I learnt "old school" butterfly widths based on training videos and subscriptions from about 2 years back. Even then I wondered why butterfly widths should be so - if a 30 point width (rhetorical number) was defined when Russell was 700, how can it still work when Russell is 1400 (now 1700)?
While a good strategy can still make it work via adjustments, I think we are ending up with lot more adjustments than before. For example, if a 1% move on a 700 point index was 7 points, the same 1% move on 1700 is 17 points so having a 40/50 fly implies such a move is covering 1/3 of the range on 1 side.
It will be better if we adjusted the trades a bit less because the butterfly span wide enough for most movements. I do not want to increase wing 2.5 times blindly because a) you can still use adjustments to make the trade work, b) the flies start costing quite a bit more and have bigger drawdowns and we should model them too, c) increasing wings 2.5 times is going to change the shape of T+0 line and potentially make it (more) unmanageable.
I wonder if someone else has done some work on this?
To share my side of story, I am experimenting with the following BWBs:
RUT - 60 downside, 50 upside
SPX - 75-60
NDX - 175-150 (I would like to maintain a 4:5 or 5:6 ratio but I do not like to use the 10 strikes, only the 25s so that's my current and 1st experiment on this).
DIA - 7-6
I am still not satisfied with the changes I have done and hopefully will have a better understanding of this in a couple of months.
For those curious, I (try to) trade NDX at half the size of SPX and RUT (because of massive moves it can have), and I trade DIA at half the size of NDX (because of commissions). But I would also like to know if people have been able to trade NDX flies successfully. BTW most of my flies are 45 days out.
While a good strategy can still make it work via adjustments, I think we are ending up with lot more adjustments than before. For example, if a 1% move on a 700 point index was 7 points, the same 1% move on 1700 is 17 points so having a 40/50 fly implies such a move is covering 1/3 of the range on 1 side.
It will be better if we adjusted the trades a bit less because the butterfly span wide enough for most movements. I do not want to increase wing 2.5 times blindly because a) you can still use adjustments to make the trade work, b) the flies start costing quite a bit more and have bigger drawdowns and we should model them too, c) increasing wings 2.5 times is going to change the shape of T+0 line and potentially make it (more) unmanageable.
I wonder if someone else has done some work on this?
To share my side of story, I am experimenting with the following BWBs:
RUT - 60 downside, 50 upside
SPX - 75-60
NDX - 175-150 (I would like to maintain a 4:5 or 5:6 ratio but I do not like to use the 10 strikes, only the 25s so that's my current and 1st experiment on this).
DIA - 7-6
I am still not satisfied with the changes I have done and hopefully will have a better understanding of this in a couple of months.
For those curious, I (try to) trade NDX at half the size of SPX and RUT (because of massive moves it can have), and I trade DIA at half the size of NDX (because of commissions). But I would also like to know if people have been able to trade NDX flies successfully. BTW most of my flies are 45 days out.