|
|
|||
| |
 |
Ed
Seykota's FAQ
|
| Home | Send Mail to FAQ | Resources | Ground Rules | FAQ Index |
Epidemic
Lab
|
| FAQ October 1-31,
2023 <== back | down | forward ==> |
| Contributors Say | Ed Says | |||||||
| October
29, 2023 Core Damage Hi Ed, I am done designing a new risk management indicator and I have an official name for it, Core Damage. It tracks the damage / reduction to the core when a new position opens. It changes when I move protective stops. It's different from the core itself, also different from the concept of open-trade risk. It tracks the damage to core by each individual trade or position and each market, it can also track total Core Damage for the entire account. I intend to use it to avoid a situation where I keep adding new positions in one market, or opening a bunch of new positions across many markets during a short period of time, while none or few of them are ready to move up protective stops. I find in my actual trading I have no current method to mitigate risk in such situations. I find a frequent scenario happens to my actual trading, specially when my account is small, and now it's possible to test it. The idea under test is that if several strong markets, from different industry groups move in a common time period, there might be a way to space out / even out the entry damage, by entering different markets at different phase, instead of entering them at same short period of time, by combining entry techniques. The worst case is missing some market but I can manage the downside better. Have fun! |
Thank
you for letting me know about your new indicator. I wonder what advantage it shows on your back tests. |
|||||||
| October
27, 2023 New Headquarters Ed, I enclose a photo of our new headquarters in Singapore. Regards
|
Thank
you for sending me your photo. The space looks plenty big enough to run your trend system on a laptop - and plenty small enough to avoid the temptation to hire a lot of economists. |
|||||||
| October
24, 2023 Risk Number Hi Ed, I’m always trying to prepare my clients for the inevitable downturns that are part of investing. However, it’s hard to accurately pinpoint exactly how you’ll react to a bad month or a bad year. A critical part of my job is understanding and managing investor psychology to benefit your financial future. [Firm] utilizes the Risk Number to mathematically quantify something as complex as your unique emotional reaction to losing money and boils it down to a single number. Calculating your Risk Number is just the first step. Once I have it, I’ll use it to review your existing portfolio and stress test it by simulating major market events and show you in dollars and cents just how your portfolio would have performed. I want us to have an informed discussion about the risks you’re taking now and what could happen during downturns and volatility. Can you confidently say your portfolio represents your comfort level? Go ahead and get your Risk Number today. It takes 5 minutes. Find out more here. Regards, |
Thank
you for offering to reduce my emotional reactions to a single number
that you deduce from a list of intellectual questions. I wonder if you might consider other ways for deducing client feelings about risk and aligning your portfolios to reflect their preferences. |
|||||||
| October
23, 2023 Picking Bottoms Hey Ed, I hope you are having a great Monday! I just got back to the US from Switzerland. I want to come out to Austin and see you at some point in the next few weeks if you are in town and available. I put a long trade on Nvidia using the pattern I mentioned above. The region around $410 (and to a more important extent, $400) seemed to be support in a strong uptrend. I set a protective stop at just under $408 and got my entry at $410 on 1400 shares, for a total base case risk of around $2800 on my trading account size of $1 million. The pattern "worked" again which is cool to see. Below is the setup.
|
Thank
you for sending me the charts for your trade. I notice that after your purchase at $410, the market briefly trades under $408, electing your stop and closing your position at a loss before rising.
|
|||||||
| October
22, 2023 Equity Curves Ed, I attach charts of 1. equity of my total equity 2023 (including cash trading account and various IRAs for me and my wife). I was down a lot at beginning of the year and am able to regain some ground with your help. Now I'm trying to stabilize. I'm up about 25% YTD. 2. The equity chart 2023 of my personal cash account. I lost quite a lot money on my cash account over years (60% drawdown from original capital 5 years ago) and this account starts low as $54k beginning of this year. I am trying to catch up and trade heavier than other accounts. This causes more drawdown than other accounts. now I am down to smallest risk config for this account. I'm down -13% YTD on this account. Thanks,
|
Thank
you for sending me the charts. I notice the account that performs best also trades with less heat (risk-to-equity-ratio). You might consider taking your feelings about <wanting to catch up> to Tribe as an entry point. |
|||||||
| October
17, 2023 Tubes and Cones Dear Mr. Seykota, I enjoyed reading about your demonstration with the tube and cone and air flow (). I have known for a long time about the misinterpretation and misapplication of Bernoulli's principle in many demonstrations. I really like your cone version! I can easily get my cone to do the same thing conspicuously. My question arises from my observations that when I make a tube out of the same material (newspaper), true, it does not collapse, but it definitely tries to. There are sections of it that are trying to collapse on itself. It does not collapse though due to its stronger geometrical structure. If I make a tube with thinner and thicker walls of newspaper (by using more or less layers of paper), I can see the collapsing effect more or less, yet it is still there. I cannot seem to reconcile this with your webpage's descriptions. Any advice would be most appreciated. Thank you. Sincerely, |
Thank
you for raising this issue. The tube and cone demo aims to show a substantial difference in behavior between the tube and the cone; it does not aim to explain differences between an ideal tube and an actual paper tube. You might consider examining your paper tube for (1) constant diameter, (2) perpendicularity of the exit orifice, (3) means of holding the tube together at the exit end, and (4) sources of tube vibration. Small irregularities and vibrations may occasion small deformities in the exit orifice, leading to some "cone-ness" of the tube. You might try increasing the air flow and also taking some high-speed close-up videos of the orifice to find out more. In theory, we have no difference between theory and practice; in practice we do.
