Jason,
I am currently short ZB and got an email from TOS saying I need to close or roll the position next week. I want to roll - exactly how do you go about doing that? Obviously I could just buy to close my March contracts and sell the next cycle but that might result in bid/ask slippage and random fluctuations in the price between the closing and opening of the positions. Alternatively I could try to outguess the market and close March at a perceived low and then sell the next cycle higher, but this may or may not work. My question is if I just want to maintain my same position, is there a more efficient way to roll it out at my same prices and avoid the scenarios above?
This is the dilemma of a longer term position in futures. Rolling is just what you've described, you just have to close out the current month and open in the next contract. So you either have to give up some slippage in order to roll smoothly with no gambling, or do what you've suggested and try to leg out of one and in to the other for a net zero or even positive roll.
ReplyDeleteHowever, in addition to this dilemma you've also got contango or backwardation risk. For a short ZB position right now you get screwed a bit as the JUN contract is currently trading for more than 1'00 point lower than the MAR. This is why some people prefer to use TLT or TBT instead, but the misnomer there is that the managers of those funds have to roll futures contracts as well so you pay the piper either way. You just don't have to do the actual rolling if you're using an ETF instead of the futures.
One thing you might consider to help your cost basis out on the short ZB roll where you lose a point is to sell an OTM put.
Is this contango thing a true loss or gain or just more of an accounting thing? In your example above, if you do the roll you will be selling JUN 1 point lower than MAR, but when you close JUN won't you be buying 1 point lower also so it nets out to zero?
ReplyDeleteTo answer the second question, you don't know what price you will close the JUN short so you don't currently know if that trade will be a profit or loss. What if ZB keeps going up? You have to look at this as two separate transactions. The industry calls it rolling contracts to stay short the same position, but they are entirely two different positions. If you choose to get short the next contract at a different price then so be it.
ReplyDeleteMy comment about losing a point is in your personal cost basis to profitability should you decide to get short the next cycle contract. Let's say your original short was 140 but you've got to roll today at 142, you lose 2 points on that trade, it's over. Now you short the next contract currently trading at 141, 1 point lower. If you close at 140 which was your original short price, you've lost a net of 1 point for the two trades. So to the people who have been short since last July 2011 like I was at 127ish, because of contract rolls you would need a price target of something near 124 on the current JUN contract just to break even. Although the differential in prices when it came time to roll in November on the DEC contract was only 8 ticks.
I understand that you can look at it as two separate transactions, and ZB can keep going up, but even if this is the case I am not convinced that you are really losing anything on the roll except for transaction costs and possible bid/ask slippage. ZB might go up or down from here, but if the cost basis of the Jun contract shifts up or down relative to the March contract I think it evens out in the end. I think Tom Sosnoff may discuss this on tastytrade this week as I asked him about it and I think it is a point of confusion for many, especially me.
ReplyDeleteFor me, the "losing" aspect is mainly psychological as if I'm short and the price to roll is always not in my favor, then when I get back to near my entrance price I'm still not back to even. However, this works both ways and I do remember rolling once and was more than happy to enter in to the new contract at a higher price. I've also noticed in the time I'm been trading ZB the last 18 months that 100% of the time the gap from the roll gets filled, so I'll keep an eye on it again this time. I just don't want to go long at these levels to try and play for that gap.
ReplyDeleteI was just a coffee shop listening to an older episode of tastytrade and they did discuss this a little bit but I'm still not clear on it. They clearly said that the difference in price is the cost of carry and that it has no bearing on your P/L. But then they said something to the effect that the cost of carrying is the cost you have for controlling a $100,000 for $5000 - so maybe there is a real cost associated with just holding/rolling the position.
ReplyDeleteGo to the Feb. 14 Get tasted episode, time about 188 minutes for the discussion. I've listened to it a few times now, still not clear to me though.
ReplyDeleteOk - maybe I'm starting to get it. They seem to be saying if you are short ZB and if ZB Mar is 143 and ZB Jun is 141, then the act of rolling itself does not mean that you are losing 2 points. Your profit or loss is already there and marked to market every day - it is determined by whatever price you are originally short ZB, so for example if you were originally short ZB Mar at 135 and ZB Mar is currently 143, you have a loss of 8 points. This loss of 8 points will not go down to a loss of 6 points just because you roll from Mar trading at 143 to Jun trading at 141. Rather as you said above you will just realize your loss of 8 points at the time of roll when you buy to close Mar at 143, then you will sell Jun at 141 so in the end your P/L has stayed the same and you now have the rolled position with the new cost basis. That part seems straightforward.
ReplyDeleteThe part I still don't understand is at about minute 190 in the discussion when they talk about how you have to pay the carrying cost and regardless of how you do it the carrying cost gets paid. At what point in this process did we pay the carrying cost? Does the carrying cost get paid by the buyer of the future or the seller?
