Talon, I think that you did an excellent job of explaining the trade. It really is a very simple event driven trade developed from a top-down approach to seeking, verifying and exploiting temporary supply/demand shifts due to market/index mechanics. This index change phenomenon is very simple and well know by most people intimately involved in the markets. I wonder if you have looked at basing exits on the relative movement in the announcement-effective date period. On a seperate topic, I am hoping that you might describe an example of how you might perform the following on any system in general - "The basic idea is to disassemble the pieces of the system and test them separately."
The whole premise of this thread is wrong headed, not because of the methods used, but because of suitability for individual traders. Here are some questions: 1. Do you not think that mutual funds and hedge funds are already researching/testing/etc ideas and have teams doing just that? They still lose as a group. What are your chances against all the army of research staff doing that 24/7, and across thousands of funds? Do you honestly believe that the individual trade will beat all these thousanda of people if he does something similar to what they do? 2. If you agree with point 1, why then not just find the funds that has good ideas and good teams, and park your money there. The edge will then cost you nothing to get (or little. if you stay long enough in the fund) 3. If you want to have idea that is already turned to an edge, why not just buy it? There is one problem though. If you buy it from place that sells ideas/edges, then many people would buy it and the edge disappears. There is also the risk that the vendor has incentive to lie. So if you buy an edge, I would want to make sure to buy it from a little guy, and also do it in a way where the vendor is compensated only if there are positive returns. Guys: do not confuse analysis time/ideas with making money. The market does not care if you spent time analyzing. You play that game with your employer. With the market you can spend you time and effort, and on top of that you pay if you are wrong.
So are you a failed trader then?? Why do you trade then? Anyone who thinks that a hedgefunds size is its advantage does not have a clue what he is talking about, sure it can be an advantage from time to time when bullying stocks but for the most part they have thousands of traders cause they need to get in and out of billions of dollars worth of stock and that is a huge disadvantage. Trading does not necessarily have to be win/lose, it can be win/win alot of the time, funds which have to move tons of stock know that they are leaving money on the table based on slippage but they dont have much of a choice, there are all kinds of discrepancies you can take advantage because of this and if you dont believe that i dont understand why you would even trade.
How the f#%$ do you think buying a system is gonna work if creating your own will not work because thousands of traders at hedgefunds are researching stuff?
Good work. Exactly what time frame does your backtest reference? Also make sure you did not include deletions due to bankruptcies, mergers, etc in the sample. I have never looked at those really, so I dont have any intuition about what they would look like. As for the change between announcement and effective date... make sure you are looking at the close BEFORE the announcement. For instance, if the announcement hits the wires at 4:10 PM on Thursday, then Thursday's close is the reference point. If it hits at 3:55, then Wednesday's close would be the reference point. The second part of that is that you will only enter that trade on a limit at that price, which means a certain percentage of those trades will be "unable". So if you want to look at it properly you have to account for that. These typically make large moves immediately on the announcement or on the next open... and the idea is that you are not going to pay for a stock that just ran up 8% on this news... either it comes back to that entry price or you don't trade. Make sense? This is one way we trade around the concept and don't follow the system. We will work into some of those trades that you may mark as being unable, and we look at the stock/SPY spread and try to enter that spread at the entry price. You may never see a trade on the stock price, but we may have been able to do the spread. Keep in mind this is not a magic bullet... we will also have some losers that you don't have in your sample. The second part of the trade (effective - later) is very easy though. As for how you make sure you don't miss those... I guess filter the newswires. We have several people reading and watching news and between us we don't miss much. usually these come out after the close so you have some time to figure it out overnight. There are lots of ways systems and trading ideas can go wrong and i have seen far too many good ideas that are impossible to execute. People ignore this but it's a real issue... also don't forget the very presence of your order in the market will change the price. A guy wanted us to fund a strategy that traded the open of NYSE stocks... had an amazing backtest and he'd traded it with a few hundred shares and had good results (this btw is a good development track = idea / test / trade small size / deploy) but we wouldnt fund it because having the orders in the NYSE open auction would push the price of those opens around enough to kill the entire edge of the system. He didn't believe us and found someone with money who didn't know a lot about the market... ran the idea with $1M and lost significant money. So yeah you always have to consider can i execute at that price and how much size can I do? good stuff. thanks for the post.
