentive to help or hurt you. I'm not incentivized to maximize subscription fees when analyzing or discussing this or any other system. I just like to provide helpful information to retail when I have the time. It's the choice of retail to take my words of advice or ignore them; I'll keep chugging along regardless.
1
quote:No. Does something I've typed imply that?
Do you think I’m saying it’s risk-free?
quote:Tail risks, specifically for the 9Sig system and effectively the risk premium of that particular system. From a glance, 3Sig doesn't have the same issue(s).
What argument are we having?
quote:I know; he somewhat addressed that in the original quote you gave. But he was also dismissive about the risk, disparaged metrics and methods to identify and understand such risks, and bragged about returns in that quote.
Your point about tail risk is fair, and Jason actually addressed that
quote:Just to be clear about terminology, this is called a Monte Carlo permutation test, not to be confused with a regular Monte Carlo simulation. MCPT does indeed check to see if one is overfitting or not, while a regular Monte Carlo sim is to model uncertainty but cannot (directly, anyway) check for overfitting. The only danger of MCPT is if it was used to decide which variant of 9Sig to use, which is p-hacking. I'm not saying your post implied they did so, but since I can't see this information, I would recommend you make sure they didn't for your sake.
Roger generated 100 different simulated 30-year Nasdaq-like markets using historical limits for daily gains, losses, volatility, etc., then tested 972 different combinations of 9Sig rules across them -- 97,200 plan runs in total. The goal was specifically to avoid curve-fitting and see whether the system could survive very different sequences of returns.
quote:When you consider the historical backtest, that's natural and expected. Technically they should have created several thousands, or at least several hundreds, of simulated markets rather than just 100 for their testing, but the 100 still presents the problem. And if they truly only ran a Monte Carlo permutation test, all of the simulations are just rearranging historical data rather than creating new hypothetical markets like a regular Monte Carlo would do. Because the actual market data contained those declines, the MCPT simulations will contain similar ones.
Some of those simulations got extremely ugly.
quote:Not really if they didn't do a regular Monte Carlo, and even if they did, 100 unique market simulations is too small for 30 years of market history. That said, you want a large number of simulations to make sure the system survives a wide variety of possible futures, but only one failed test can discredit a system. Since historical data already points to 9Sig's capability of catastrophic losses (even if it can eventually come back), a regular Monte Carlo is unnecessary. You don't need to concern yourself with testing for potential black swans because you have a known and dangerous white swan.
9Sig was specifically tested against severe, randomized, path-dependent drawdowns rather than just assuming the historical market repeats itself.
At this point, I'll just flatly state the problem. 9Sig's problem is market declines lasting multiple years and/or several quarters that drain both its bond sleeve and fresh capital contributions trying to catch falling knives and thus exasperating losses. Outside of that, you're either experiencing shorter versions of that or, more typically, collecting risk premium. It's just not good risk premium when you consider those drawdowns and their effects on the investor.
You will find very few investors capable of sticking to a system when down 80% or even 90%; depression, stress, anxiety, and psychosomatic pains all appear long before reaching those points. And since investors can't control when those declines happen, they can occur at already bad times for some of the investors. Many can say right now that it's the risk they choose for the returns they're chasing, but their tune will change when that risk is staring them in the face.
That's why I said on the first page that I don't trust Jason Kelly. He disparaged metrics and methods that are used to choose systems with good risk premiums and determine exposure levels people can realistically handle because those metrics and methods show the weakness of his 9Sig system. Then he had the gall to say that those metrics don't define superiority, perhaps implying 9Sig's returns since 2010 do. It would have been great if he had just acknowledged the serious risk his subscribers were taking and that he wanted to mitigate that with an improved version, but that's not what he did.
And after saying all of that, I still find the base idea of 3Sig, not 6Sig and 9Sig, interesting. It's given me some ideas to test, even if they don't look similar to that system.
quote:Are you being deliberately obtuse? Every example I gave was them encountering their specific tail risk. It was never me accusing Mr. Kelly's systems of having the exact same flaws as them.
9Sig is structurally different than LTCM or Niederhoffer's plans.
With normal 9Sig, you aren't borrowing money in your brokerage account. You own TQQQ and a bond fund. There isn't a broker standing there saying, "Send me another $500K by tomorrow morning or we're liquidating you."
Jason also built rules intended to preserve buying power. For example, the 9Sig buying-power throttle prevents a single quarterly buy from consuming more than 90% of the bond reserve, and the spike-reset rule can reduce an excessively high stock allocation after a huge rise.
