Double Butterfly spread

[i960]

I'd agree that's one of the key advantages, that is: variance and trendiness (if it's a decent spread). Margin offset/reduction is also fair/decent but of course commissions are a con (unless a prop trader).

 

[cdcaveman]

I'd agree that's one of the key advantages, that is: variance and trendiness (if it's a decent spread). Margin offset/reduction is also fair/decent but of course commissions are a con (unless a prop trader).

Trendiness or near stationarity. . If you get with the right broker your commissions should be 2.25 one way per future... Which makes a lot viable... In and out of a fly your less then 20 bucks or .02 cents in crude... Which if you can't make sense of that prop commissions won't help you

 

[i960]

Trendiness or near stationarity. . If you get with the right broker your commissions should be 2.25 one way per future... Which makes a lot viable... In and out of a fly your less then 20 bucks or .02 cents in crude... Which if you can't make sense of that prop commissions won't help you

Yep, I agree on that, but given that this is a DF thread I was considering the context of more complex 6-8+ leg stuff and the fact that stationarity is of course higher in those constructions.

For instance, I'd never consider doing a 1:4:6:4:1 or 1:5:10:10:5:1 without prop rates (and maybe not even then either).

 

[cdcaveman]

likely in a significant dislocation you would never need to extend out that far to isolate a disjoined month in the back in order to get a good return.. A single fly will do the trick... and a double fly will definilty be more then enough..

 

[i960]

Would love to hear some thoughts on when someone would use this... I imagine you could +1/-2/0/0/-2/+1 as well.. but when does this make sense

Found a reference to this EDF spread on Curve Advisor's website:

http://www.curveadvisor.com/butterfly-trading-basics/

 

[cdcaveman]

but the question lies.. where does the out of sink future go in the fly.. typically in a single fly if say june is out of sink and trading at a premium.. you sell that as the meat and buy the wings

 

[i960]

but the question lies.. where does the out of sink future go in the fly.. typically in a single fly if say june is out of sink and trading at a premium.. you sell that as the meat and buy the wings

As you know, a fly is a spread between two spreads, (A-B) - (B-C), but the end result is that one is typically focusing on body vs wings or vice-versa.

A DF is really just (A-2B+C) - (B-2C+D) and could be considered a spread of flies, in the same way that a fly is a spread of spreads. If you look at it that way it's conceptually more of a spread than a fly. However, another way of looking at it is that a DF is also AB-2*BC+CD meaning that the body is the BC spread and the wings are the AB and CD spreads. You're not focusing on an out of sync future you're focusing on an out of sync spread. I'd imagine that for the more sophisticated STIRs guys that are trading these regularly they're doing them for all sorts of reasons and not necessarily isolating something specifically.

Remember the stuff @Wingz was talking about? If you have a 6m spread there are 6x1m spreads, 3x2m spreads, 2x3m spreads, etc. that when adjacently summed together they result in the original spread. These are all the combinations out of a 6m spread:

o: M N Q U V X Z
1: MN NQ QU UV VX XZ
2: MQ NU QV UX VZ
3: MU NV QX UZ
4: MV NX QZ
5: MX NZ
6: MZ

e.g. using a 6m spread and 2m adjacent components summed:

CLM7:CLQ7 + CLQ7:CLV7 + CLV7:CLZ7 (Q and V cancel out)
=> CLM7:CLZ7

But rather than spreading all the component spreads against the 6m spread (which would be kinda pointless) it's instead possible to hone in on a more specific spread that is perceived as out of line. In the above there is an obvious center, CLQ7:CLV7, and if you combine 3 of those 2m spreads against the same 6m CLM7:CLZ7 spread it's part of you get:

CLM7:CLZ7 - 3*CLQ7:CLV7
=> CLM7 - 3*CLQ7 + 3*CLV7 - CLZ7 (DF)
=> CLM7:CLQ7 - 2*CLQ7:CLV7 + CLV7:CLZ7 (DF)

Which is a double fly with QV as the body and MQ and VZ as the wings.

 

[propwarrior]

Trendiness or near stationarity. . If you get with the right broker your commissions should be 2.25 one way per future... Which makes a lot viable... In and out of a fly your less then 20 bucks or .02 cents in crude... Which if you can't make sense of that prop commissions won't help you

The starting all in per side cost is $1.15 at prop firms. The higher volume players can get <50c. This opens up a lot more strategies. If you can get a fly on for $2 or double fly for $4 it's a huge deal. You also have to consider that big players will often average into positions because their bankroll allows it.

 

[propwarrior]

likely in a significant dislocation you would never need to extend out that far to isolate a disjoined month in the back in order to get a good return.. A single fly will do the trick... and a double fly will definilty be more then enough..

From my experience trading decisions are not just identifying dislocations in a curve. Many are pure stat arb plays. Spread positions can be morphed into longer term plays. So for example you may be building a favourable price in a fly knowing in advance what you will do with the position if you cannot exit favourably. So you start with fly as a base. Sometimes it is necessary to morph a position into a longer term play further back in the curve or possible into an Brent/Crude arb.

GL

 

[propwarrior]

My observation was based on eye balling the charts shown earlier in this post. What objective measure would you use when comparing the volatility of two instruments - spreads vs outrights? Spreads may very well be exponentially less volatile than an outright but my curiosity is piqued and I'd like to put some data in a spreadsheet and see if this is indeed the case ... as it may well be.

Its not just about volatility. If you look at the front future for any commodity you will see many cases where price is pressured towards areas where hot money places it's stops. If you check out the calendars down the curve you will see less of this behaviour. This is why a lot of the big players will use calendars or stay away from the front to express a directional trade idea.