|
|||||||
| October
14, 2023 Volatility Filter Happy Saturday Ed, My standard trend following systems trade a universe of ~60 diversified futures contracts denominated in USD, use the simple support/resistance and position sizing system you show me (fast length of ~100 days, slow length of ~200 days, size based on risk per lot) for all instruments, 1% heat per position, with max exposure of 500% for both the long and short postures. As I mention to you, the average MAR of this system over the past 30 years tends to be in the 0.2-0.4 range. I add a volatility filter to my entry and exit rules. For each instrument, I define a fast and a slow volatility look-back period. The fast look-back ranges from 10-30 days and the slow look-back ranges from 50-150 days. I require the historical volatility over the fast length to be less than the historical volatility over the slow length as an entry rule. I also add an “or” statement to my exit rule — I close a position if it trades through my protective stop OR if historical volatility over the fast length is greater than the historical volatility over the slow length. I then vary these parameters to give a heat map and find an optimal region for the look-back periods on the volatility variables. I find that by adding this “volatility filter” that takes me out of the market and brings risk down when there is unusually high volatility increases the average MAR of the systems dramatically — now the MAR on most systems is in the 1.5–3 range. Does this drastic change in results seem to be directionally consistent with what you observe in your studies? I can send you my heat map if you would like to see it. I want to make sure I’m not missing something — as Richard Feynman says, “The first principle is that you must not fool yourself, and you are the easiest person to fool!” |
Thank
you for raising the issue of using increasing volatility as a long exit
rule. Volatility normally expands during a major bull market. Your rule might sell you out early in the move. You might consider back testing your system over some few-in-a-lifetime bull markets to see how well your positions survive. You might also consider taking your feelings about <taking profits on the way up> to Tribe as an entry point. |
|||||||
| October
14, 2023 Wants to Lose the Need to Win Hi Ed, I just finished [a book]. My goal was to go through the whole course twice and I’m getting close. I think the mechanical trader he is referring to in the book, is you. I could be wrong, but based on the answers, it sounds like you. What are some of the ways you suggest to get rid of the need to win in trading? “the need to win drain him of energy” Eckert Tolle. I’m starting to think maybe that is part of my problem. Although I have blown my trading account many times, I somehow still have that stubborn confidence that I am going to make it in trading. |
Thank
you for reaching out about blowing your account many times. You might consider examining your life for other evidence of dopamine-centric behaviors. You might also consider taking your feelings about <needing to win> to Tribe as an entry point.
|
|||||||
| October
11, 2023 Wants Help Hello Ed, I ran the program using this input file SP----C.csv my output files are attached. Do you know how I can fix the trouble I'm having with the metrics file. Looks like every line wrote over the previous one so only the last line remains. Do you know how I can run the tests using the files below? 7-7.htm 0-0.htm Thank you all for your help, |
Thank
you for asking for help. Yes, I know how to fix your trouble and how to run the tests. You might consider taking your feelings about <helplessness> to Tribe as an entry point.
|
|||||||
| October
11, 2023 Wants to Guest Post Hey there, I wanted to reach out and see if your website accepts guest post contributions or link insertion in existing posts? If so, I'd love to hear more about your guidelines and any specific topics of interest. Thank you for your time, and I'm looking forward to your response.
|
Thank
you for reaching out to me about guest posting. You can find out about the guidelines at the Ground-Rules link above. You might also consider studying your candidate sites a little more before sending out email blasts. Also, you might consider using a more inviting salutation than "Hey there," or at least use the more respectful formulation, "Hey There." |
|||||||
| October
10, 2023 Wants to Visit Dear Ed, It has been a few months since I wrote to you the last time. I am still hoping to be given the chance to spend some time sorrounding you and thus wanted to ask again whether I could visit you and learn. Last time I mentioned that I was not readily available as I was planning a move from Munich to Hamburg and so on and so forth. I can understand that from such statements you might sense that I am not really prepared and ready to take on this journey. As one saying goes "If the student is ready the teacher will appear." Now I am surely a little closer and willing to come by from Germany at any time that suits you Regardless of a yes or no I hope this eMail finds you well and wish you all the best, |
Thank
you for reaching out to me. I wonder if you can tell me a little more about what specific goals you might have in mind for such a meeting. I plan to host a Tribe series in Austin and also consult privately. |
|||||||
| To Top of Page |