So 9Sig doesn't have Niederhoffer's obvious Achilles heel of market plunging, margin call, forced liquidation, and game over.
We do have actual live stress tests, though. It was running during the February 2018 volatility explosion that killed XIV. It kept running.
quote:
It also went through COVID. During March 2020 9Sig was roughly 36.5% below its target and signaled a strong buy rather than being forced to liquidate.
2022 was the more interesting test where TQQQ lost roughly 79% for the year. By July, after the first half of the bear market, 9Sig had deployed almost all of its buying power and was 96% TQQQ / 4% bonds.
quote:So now you're admitting that Covid wasn't a specific tail risk for 9Sig. You know what I'm talking about, but you're dancing around it as best you can.
Tail risk for 9Sig isn't a one-day crash. Imagine the 2000-2002 Nasdaq collapse.
quote:-99.7% according to this backtest, which is worse than XIV.
At that point 9Sig hasn't "blown up" like XIV, but economically it could get awfully close.
quote:It's called overfitting, and yes, it matters enormously. It's why Mr. Kelly's backtests of 9Sig never go beyond 2010. He could do longer backtests for 3Sig that show it holds up fairly well through the Great Recession and tech bubble, but then he'd have to answer why he doesn't do the same for 6Sig and 9Sig.
Was I lucky? Was the system just that good? Does it matter?
quote:In some broader ways, yes. But I think you know what I mean: a rare event that will blow up the system and accounts that follow it due to the system's weaknesses. I speak of tail risks specific to the system, like February 2018 for the short vol crowd who thought they could just use that as a buy and hold investment, Victor Niederhoffer's 2 blow ups, or LTCM's blow up.
Covid was a pretty big tail risk.
Mr. Kelly likely recognizes that risk even if he seems to dismiss the naysayers in your quote. If it blows up, his subscription money does, too.
Bragging about your high beta approach having high returns in a bull market and that tail risks haven't materialized yet is certainly a choice on his end. I wouldn't trust him, especially after this part:
quote:
Many highly ranked rules-based plans focus on maximum drawdown, Sharpe ratio, smoothness, and such. Those are useful measurements, particularly for risk-averse investors, but don’t define superiority. A strategy with substantially lower return usually looks better by those measures because it carries less equity or leverage exposure.
Indeed. My thing is that anytime I see something I consider flawed (if not outright garbage) like Kelly Criterion or Sharpe ratio, I point out its problems because initial information on the internet often just mentions the benefits and what it should do in theory.
I like to think it's the financial equivalent of warning a friend to avoid the hot chick with BPD.
I like to think it's the financial equivalent of warning a friend to avoid the hot chick with BPD.
quote:
dstone12
You posted the Kelly Criterion, which is theoretically how much you should risk on each trade/position or gambling bet. In reality, it is far too aggressive and reckless when trading because KC doesn't consider the distribution of losses (only the average loss) and the possibility of multiple consecutive losing trades. Too many of my backtests show it blowing up an account with half Kelly due to a single bad trade from the tail end of the distribution or with quarter Kelly after a string of bad trades.
re: Why is Warren Buffett so revered?
Posted by Omada on 9/25/26 at 7:15 pm to Shepherd88
Even after their high fees, RenTech's CAGR is about 40%.
Others off the top of my head:
Jane Street's net trading revenues basically doubled in both 2024 and 2025, thought their overall CAGR isn't public information.
Victor Niederhoffer's CAGR before his first blow up was 35%, and his second fund did 50% before it blew up.
Paul Tudor Jones is just 19%, but he beat VN over the long run.
Others off the top of my head:
Jane Street's net trading revenues basically doubled in both 2024 and 2025, thought their overall CAGR isn't public information.
Victor Niederhoffer's CAGR before his first blow up was 35%, and his second fund did 50% before it blew up.
Paul Tudor Jones is just 19%, but he beat VN over the long run.
quote:
His last 20 years tracks the market, not outperfroms it.