So using the example you gave above:
ReplyDelete"Let's say your original short was 140 but you've got to roll today at 142, you lose 2 points on that trade, it's over. Now you short the next contract currently trading at 141, 1 point lower. If you close at 140 which was your original short price, you've lost a net of 1 point for the two trades."
So you lose 2 points on the original trade at the time of the roll. In the next contract cycle you gain 1 point, so you have a net loss of 1 point. But if the contracts did not expire and we did not need to roll the result would be as follows: You would be short at 140. You would be down 2 points when it was trading 142. At some point in the future you would close at 141 (same as the rolled position going from 141 to 140). Your net loss would be 1 point - same result.
I would disagree that losing one point on each scenario is the same. On one you are closing at 140 and one you are closing at 141. Let's say you are a technical trader and you shorted at 141 and your target, regardless of the function of rolling, was to close a short at 140 and you would not be interested in being short at any price under that. What if it took six months to get to 140 but because of rolling your net on all the trades was actually negative. The end result is ZB at its then current price is where you would want to cover any shorts but yet you could lose money on that trade.
ReplyDeleteFor a better definition of the cost of carry I would consult Tom, I wouldn't be able to do the definition justice. But to answer your question, we as the trade don't directly pay that carrying cost, its a function of the other side of the trade which is holding the actual physical commodity like a 30yr bond. The reason futures settle in to physical every 90 days is because of this carrying cost. What he means by you paying it either way is whether you are the physical holder of a commodity or the trader who pays by virtue of the differential in prices on the next cycle futures contract (the rolling cost). That's the best I can explain it, I would defer to the master. From my perspective its simple, I don't actually care about the mechanics behind it, I just need to be aware that there is a differential in price on futures contracts and when it comes time to roll there can be contango or backwardation.
Here is what I don't understand about your technical trader example above:
ReplyDeleteSuppose you bought a stock at 100 with a price target of 110. If the stock had a 2:1 split so your stock was now $50, you wouldn't say I'm holding until $110, your new target would be $55 to account for the shift in the price of your underlying.
So in the ZB example, if you start the trade at 141 in one cycle, but in the next cycle the cost structure is shifted (by the cost of carry), why would you think your target price should be the same as it was in the beginning?
Your example is my point! What I'm trying to say is you need a different price target to get back to even, and the more times you have to roll the worse it gets. And since all commodities are bounded by zero on the downside, at some point you would probably choose not to stay short. This isn't a right vs. wrong discussion. I'm simply pointing out that the cost of carry can affect your total profit and loss even if over the long run the price of initial entry and eventually exit are exactly the same. That's why I was saying you need to look at futures contracts as three month maximum trades and deciding to roll a trade is really a decision to close one and enter another.
ReplyDeleteI completely understand the concept of looking at the futures contract as a 3 month trade, and deciding to roll is closing one trade and entering another. I also agree that after a roll there will be a different price target - but that is irrelevant - we agree that the roll is a brand new trade, it has a different entry price so naturally the price target will be different than the original trade.
ReplyDeleteThe part we seem to disagree about is that you believe the more times you roll the worse it get. The way I understand it the roll is irrelevant to your profit or loss (with the exception of commissions). It is merely finishing one 3 month trade, and starting a new three month trade at a different price point. In the original trade you may have a profit or a loss when you close it at the time of the roll. After you roll, as you said, it is a new trade and it starts with a P/L of zero. If you want to think of it as a continuation of your original trade that is fine - so if you lost 5 points on the original trade you will need to make 5 points on the "rolled" trade to get back to even. So if you went short at 140 and ZB is 145 at the time of the roll, you have lost 5 points. If you choose to roll and Jun ZB is trading at 143 at the same time Mar ZB is trading 145 (because carrying cost is 2 points in this example), then you become short Jun ZB at 143 and now you need a 5 point move to get back to even. The fact that your price target is now 138 instead of the original 140 is irrelevant - you still need a 5 point move and it is a different target because you have a different starting point. So it's not like the roll is costing you any money (excluding commissions) - you don't need more than 5 points to get back to even after the roll.
I had an email discussion about this today, I will send it to you.
Spent a lot of time trying to figure out how do to a simultaneous roll from Mar to Jun, but in the end just ended up doing a two day roll. With bonds down yesterday bought to close Mar ZB for about 141'12 and with bonds up today rolled into short Jun ZB for about 140'26. Bonds could easily go up from here but I think as far as the roll goes this worked out okay.
ReplyDeleteI would say things worked out better than okay, you effectively kept your short on but did a short term scalp. The differential on your trade is 18 ticks and if you just rolled from MAR to JUN and paid the current price you are looking at an 40 tick differential or so.
ReplyDeleteRight - I kind of wanted to wait longer before re-establishing my short position to see if bonds would rally more in the next couple of days but in the end I decided to just be happy with getting a good roll.
ReplyDelete