Really excited about latest book by Jack Schwager. "Unknown Market Wizards: The best traders you've never heard of". Expected to be out by Nov 3. A preview of some pages is already available here. Besides the obvious reasons - it will be featuring my friend and a mentor, Marsten Parker. He is also the creator of RealTest, backtesting software I am currently using. Marsten was one of the founders at a tech company that IPOed on 90-ies and after successful exit got into trading. As his company pioneered automated testing software he had a brief interaction with Jim Simons, who apparently "didn't like bugs in their trading software", and Marsten's team was consulting them on better practices. After getting into trading he used his tech skills and quickly realized he wanted to stick to systematic and later mechanical approach. On the way he developed RealTest to test his ideas and been basically iterating on it ever since. Long story short, 22 years later, averaging 20% annual return trading only mechanical systems, he is now, for a first time getting some publicity. Which I always thought he deserved taking into account combination of his performance and longevity in the market. But he just doesn't like going around and bragging about his trading. While still actively trading he is considering releasing RealTest to a wider audience. Hopefully soon! If anyone is interested what it is about here is a website - https://mhptrading.com/realtest/ Val
Wow, didn’t know Jack was releasing another! huge fan of all the wizard books! I got em all, surprisingly my favorite of them all is the hedge fund one, lots of gems even for the retail trader in that book!
Thanks, Val! I have enjoyed our correspondence and collaboration over the past few years and am impressed with how quickly and successfully you have pivoted into the systematic approach. And I very much appreciate all of the encouragement and feedback that you've given me about the software. Keep up the great work!
I just started to read your journal here and it looks great. Thanks for sharing. I hope your journal helps you just as much as it helps other traders. I read that you started to trade with Amibroker (which I use myself) but now you've migrated to RealTest. Is there any particular reason that you switched? Do you find any advantages with RealTest? I'm running some mean reversion systems myself on the American markets where I typically trade on the open, relatively large numbers of low volume, low priced shares. What works for me right now with Interactive Brokers, is to buy or sell with Midpoint pricing with the default adaptive algorithm so I'm not flooding the market. On average, that seems to get me close to the opening price. I've also recently set IB for Tiered Commissions, otherwise I'll have to pay far too much when buying or selling large numbers of shares. I'll see how that works out. I also read that you're from Canada too. Have you been getting away with paying just capital gains tax on your frequent trading? I've been paying capital gains only on all my trades so far. I don't do any short selling or trade any futures. I try to keep my most frequent trades with my RRSP accounts as I read that Revenue Canada doesn't care about them too much. They know they'll get their tax anyway when you make any withdrawals.
Bensdorp seems like a scam artist but I haven't read his books yet. Whenever someone tells you they were this big shot hedge fund manager and now are "mentoring" complete newbies, alarm bells go off.
Perhaps. Other ET members messaged me about their suspicions too. I personally don't know him and his backstory does seem a bit ridiculous, but I did like his book And it does outline similar approach I use for trading. The completeness of the rules for the strategies he gives is pretty exceptional comparing with other authors. Emmett Moore, founder of the tradingschools.org is pretty controversial figure himself. While I enjoy reading his reviews I became somewhat sceptical of them too. Val
Thank you Sam, those are very thoughtful questions. When it comes to AmiBroker vs RealTest - AB is super flexible, too flexible. Short answer is - In RT my strategies code is way shorter / simpler / easier to read => has less bugs For the type of ideas I typically test it much faster for me to write code in RT RT is designed around modeling multiple strategies all together. Which is super important for me AB is a truly great product for pretty much anything backtesting related. But it makes it a bit too generic so too many things need to be described to model some pretty typical strategies, that leads to more code, which leads to more support and a greater possibility of making mistakes. AB pretty much completely lacks support of testing multiple strategies together. It goes as far as long+short which is fairly primitive way of doing it and if you want to model more than that you really need to get deep into coding and even then it is cumbersome. On the other hand RT doesn't have ambition to support any type of backtesting possible. It is more of a niche product. If it works for you - it will offer unique powerful features and save tons of time on testing. In some of my earlier posts I included screenshots of how RT strategies code look like, check it out. I think they speak about the difference quite well on their own. My single stragy code with RT is typically half a page long to give you an idea. Unfortunately I have to pay full tax on my trading profits. Which is crazy considering my tax bracket. My volume of trades is too high to be unnoticed and a big chunk of my profits comes from shorting, so there is not much choice. So I run it as a business but without incorporating. At least I can offset a bit with subtracting costs. Val
Thank you so much for your reply. I think you do some mean reversion trading with multiple positions using daily bars? Are you using limit orders? I find most mean reversion systems work better with limit orders and they are cheaper and easier to trade with Interactive Brokers. With Amibroker it's somewhat complicated to code and backtest limit orders when you run multiple positions. You pretty much have to use a custom backtest for that. Can you code and test for limit orders for multiple positions with RealTest?
true, but not all mentors are scammers. If you have any sense of aura you can tell from the onset if someone is selling snake oil, or legit shit. It all depends. I have many books I’m reading rn, and Bensdorp is one of em, I’m on chapter 3, so far I haven’t sensed anything fishy.