I'm not a WB fan, but it's harder to find meaningful opportunities the larger you grow. Eventually you get too big for your pond and see your returns diminish unless you can diversify into other things that can match your prior returns (which is rare).
re: Can any physicians or healthcare folks comment on Omada Health? OMDA
Posted by Omada on 9/22/26 at 11:32 pm to bayoubengals88
Bastards using my name without permission :angry:
re: Where to put kid's money
Posted by Omada on 9/15/26 at 3:07 pm to bayoubengals88
quote:
Courtesy of Steve Ellison
All I did was link it here
Alpha and Sharpe Ratios according to Morningstar:
Ticker; 3 Year, 10 Year Alpha; 3 Year, 10 Year Sharpe Ratio
PBP 1.84, -2.44; 1.23, 0.5
VEGA -1.78, -3.5; 0.95, 0.51
XYLD 1.58, -2.04; 1.19, 0.57
QYLD 2.7, -0.56; 1.26, 0.69
FTHI 0.13, -2.7; 1.14, 0.54
FTQI 0.95, -1.76; 1.22, 0.56
In other words, all but VEGA have done well in the past 3 years but have done quite poorly over the past 10.
Ticker; 3 Year, 10 Year Alpha; 3 Year, 10 Year Sharpe Ratio
PBP 1.84, -2.44; 1.23, 0.5
VEGA -1.78, -3.5; 0.95, 0.51
XYLD 1.58, -2.04; 1.19, 0.57
QYLD 2.7, -0.56; 1.26, 0.69
FTHI 0.13, -2.7; 1.14, 0.54
FTQI 0.95, -1.76; 1.22, 0.56
In other words, all but VEGA have done well in the past 3 years but have done quite poorly over the past 10.
re: Looking for recommendations on books/courses for a teenager
Posted by Omada on 9/10/26 at 8:28 pm to sheepshead
quote:
The Most Important Thing by Howard Marks.
I really like this one. It's a book about abstract ideas rather than formulas, so teenagers can pick it up and remember the principles without trying to figure out what a standard deviation is.
Some of the Little Books, Big Profits series are good, though I haven't read the vast majority of them to know if they all are. I believe someone else recommended Bogle's; I can recommend Joel Greenblatt's, Christopher Browne's, and Aswath Damodaran's. I'd avoid Michael Covel's since I consider him to be scummy with some of his marketing antics and thus unreliable.
I've gifted Fooled by Randomness and The Black Swan to high school and college grads.
quote:
A Random Walk Down Wall Street
I'll start by saying that I haven't read the book and that I agree with the premise that regular retail should play it safe since they have no edge, but I disagree with some of the author's arguments to try to make that point. I wouldn't call it a terrible book, but if your kids do enough digging into the world of financial markets, they can find some conflicting evidence and/or additional information that puts some arguments in a different light.
re: Heather Dinich is on my flight from Houston
Posted by Omada on 9/4/26 at 5:37 pm to Jack Crevalle
Someone out there thought that was a good picture of her to post on the internet. Yeesh
quote:
So my utilization is high because apparently the AC company ran the 0% thing as a wells fargo credit card? with a maximum of 18k and I still owe about 12k? idk if thats usual or not; its still 0% for 36 months and I will definitely pay it off before then, but thats how it shows up on my credit report. Could that be nuking it?
If it is actually a credit card, then yes. The credit utilization aspect only considers revolving lines of credit, not something like your mortgage (unless you put that on a credit card!). And in that case, as far as the credit agencies are concerned, you just opened a new credit card and are using about 67% of its limit (but not your overall limit). However, if it is just a traditional loan, then I don't think it would (someone correct me if I'm wrong, please).
I wouldn't sweat it, though. As you pay it off, your credit score will go back up. Personally, unless you are planning on buying a new house within the 36 months of 0% APY, I would figure out my payments so that I'd pay it off just before any interest would apply. This would see your credit score recover at a slower rate, but you could put any excess cash into retirement, savings/emergency fund, or your mortgage.
One other possibility to lower your credit utilization is to accept any offers to increase the limit on your credit cards so long as the offer won't result in a hard credit inquiry. These offers usually only do a soft inquiry, but you'll need to make sure. You generally won't want to make a request for a credit increase because that will require a hard inquiry, and those ding your credit score.
re: Statement from the SEC
Posted by Omada on 9/3/26 at 9:54 pm to burreauxsballz
That's cute, SEC. But before you do anything else, just remember we're ready to spit on our hands, hoist the black flag, and begin slitting throats.
If you have an account with Capital One, you can download their app and use CreditWise, which shows your approximate credit score and how the different factors affect it. It also has a simulator that allows you to experiment with changes.
The factors and their weights on your credit score are:
1. Payment History (35%): whether you have late/missed payments, bankruptcy, collections, etc.
2. Credit Utilization (30%): how much credit you are using. More utilization generally means worse credit, but a marginal amount (like 1%) is better than 0% for the calculation.
