Curve AdvisorKeymasterJuly 13, 2015 at 7:40 amPost count: 612
I wanted to expand on the CME article I wrote on directional trading. Here is the link: http://www.cmegroup.com/education/files/curvature-trading.pdf.
But before I go on about why I prefer doing trades with non-linear payouts, I want to discuss Bill Gross. In April 2015, he stated that bunds were the “short of a lifetime.” Well, it turns out that rather than being short bunds, he had a short straddle position in bunds vs a short straddle position in Treasuries, where he was short less puts in bunds than calls in bunds and puts in Treasuries. So while he had a short bund position in options relative to treasuries, he did not benefit from the bund selloff because vol exploded. In June 2015, he stated that the Shenzhen was the next “short of a lifetime.” He had no short position in Chinese equities when it came crashing down. Instead he had a short in US equities and some emerging market currency positions, which he claimed worked. Perhaps – but I’m sure they didn’t drop 30% like Chinese equities.
In his mind, I’m pretty sure what he thought he was doing was clever – choosing some structures he felt were better risk / reward ways to express the same view. In his own way, he was choosing what he probably thought were trades with non-linear payouts. I suppose I can give my two cents on what I thought his thinking was, and why it made sense to to do (if there is reader interest). But I mostly wanted to highlight what I think is a mistake in directional trading, even by a very experienced trader.
My number 1 rule of directional trading is – if you have a strong directional view on something, express at least a small percent of it in the most direct way possible. In BG’s case, it would have been to sell bunds and sell Chinese equities.
This is probably not what you were expecting from the Curve Advisor.
[to be continued]
Curve AdvisorKeymasterJuly 15, 2015 at 3:26 pmPost count: 612
I’m a big believer in diversifying one major directional views. This is because there are pros and cons to the various ways of taking a directional view. Below are the five major ways of taking a “directional” view in interest rates. I just made this list up, so I may add or modify as necessary:
(1) Outright directional trade: A single contract position in the part of the curve that best expresses the view.
(2) Slope or curvature trade: Any time you have a move in one part of the curve, there will be an impact on the slope and curvature around that area.
(3) Options trade: Options allow you to express a delta view, but you may also have a view on the gamma, theta and vega.
(4) Secondary rate market trade: Depending on the cause of the move, sometimes when you have a rate view on EDs, there may be times when you prefer putting on a trade in another rate market (say BAs) that may be correlated.
(5) Non-rate market trade: Depending on the cause of the rate move, there may be impacts on other markets, like equities, commodities and foreign exchange.
(6) Cross market rate trade: Taking the outright directional trade, but doing an offsetting trade in another market.
After reviewing the cause of the directional move, think about the impacts in these other areas and see if it makes sense to consider alternate positions.
[to be continued]
Curve AdvisorKeymasterJuly 20, 2015 at 12:03 pmPost count: 612
You may be saying to yourself, if I have a strong directional view, why wouldn’t I just do 100% of the trade as a single contract trade? The problem is with the P&L noise. It rarely happens that you catch the absolute top or bottom of a huge move. You may be able to pick an occasional top or bottom at some technical level (for a modest gain). But if you wanted to say, pick the top of the Chinese equity market in 2015, it’s somewhat unlikely you would have sold at a level where you would not have gotten stopped out (multiple times). Some macro views could take months (or even years) to materialize. You never want to manage your P&L so that you get stopped out on a trade you really believe in.
Your only alternative when trading a single contract position (and not get stopped out easily) is to size the trade absurdly small. I suppose you could also do some kind of momentum trading in large size when you think the timing is right. I’ve just never seen it done consistently well by anyone I’ve worked with. And all it takes is that one year of choosing badly, and you get canned.
What you ideally want is a structure with nonlinearity – something where you can pick up value when you are right, but not lose value when you are wrong. I’ve listed some of the other ways you can do a “directional” trade in the previous post – you will be find other ways of expressing the trade. You will need to determine what version(s) will provide the best risk/reward on your view.
Curve AdvisorKeymasterJuly 22, 2015 at 10:54 amPost count: 612
I’ve sat with dozens of traders with different styles. I’ve really only seen two guys who have “consistently” (mostly) been right on their macro directional calls. They both started their own hedge funds, and they both closed for poor performance. Interestingly, one guy was long Chinese equities (a long time ago), but had to stop out on a dip before the subsequent spike. The moral is… macro directional trading (not high frequency directional trading) is going to be volatile. The problem with relying solely on directional trading is that you will eventually have a “bad” run, which could turn into a bad year. And in trading, you need to be profitable “almost” every year.