1. I respectfully disagree completely with this point. You need to find the post I made (I think on this thread) about how you can win at trading because you aren't playing the same game as everyone else... that someone doesn't necessarily care of I buy here and sell 1.00 higher. Also... these funds can't beat the market as a group because as a group they ARE the market. Think long and hard about the irony of that... few people understand the significance of that statement. Mutual funds aren't that smart and most of them don't trade. Hedge funds do, but a lot of funds are poorly managed, etc. So don't be intimidated. Some of the smartest minds in the world do play in this business but I have known plenty of engineers / physicists, etc who understand everything anyone can know about the math, but they still can't trade. The biggest answer here is size. You can do things with your $5M account and make a good return that are impossible even with $100M... and at $500M and above the size becomes a serious problem. Yes, the individual has a chance. No, he doesn't if he follows the kinds of ideas that, in my experience, are popular on the internet. Planning to make a living trading S&P futures when it touches this or that band and the CSI or RSI is above this level blah blah blah. No chance at all. If this describes your trading experience to date and you aren't having success, then take a step back and think about what you're doing. 2. See #1 3. Ridiculous statement. I would never sell an idea that has an edge. At best, if I was a non professional, I would take my idea to someone and have them trade it for a share of profits. Finding a great idea available for 3,000 is ridiculous... doesn't happen. Also... if your argument is your original 1 and 2, then 3 is impossible by definition, no? I think trading is a lot harder than most of these forums make it out to be. I think your chances of success are a lot smaller than most people would have you believe... but it IS possible.... and it's possible if you find good ideas, have adequate capital, and develop the behavior that allows you to implement these ideas successfully. Figure it will cost you $50,000 and 2-4 years to figure it all out out, on average.
Talon, Just to make sure I am not mis-understood: I think you are an intelligent person, you are doing good work here, and you are sharing valuable knowlegge. I am running late to a dinner, but I want to highlight some points: 1. Funds as a group lose: That is the basis of my post. They lose, even if the make the research. So doing research is not a guarantee for profits. Why would then a lonely entrant to the field be expected to have higher chances? Some of that, you point out below, but I am not sure the typical average new trader understands it unless he has the experience, the energy and the intelligence you have. If that is the case he is either participating now in this discussion , or simply pouring over his analyses and exploiting the many pieces of insights that other write somewhere in here. 2. An advantage the the little guy has is his small size. I agree with this. It makes him agile, and he can trade on smaller time frames, but he needs to be skilled. If he is on a wrong side he can get cleaned quickly. 3. You are right on not to follow what othwers follows. Making money is a "lonely men" business. There is a need for a trader to think, be original, and carve a niche. I was more directing my post to the the individual traders who might under-estimate what he is getting into, and not to fall into the assumption: I will work and then I will make money. The funds highly educated, but yet loser analyst's, is proof that hard work and proper research, from a process point of view, while it might be necessary, is not sufficient to make money! I also agree with you on the number of highly qualified scientists who do not make money. Markets require from a person to shift on a dime if he is wrong, in the sense of recognizing a change fast, deciding quickly, and have no heart to the money already lost. Most scientists were trained to change only after they have done an analysis, and they may have psychological limitations because of lack of experience in situations where they lost and were proven wrong by markets. In the latter case, they have to deal with the biais from previous analysis, the money lost and not yet realized, and their ego. These guys rely on analysis as a corner stone, and they may not let go easily even if they were to blow out their accounts. If I were to hire these guys, I wuld make sure not to allow them to trade. The best of the analysts however can be a devastating money machine to the opposition, because of their superior knowledge like best analysts, and the quickness and lack of heart like the best risk-calculating gamblers. Best wishes!
Ok I think we've more or less exhausted that little index add/drop system. We haven't seen a complete backtest and you certainly should do you due diligence before trading anything, but I think all the information is there if someone wants to trade it. Let's talk a minute or two about the academic research. I think it's important for any serious trader to know the literature and to keep up with it, but much of it is very math heavy so it's not easy reading. This is a constant challenge, but academics do sometimes publish real, quantifiable edges. They also publish some garbage and most of the edges they publish degrade pretty soon after they're published. Still, keeping abreast of this literature helps us to better understand "how the market works", which to me is also a big part of why I trade. Don't get me wrong, not every idea has to have academic support... many I trade do not, but I chose some ideas that do have this support so you will be more confident you are not wasting your time on some silly idea I made up. So... Let's look at another idea. The reason I am choosing this idea is 1) it is simple and 2) i know it works and 3) there is a lot of academic research available to back it up. I also chose this idea because it is MUCH harder to backtest... and remember the point of this thread is that discussion. The idea is "earnings drift". When a stock reports a large earnings surprise and gaps in one direction (see RIMM and STEC for recent examples) there is a strong tendency to keep moving in the direction of the gap over the next 1-3 months, sometimes longer. Let's open the floor for discussion. What are the issues with testing this system? How would you set up a testing framework? Where do we start?
This thread had me thinking about other things I could possibly look at and surprisingly large movers after earnings was something I was thinking about. Back testing this requires a lot of research and manual work. To back test this first I would have to decide what would qualify as a surprise earnings gap. If I decided 5% was the number I would then go through the charts and find some companies that had surprise earnings which gapped 5% or more. I would create a list and map out the % it moved in the gap direction after 1,2 and 3 months. I would then use that data to decide if this was worth trading.