3. Credit History Length (15%): the ages of your oldest, newest, and average account. Older is better here.
4. New Credit (10%): the fewer new credit lines you have, the better.
5. Credit Mix (10%): the different types of credit accounts you have (credit card, mortgage, auto loan, etc.). More types is generally better here.
So if we look at these, you've said you have a problem with #1, you paid off an auto loan that reduced your credit mix (#5), and you opened a new credit line that worsened #2-4 (but may have helped with #5).
Since 1 and 2 have such heavy weighting, they should be your primary focus. For 1, you can either try some sort of negotiation or just wait for it to roll off, which will happen 7 years after the original delinquency date. 2 can be dealt with over time as you pay off loans. 3 and 4 are just a matter of time. I'm not sure if you can do much about improving 5, but it has low weight anyway.
The factors and their weights on your credit score are:
1. Payment History (35%): whether you have late/missed payments, bankruptcy, collections, etc.
2. Credit Utilization (30%): how much credit you are using. More utilization generally means worse credit, but a marginal amount (like 1%) is better than 0% for the calculation.
3. Credit History Length (15%): the ages of your oldest, newest, and average account. Older is better here.
4. New Credit (10%): the fewer new credit lines you have, the better.
5. Credit Mix (10%): the different types of credit accounts you have (credit card, mortgage, auto loan, etc.). More types is generally better here.
So if we look at these, you've said you have a problem with #1, you paid off an auto loan that reduced your credit mix (#5), and you opened a new credit line that worsened #2-4 (but may have helped with #5).
Since 1 and 2 have such heavy weighting, they should be your primary focus. For 1, you can either try some sort of negotiation or just wait for it to roll off, which will happen 7 years after the original delinquency date. 2 can be dealt with over time as you pay off loans. 3 and 4 are just a matter of time. I'm not sure if you can do much about improving 5, but it has low weight anyway.
re: Red! Green!! Red! Green!!
Posted by Omada on 9/2/26 at 3:07 pm to Upperdecker
Taken from dailyspeculations.com; posted by Steve Ellison in 2024 and called the wall of worry.
EDIT: I'm posting the chart without my own commentary.
EDIT: I'm posting the chart without my own commentary.
August and September are generally poor months for the S&P. Maybe this month bucks the trend, or maybe you'll have to wait until October or November.
re: Day Trading Noob
Posted by Omada on 8/20/26 at 9:22 pm to LSUTIGERS74
The vast majority of my backtesting has been with 1D or larger data, not intraday. In my opinion, swing trading edges are pretty easy to find compared to intraday ones. Maybe one day I'll get back around to doing intraday backtesting just to see what I can find, but it's something I don't need to do at this point.
Are you saying you've only done 29 backtests?
quote:
On my own journey I have tested pretty much every major statistical catagory, research paper theory, or social media trader setup over the full databento historical CME and OPRA datasets to no avail (29 of 29 kills on backtested theories
Are you saying you've only done 29 backtests?
I love the first 2 comments I can see of that tweet, especially this one:
That said, it's not really a surprise that most of them fail to outperform. Since the fund fees are just a percentage of AUM and not a percentage of returns, the name of the game is principal retention, which means managers often won't stray too far from each other and the index. If they do, overperformance will attract more principal and produce more fees, but underperformance means principal withdrawals, getting fired, and potentially becoming a black sheep in the industry and/or a scapegoat for the next firm that hires you. The juice is often not worth the squeeze even for the ones capable of outperformance.
The steeper fees are also a drawback, obviously, and fund size can limit viable investment options that materially impact the portfolio. Warren Buffett will tell you it's not all peaches and cream being a big fish.
quote:
i beat the s&p 500 and im just a retard who throws money at palantir and anything indians on twitter tell me to invest in
That said, it's not really a surprise that most of them fail to outperform. Since the fund fees are just a percentage of AUM and not a percentage of returns, the name of the game is principal retention, which means managers often won't stray too far from each other and the index. If they do, overperformance will attract more principal and produce more fees, but underperformance means principal withdrawals, getting fired, and potentially becoming a black sheep in the industry and/or a scapegoat for the next firm that hires you. The juice is often not worth the squeeze even for the ones capable of outperformance.
The steeper fees are also a drawback, obviously, and fund size can limit viable investment options that materially impact the portfolio. Warren Buffett will tell you it's not all peaches and cream being a big fish.
Popular
30
