I was recently sent a copy of “Short Term Trading Strategies That Work” by Larry Connors and thought I’d post a review. The book covers many subjects that I’ve spent a lot of time looking at over the last two years. I gather that some of the book comes from previous articles posted by Mr. Connors on the TradingMarkets website – so for some it may be material that you’ve already read.
Overall, I found the book to be a pleasure. Well written, short and to-the-point. Most trading books I read, frankly, are fairly worthless because they don’t contain any testing. Mr. Connors makes it a point of discussing a strategy and then showing the tests. Obviously backtesting can only take you so far, but it is a good starting point. Most importantly, the book gave me some fresh ideas to explore – something everyone can use from time-to-time. While the book may cover too much well-tred material for advanced system designers, I think the book would be an excellent starting point for someone interested in learning about systems trading.
As a side note, I love the format of the book. It is a large, thin, hardcover book. Makes it a pleasure to read and reminds me of my childhood Tintin and Asterix & Obelix books.
In this review, I’ll go over each chapter in brief, but I won’t discuss the specific strategy rules as I think that would be unfair to the author. I will also be starting to confirm the tests outlined in the book, and post some follow-up on the systems and whether I was able to reproduce the results, and how as the strategy done recently.
Chapter Overview:
Chapter 1 – Introduction: Not much to say here.
Chapter 2 – Think Differently – Rule 1 – Buy Pullbacks, Not Breakouts: This chapter makes the convincing argument that buying pullbacks has, statistically, worked far better than buying breakouts. While this is not news to me, the chapter presents some compelling information.
Chapter 3 – Rule 2 – Buy the Market after it’s Dropped; Not after it’s Risen: Basically a reverse of the prior chapter that continues the mean reversion as a strategy argument.
Chapter 4 – Buy Stocks above their 200-day Moving Average, not Below: This chapter shows how buying stocks/ETFs above their 200dma has a significant advantage over buying stocks below their 200dma.
Chapter 5 – Rule 4 – Use the VIX to your Advantage…Buy the Fear, Sell the Greed: Presents a basic VIX system in outline form (covered more in later chapters). The basic idea is to buy when the VIX is stretched.
Chapter 6 – Rule 5 – Stops Hurt: This chapter presents compelling stats on how much stops of different kinds (time stops, % loss, trailing) hurt trading systems. And this is without a doubt true. One issue I have with this approach is that I’m not sure most people could survive the drawdowns caused by this approach. A second issue is that not having stops takes away some pretty attractive position sizing techniques such a percent at risk or an ATR-based stop. His argument would no doubt be that those hurt the overall performance of the system, but in my limited experience I’ve found you can get more return out a system using position sizing. You can also translate that into options position sizing as well.
Chapter 7 – Rule 6 – It Pays to Hold Positions Overnight: Here Mr. Connor points out, again, using statistics, that most gains are made overnight rather than intraday. Thus, he argues that one should hold overnight to maximize profits.
Chapter 8 – Trading with Intra-Day Drops – Making Edges Even Bigger: This chapter shows how buying on a limit price below the open price can significantly improve the edge of a system. Obviously, there will be fewer trades as the percentage below the open increases, but Mr. Connors shows how both the percent correct and average gain per trade goes up. Definitely interesting stuff that I will be using.
Chapter 9 – The 2-period RSI – The Trader’s Holy Grail of Indicators?: I hate the title of this chapter – whenever I see the phrase “Holy Grail” I immediately think to myself that it is about to fail. But regardless, this chapter provides a comprehensive look at the power of the RSI(2) indicator. I’ve covered this a fair amount on this blog, so I don’t think much more needs to be said. This chapter also includes the Cumulative RSI strategy that I’m going to be testing myself over the next day or so.
Chapter 10 – Double 7’s Strategy: Not much I can say about this chapter without revealing the strategy – it presents a good simple trading strategy for the SPY when it over its 200dma.
Chapter 11 – The End of the Month Strategy: Michael over at Marketsci has been covering similar territory – Mr. Connors presents a strategy for focusing on the end of the month as a system.
