Two subjects follow: FIRST: I slipped into pairs trading in 2008 not even knowing it existed as a trading form. I had been day trading since early 2007 GS as the axe and Merill Lynch and Lehman as followers. I plotted all three on a normalized graph each day with their common starting point at the zero percent point on the left vertical axis. The chart was structured not by price but by percentage gain / loss (as would be obvious). I only traded GS and one of the others if they were both (the axe and the follower) above the 0 per cent line. When GS dipped to a level one-half percent below its most recent peak, I went LONG GS not giving a flying flip how much momentum it had downward. At that time, I also went SHORT one of the other stocks that was following. Invariably, GS would reverse to the up while the other stock continued to decline. I seldom experienced a loss on those trades ! The biggest short coming I had was when to close. It was more guess work when to close --which when you consider I almost always closed with a gain-- illustrates the merit of this style trading.....the sizable unbooked gain while in the trade had sizable margin for guess work approaching contemplation of closing the trade. This was short lived when Merill and Lehman blew up. I was on the hunt for a replacement......looking for another axe and followers. My point is (1) this is a powerful style of trading (I developed for myself) and was derived from my day trading DNA I still had in me from my robust day trading days of the later 1990's. (I did no trading whatsoever in anything for 8 years from 2000.) Again, I want to point this style out to you if you want to explore it. I would really, really, really love to do it again. If you are a day trader person, this is one to explore. Like what followed in my life (this was a self invention for myself) intuitively I figured others must be doing something like this ( ? ) SECOND: In my search for substitute stocks to GS LEH MER, I slipped into my next self discovery which was "inventing" pair trading, not knowing there was a universe of existing pair traders. It was a very different style for me, much more conservative, and much less time on the monitor, the latter I really appreciated. I subsequently discovered this thread and was tremendously excited. (I had just barely known that ET existed.) But what I inferred after reading a large number of posts is that trades, although not necessarily always day trading, were generally fairly short lasted. It appeared to me software lent itself to this as I also inferred. (I never subscribed to software.) I continued to trade through 2009 with my own style which was crude by comparison from the more sophisticated posters I was reading. My most compelling reason for staying with my style was the fact that I almost never experienced a loss out of 45 trades. (This does not include trades I went in and out of quickly because I did something wrong in clicking buttons carelessly. I eventually made myself a check list. Boring but effective.) To be honest, I was prejudiced to get out of all my trades with a gain because I wanted this to be the finding of the holy grail which chronically got me out prematurely. So one would not be impressed with absolute dollars made. (Although the annualized returns on the majority of trades were impressive. I had to use annualized measurement because I was frequently moving money in and out of that trading account.) What WAS impressive was where the trade I had gotten out of the vast majority of times would have gone two or three times better than my booked gain. Thank you, that was just as important to me as booking my little gains. It was 2009 as an acid test for rating this newly self-stylized trading that proved itself as safe and consistently successful. I had no compelling reason to the faster moving, software dependent trading that otherwise would have been available. I particularly like the SAFETY and the slow, leisurely pace of this style. I have learned discipline and my trades are consistently better than those early years for dollar measured gains. Without software, however, it is hunt and peck for the trades, but I find free resources offering pairs that I pick up on, examine, and many times use. I have started a small group in which we are dedicated exclusively to this style and nominate to each other trades. It is the turtle in the race with hare for style when comparing the faster moving pair trading I see elsewhere. It offers the prospect of putting on a number of small trades simultaneously and easy to manage. Slow and disciplined keeps my heart rate down and frees up time for other things in life. Certainly, a faster moving trader can participate in learning this style (very small learning curve) and try some trades easy to manage while continuing his traditional style. At the end of many months doing this, one can easily be in a place to judge its merits in his trading life.
This is an update from the Dec 3 update on the ripping trade of AET CI. Trade was initiated Nov 15. (update posted on Charts of Note where all my charts are posted.) http://www.elitetrader.com/vb/showthread.php?s=&postid=3709281#post3709281
Congrats on what sounds like positive results but I wouldn't get too comfortable with the "SAFETY" aspect of pairs trading. There are lots of hidden landmines just waiting to inflict damage. A recent but fairly benign example is NYX from Thursday. I was very fortunate to be on the correct side of my NYX/SCHW pair (I was long) but I could have just as easily been on the short side. The buyout premium was pretty small; only 30% when sometimes it is more than 100%. Unfortunately, buyouts usually do not favor the Pairs trader because inside information is typically leaked out and the stock runs up. The unsuspecting Pairs trader shorts the stock because it has run up to a level that is normally âtoo highâ. Some claim that going long only the smallest cap stock of the pair will help mitigate this risk but there are plenty of cases where this would not have helped. These landmines can wipe out months of gains and it can takes months more to recover from the injury. The same is usually true on the downside e.g. the stock drops âtoo muchâ on no news, the Pairs trader goes long and then the really bad news hits! Another decent pair was HLF/NUS but the results have not been pretty the last few days. Of course, one should always avoid questionable companies like HLF but it's very hard to know if a company is real or just Memorex. My intent is not to sound like a Doom and Gloom trader but knowing the potential risks and preparing for the inevitable losses is the only way to survive the trading game.