That having been said, I do think there is a place for directional trading in your book. In my best P&L year, I made 40% of my revenue from directional trading. So I basically grinded 12 months to accumulate 60%, and in a few really good weeks, I made the other 40%. That makes directional trading pretty sexy, and the key for me was that I had a consistent revenue base (curvature) to be able to put myself in a position where I was comfortable on the losses to take those big directional bets. But I didn’t take a purely single contract directional positional – but I did have some on. One of the two guys also had some consistently profitable P&L (mostly FX and rates carry trades) which he used to take 1+ year macro positions (many times single contract positions) and he was also well-diversified. The other guy was the head of the group, so he basically cherry-picked everyone’s best trade ideas but had a great sense of timing. This is my anecdotal spiel on some different ways to trade directionally “successfully.”
Directional trading has its uses, but if you want a long career in trading, you should try an pick up as much nonlinear protection you can. Don’t just think in two dimensions (up and down) – that “up and down” is going to affect other structures and markets, and some of those may be better risk/reward trades. So take a good look around.
Curve AdvisorKeymasterJuly 23, 2015 at 12:08 amPost count: 612
I usually try and find a “better” way to take a directional view, so that results in having a portfolio of trades to express the core view. Sometimes when you get a little fancy (as Bill Gross can attest), the “better” trade may not perform as well. So I usually take a view on the outright (usually much smaller), assuming an outright makes sense to do. I do much larger size in what I think is the best view. Sometimes, there may be multiple “better” nonlinear ways to put on the view, and those trades may not have very high correlation to each other, which makes it easier to do various versions in good size.
Plan ahead and think about what you expect to happen at various points in time. Depending on the markets, you may want to get into and out of a piece of a position quickly. You just can’t beat the liquidity of a single contract position, and you should size the single contract position accordingly. You may also use the technicals on the single contract to trade not just the single contract trade, but at times, the non-linear trades as well. Just keep in mind the nonlinear trades may take longer to unwind. I usually unwind the versions of the trade that have outperformed on the move, and this allows me to hold (or sometimes even add to) the versions that have underperformed.
This is just my trading style, and you should pick out whatever elements make sense for you and your particular view.
Curve AdvisorKeymasterJuly 28, 2015 at 12:45 amPost count: 612
Just because I have a directional view doesn’t mean I won’t accumulate good risk/reward trades that subscribe to the opposite view. For example, just because I am bullish rates does not mean I won’t put on a bearish rates position – PROVIDED that the bearish trade provides excellent risk/reward. This may seem strange to a lot of people. But it all goes back to #1 on my list of Trading Philosophies (see the “My Trading Philosophies” thread): a 2 to 1 bearish reward to risk trade plus a 2 to 1 bullish reward to risk trade equals 1 unit of profit regardless of whether we rally or sell off. I’m just all about collecting “value” on the curve, regardless of where it is coming from.
What’s interesting about this approach is that if I have good bearish reward to risk trades in my book, that allows me to size larger in my core-view bullish trades! Say I was willing to risk 4 units on my core bullish view. If the portfolio of bullish trades was 8 to 4 reward to risk on a rally, having on some bearish trades that offer 4 to 2 reward to risk on a selloff gives me the following profile on the entire package of trades – on a rally, we make 6, and on a selloff we lose nothing. Well, if we lose nothing if we sell off, you could easily do more of your core bullish trades!
Curve AdvisorKeymasterMarch 10, 2016 at 2:30 pmPost count: 612
Of course, using this type of strategy depends on you work environment. Some work situations are more friendly to structured trades than others. So you need to consider your particular situation as to how the criteria for your compensation (reward) and the criteria for how your risk is measured. For example, when I worked at JPM, the main risk criteria was VaR (based on the worst 7 days in the past year). The risk measures/constraints may be different, and you need to maximize your expected P&L relative to those constraints.
A second thing to consider is that the more “relative value” your book is, the more you may have some trades that don’t move in lock-step with direction. So you may have (for a little while, at least) P&L noise associated with the portfolio of trades. If you are unlucky, you may even have a trade that moves in the opposite direction of what you were planning! So you need to factor that into any sizing calculations. This is one of the reasons, I like expressing a directional view in multiple ways (including the simplest way – and outright long/short).
Curve AdvisorKeymasterNovember 17, 2016 at 6:54 pmPost count: 612
If you are bearish, then there are two things you could do… play for more frequent hikes in the same time frame (current flies go higher), or play for the hiking cycle to be longer (flies further back go higher relatively – in other words, play for the flies further out to be the high ones).
You must be logged in to reply to this topic.