Back in March I wrote an article called “Good Days for T-Bonds” that basically highlighted the fact that the last five trading days of the month have historically been the best time to be long t-bonds. And so of course, Murphy immediately invoked his pesky rule and the last five days of March showed a fairly sizable loss. But I refuse to surrender. So let’s take things one step further.
Jay’s Seasonal T-Bond System
If A) the Current Month is May, June, August or November AND the Current Trading Day of Month is 1, 2, 3, 10, 11, 12,
OR
B) If today is one of the last five trading days of the current month then…..
Trading Signals
*If A or B is true then hold a long position in T-bonds.
*If neither A or B is true then hold no position in T-Bonds.
So what the heck does all of this accomplish?
Results using T-Bonds Futures Daily Value Change Data
To test this “system”, we will do the following. If indicator A or B above is true then we will assume that a long position of one t-bond futures contract is held that day, and we will add (or subtract) the daily change in the nearest futures contract month for that day (a 1-point change in price is worth $1,000).
*No slippage or commissions are deducted from these results.
In Figure 1 the blue line represents the cumulative gain in t-bonds during the “Bullish Days”.
In Figure 1 the red line represents the cumulative loss in t-bonds during “All Other Days”.
Figure 1 – Gain for t-bond futures during “Bullish Days” (blue line) versus “All Other Days” (red line); 12/31/1983 through 4/17/2015
For the record, the “Bullish Days” registered a gain of $234,000 while “All Other Days” registered a loss -$117,500.
This disparity in results is what we “Highly Quantitative Analyst Types” refer to as “statistically significant.”
Using ETF ticker TMF
Ticker TMF is a triple-leveraged ETF that tracks the daily change in the long t-bond times 3. For the record, triple leveraged ETFs are NOT for everyone. To trade without leverage consider ticker TLT. Nevertheless, the results in Figure 2 show the growth of $1,000 invested in ticker TMF only during “Bullish Days” since TMF started trading on 4/16/2009.
Figure 2 – Growth of $1,000 holding a long position in ticker TMF during “Bullish Days” (blue line) versus “All Other Days” (red line); since 4/16/2009
For the record, $1,000 invested only during the “Bullish Days” grew to $5,384, while $1,000 invested in TMF only during “All Other Days” declined to $310.
So for the record:
*TMF during “Bullish Days” = +438%
*TMF during “All Other Days” = (-69%)
*starting on 4/16/09
**through on 4/17/15
Ticker TLT is a non-leveraged ETF that tracks the long t-bond and trades much greater volume than TMF. TLT also enjoys decent option volume. So one other strategy that a trader might consider (although this is suggested as “food for thought” and not as a recommendation) is buying call options on TLT rather than buying shares of TMF or TLT. By doing so a trader can commit a great deal less capital and therefore enjoy less dollar risk during those times when the so-called “Bullish Days” don’t pan out.
So let’s look at an example of buying a call option on TLT as a substitute for buying shares of ticker TLT. For this we will use www.OptionsAnalysis.com. 1/23/2015 was the sixth day prior to the end of the month so we will buy a call option on that date and hold it until the close on the last trading day of January. One way to find a trade using call options is the “Percent to Double” routine. The input screen appears in Figure 3 (you can click on Figures 3 through 8 for a closer look). 
Figure 3 – Input screen for “Percent to Double” routine at www.OptonsAnalysis.com
The output screen appears in Figure 4 with buying the March 136 call highlighted as the top suggested trade.
Figure 4 – Output screen for “Percent to Double” routine at www.OptionsAnalysis.com
Figure 5 displays the particulars for the trade. We will buy 2 call options for a cost of $524. This trade will have a “Delta” of 87 (which simply means that it will react similarly to a position of buying 100 shares of TLT).
Figure 5 – Long 2 Mar TLT calls: Cost $ and Maximum Risk = $524; Position Delta = 87 (www.OptionsAnalysis.com)
Five trading days later – at the close on 1/30/2015 the options could be sold at $4 a contract, for a profit of $276 ($4.00 – 2.62) x 100 x 2. This is a 52% return on the $524 of capital actually commit to the trade as seen in Figure 6.
Figure 6 – Long 2 Mar TLT calls – 5 days later (www.OptionsAnalysis.com)
To see how this compares to buying shares of TLT, see Figure 7. In order to get the same Delta as being long 2 March 136 calls (i.e., 87) we would buy 87 shares of TLT. With TLT trading at $134.77 a share, it would cost $11,725 to buy the 87 shares – versus just $524 to buy an “equivalent position” in the March 136 call option. This trade appears in Figure 7.
Figure 7 – Long 87 shares of TLT at $134.77 a share (Cost: $11,725) (www.OptionsAnalysis.com)
As you can see in Figure 8, five trading days later this position could be sold at a slightly higher dollar profit of $306. But remember, the trader who bought 87 ETF shares had to put up $11,725 of capital in order to make $306 while the option trader put up (and risked a total of) $524 in order to make a profit of $276.
Figure 8 – Sold 87 shares of TLT 5 days later (www.OptionsAnalysis.com)
Summary
So will the trading days highlighted in this article generate superior returns ad infinitum into the future? Ah, there’s the rub. The truth is that there is no way to know for sure. Still, if you had to bet……which of course, you don’t have to do. In any event, the long-term difference in performance between “Bullish Bond Days” and “All Other Days” is pretty stark so there may be something to it.
Also, the other part of the lesson is that are ways to use options to maximize profitability while simultaneously limiting risk.
Jay Kaeppel
Very nice analysis Jay.