Shame on me for not placing THE adjective I have been trying to apply every time I use the word SAFETY in the context I used it. If you go back for about a month, maybe more, my posts have been starting to cultivate a new habit of joining to the SAFETY hip the adjective RELATIVE. Thank you for calling me on it. I don't take the omission lightly. Having said that, I would assume every reader infers the word relative including you. However that is no excuse for not using the "relative" adjective. Readers don't know me and most have not read my series of posts. Among that population will be many who I presume would NOT necessarily infer I meant "relative" safety." That is why I say there is no excuse for not me not using both words ALWAYS together in the context of describing my style. Good call S2M. In what I did making 2009 my test year for my new found manner of trading--pair trading--I had exactly a similar illustration as you narrated. It was two banks I paired and one morning I was aghast at one of them having announced some sort of restructuring that I didn't understand nor did I care. I found myself on my fanny with a blow out on my hands, and no amount of adding layers was gonna bail me out. I closed the trade which neutralized a number of previous gains. This brings up in the same breath as "relative safety" the understanding there is always risk in trading. After 2009 which proved my style of pair trading to myself, I experimented with some deviations on what otherwise was disciplined trading. One had my first level get out of hand on me. I believe I put on four levels all together. This was an unusual occasion where I watched very, very closely to the trade for many days to take advantage of movements that allowed me the benefit of those additional levels....booking gains on the 4th level and the 3rd level and subsequently adding them back on again and trading them. Altogether, after all the dust settled, I got out net ok, but (1) the first level never recovered for me, and (2) I was tethered to the monitor for a long time watching for opportunities to extricate myself with sub-set trades using those additional levels. But being tethered for days to my monitor WAS NOT MY STYLE since the day trading days of yore, late 1990's. (By the way, see my next post about why I declare my style of pair trading as a wonderful approach using this very example as a model.) So risk comes from more than the exogenous event; it also comes from having to work harder on a trade at times. I could have just bailed on the first and second level (applying some notion of stop loss), but I worked the trade so I ended much, much better than that. I don't like working harder just as much as taking a loss on a first level. Working hard (for me these days) is a loss, too. It is not only the "work" but also to have to tie up capital that could be used more productively else where.....The notion of lost opportunities. Thanks, S2M.
(The above quote is from my last post.) Relative Safety to a trade: There are times I find myself adding a third and then a 4th layer (the latter infrequently). I do this cautiously by double checking previous work and looking for the obvious driver/s that may be propelling the widening spread in this trade I already am in. Once content, I confidently add those layers. And here is the point: I am a poster child for Reversion to the Mean. I do it to the extreme. Extreme = lots of history. I not only want to know how much that rubber band is currently stretching relative to where the pair plots have been very recently, but also recently, also not so recently, and also farther back. With that examination, what I am looking for is to initiate a trade and one that later I can confidently add additional layers relatively safely.
My first pair posted in 1 1/2 months. Only for the aggressive, though. HOT / HST Earnings coming out on both soon. Recent long term parallel feature makes it dicey for swing traders to stay in for very long. This adds extra risk to future added layers. 4 charts of varying time lengths posted at Main / Trading / Charts of Note http://www.elitetrader.com/vb/showthread.php?s=&postid=3737253#post3737253
Theres alot of discussion on the entries but very few on when a pair is out of weck and should be removed. What criteria are used on removing out of weck pairs and which are more effective? Any veterans contribute?
Probably not the answer you are looking for, but for my style there is lots of discretion when to get in and even more when getting out. There is no analytics involved with my style. It is more an art than any tripping thresholds signaling green lights and red ones. I have followed only some of those that do, and I am not impressed. But then again, mine is a different approach, and I'm probably wrongfully comparing apples with oranges. My challenge is making a program that can screen for an "art."
See NEWS on these chart series with today's post of current chart. Decent trade if you were in it ! Hold position or take profits ? Note the considerations (1) posted on today's chart and (2) advisory in red below chart ! http://www.elitetrader.com/vb/showthread.php?s=&postid=3745842#post3745842
HOT / HST update to Feb 5 original posting advisory on this pair. Very nice trade if held on to the trade thru Feb 20th. http://www.elitetrader.com/vb/showthread.php?s=&postid=3746823#post3746823 (This posting is a re-post from Feb 20th at this forum as the original Feb 20th post here fell thru the crack as ET changed to its new format and back to the old, and it lost this post.)