Chapter 12 – 5 Strategies to Time the Market: This chapter covers, not surprisingly, 5 strategies. It builds out the VIX Stretch strategy, and then showcases another VIX strategy, a TRIN strategy, another Cumulative RSI strategy, and finally a short strategy for the SPY. All are interesting.
Chapter 13 – Exit Strategies: Here Mr. Connors reviews different exit strategies and their qualities. These include: time-based exit, first-up close exit, new-high exit, close above a moving average exit and the RSI(2) exit. The author provides detailed analysis of each exit. One criticism of the book here: he does not include stats on the first-up close exit and the new-high exit.
Chapter 14 – The Mind: I thought this chapter would bore me, but I found it surprisingly interesting. Rather than going on with the classic “trading in the zone” approach that has been covered numerous times, the author structures the chapter as a series of questions related to systems trading. Example: “You lose money for eight consecutive days and you’re long multiple positions as the market is imploding. What do you do? Get out?” My reaction to many of the questions (not that one because I have an answer) was “hmmm….I don’t have a good answer”. So more food for thought. There is also a long interview with Richard J. Machowicz, a former Navy SEAL, that may be of interest to some people but not me.
Positives:
- Well written and easy to understand – even for someone without a lot of trading experience.
- Very interesting strategy ideas – gave me lots of areas to explore. Favorites: Cumulative RSI and Double 7’s Strategy. Also interesting: the chapter on Exit Strategies.
Negatives:
- The systems are a little lacking in detail. Are we entering on the close of the day the signal is generated, or on the open of the next day? While this may be somewhat obvious when looking at the charts showing entries and exits, I think beginning systems designers may find it confusing. The charts showing the sample trades don’t show dates – so if you’ve programmed the system and want to double-check the results, you have to do some guessing such as “When was the QQQQ between 52.50 and 52 on or around the 22nd of a given month?”
I actually find this a problem, in general, when people describe trading strategies in general. I’d love there to be some standard such as:
Entry:
- Indicator calculation (if appropriate): n/a
- Buy on: Open of the day following a signal.
- Position Size: Current Equity divided by number of positions
Exit:
- Indicator calculation: n/a
- Sell on: Open of the day following a signal.
Etc..
- Much of the work in the book revolved around stocks/ETFs being above a 200DMA – and thus, there aren’t a huge number of strategies that will currently work.
- As mentioned above, the systems don’t use stops. This may be unrealistic for many people, and limits your position sizing options. Simply saying “take your equity and divide it by the number of positions” keeps it simple but also gives no idea, for the trader, as to how much to risk.
Most of these negatives are pretty small – and I would strong recommend the book. For the beginning system designer, this book would be indespenible. For the experienced designer, you’ll probably find a gem or two that will spark some ideas of your own.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=747d319d-40c0-40e9-ae4f-c2fd9b553973)
This is a great book..I’m glad I found it. Connors is great……I also learned a lot about hedge fund trading strategies from 2 other great books. Hedge Fund Trading Secrets Revealed..by Robert Dorfman..and Richard Rms’ Stop and make money…..both are riveting and very informative. You should check them out if you like reading behind the scenes stuff about hedge funds and what methods they use to trade
good review, i have read the earlier edition before, and have purchased and bought trading systems from the connors group. on balance i think they do very good practical work that is simple and easy to apply.As long as the patterns persist, traders that follow their research will definitely make money consistently.
My major criticism is how the research relates to data mining—essentially this is the methodology used to come up with most of the trading systems without a lot of theoretical justification or integration. Much of the research works very well on large cap stocks and the s&p500 and the nasdaq, but would do poorly on the russell 2000, small cap stocks, and also commodities. They capitalize on the countertrend nature of these investments that has been induced by intense competition and increasing volume.
This behavior is due to the “averaging” or scaling process used to buy stocks by mutual funds, as well as the index arbitrage programs which focus on the S&P500 primarily . Note that these two factors largely do not exist in the commodity markets and for the more junior indexes like the Russell 2000. As a consequence they tend to trend much more, and fading strategies are dangerous. My concern is what will happens when we have a net money outflow from the stock market, and a collapse of proprietary trading desks and funds that focus on arbitrage (as we are having right now). Perhaps this means that large stocks and indices will trend more in the future than in the past.
In conclusion, looking for holy grail indicators with fixed parameters can be dangerous advice, and only an adaptive approach can save someone from abruptly changing conditions. Perhaps a crude method of doing this would be to use a moving average of the performance of a basket of countertrend based strategies, and sell when the equity curve goes below a long moving average.
Excellent comments – I agree that a single set of parameters is dangerous because the market changes over time. Rob over at Quantifiable Edge and Mike at MarketSci have both shown that the market rewarded buying higher highs until it didn’t – about 1980 if I’m remember correctly. Thus, any strategy that had worked on that property (trending) stopped working. Now we’re in a period of mean reversion, but who knows if that will be maintained. One thing that Rob looks at that I like is just running two simple systems – one that buys a higher-high and one that sells it. The result is that you may get a better head’s up when either one starts to fail. This approach is similar to your equity curve trading technique that you outline. Then the next question becomes what does one do to adapt or change a strategy. Obviously, you can have two different strategies for each market and trade them accordingly. Other approaches could be to adjust position size or adjust/extend the parameters of the system to require more extreme settings (in the case of, say, a mean reversion strategy).
All of these, of course, will suffer from data mining bias – which, in my opinion, is very tough to get around. But I don’t use strategies as a holy grail – I recognize that they fail (all the time), and have to be adapted, and, finally, at the end of the day, they really only offer probability – not an outcome.
I do find it interesting that you say that the methods don’t have a lot of justification for why they work. Two comments on this:
1. The market is always multivariate – so assigning a reason for behavior is, at the end of the day, just a guess as to why something works. The reality is that there are probably hundreds of reasons why the market behaves as it does – therefore, for me, I don’t find it that interesting to look at. The simplest form of this is at the end of the day, you’ll have CNBC give you the reason why the market was up or down. It’s usually one reason. It may or may not be right or wrong.
2. You give a great reason (a few actually) for why the systems might work. So there you go! You’ve got it wired already.
At the end of the day, financial instruments either trend or mean-revert – the trick, in my opinion, is figuring out which they are doing and then design systems to trade it.
Thanks very much for your thoughtful comment – great stuff.
Dear Sir,
Where I can free download “Short Term Trading Strategies That Work” by Larry Connors.
Regards
Anand
Uh, you can’t. You have to purchase.
I received the book on new year’s eve and have coded up three long-only strategies using Profunds ULTRA ETF’s using Double 7’s (also in TASC, January 2009), RSI2 and Cumulative RSI2 in each of the strategies. Of course the rules in the book didn’t work and I had to develop other rules, but I did get some very strong performance numbers using EOD numbers and buying and selling on the open of the next day for 2008 with April and July through December being very strong.
These strategies are definitely worth pursuing if the market continues mean-reversion with lots of volatility. However, tuning these strategies is difficult since there is so little data. I decided to tune on three ETFs and then test across all of them for returns. Not quite how I learned to do back testing (prefer to take my historical data and split into two sets of data — one for tuning and one for evaluating the tuned data).
I’ve yet to figure out a short strategy that works using these indicators. I was hoping to develop a strategy (long/short) using the following ETFs (UCD, UCO, ULE, UGL, EWV, EFU, EEV, EWV, AGQ, YCL) which don’t have bear counterparts.
Anyways, I am very happy to have found this book, use his ideas for fodder to developing my own systems, and try to apply these techniques. However, the proof in terms of dollars earned on the strategies has not arrived.
Interesting comments Mike – I coded up the strategies as I was reading and found that most of them were within the margin of error in terms of performance.
In terms of the short side – that was one of my points that the book lacks a lot of short strategies. Now having said that, others have suggested using short ETFs to get the job done. My issue with that is that short ETFs are essentially broken – so that is unlikely to work.
Finally, as I mentioned – I think it is a great book for idea generation – I don’t think any of these systems should really be used straight out of the book. It is also a good book in terms of getting your mind around mean reversion.
Thanks for the comment!
[...] trading strategies…” As one would expect of a simple trading strategy published in a popular book, the Double 7s strategy has been discussed on various sites and in the [...]