Favorite Futures: equity and fx futures and options
Posts: 26 since Sep 2013
Thanks: 6 given,
I never adjust. My reasoning is that an adjustment affects the analysis in different ways depending on the time frame and tools you are using. I don't think that one method is more accurate than the other and it is a waste of time and work trying to sort it out (it was for me anyway).
Check out the ES for the last TOS rollover. There was a small gap attributable to the actual rollover trade (the spread from one to the other) and a larger gap attributable to the overnight trade on the day they stitched the 2 contracts together. Look at the huge difference on adjustment between a session and globex chart. Neither is useful imo. And keep in mind that they stitch them together a week before the actual expiry, which seems irrelevant to any meaningful rollover trade (big money holding their position).
For long term charts that span several contract cycles, I think that the carrying costs average out. Those costs are highest at the start of the contract period and decay until expiry, so there is a skew that I don't care to account for. I use the cash charts (SPX) for the long time frames and I keep in mind that S/R is more of an area. I use the ES for volume profile aggregation. Keep in mind that the TOS profile aggregates by bar rather than price-by-tick, so nodes are areas rather than precise price points.
For intraday charts, I use both a session only and a globex chart. The globex chart is only used to view ONL/ONH and the reactions that occur when the session traders test these levels. I use the session only to plot inter-day trend lines and S/R.
An alternative is to use the discrete contract chart but I don't use it due to the extra work of making new drawing sets, alerts, etc. The discrete contract is a better alternative to adjustment. I use the continuous chart and plot the rollover line (the seam - where TOS stitches the two together). For the first few weeks of the new contract, I ignore the structure of the expired contract, especially that which occurred late in its period. As the new contract matures, I begin to consider the old structure and consider it to be increasingly relevant. If you believe that the carrying costs are manifest as a repeating skew, then there is a period when that amount is equal from one contract to the next. For me, a simple non-mathematical view is to consider that it occurs half-way through the period. Therefore, if I have a pattern that occurred in the expired contract (on my 90d session only chart), I will look for a small intraday pattern to set up in that area. If it does, I then ignore the older structure - because my area is proving to generate interest and then just concentrate on the new small pattern for execution. The expected range of my trade is usually a measured move off of the new pattern, that will also coincide with recent structure and also to historical areas.
Also: the software is notoriously unreliable for plotting rollover lines. The ES is not so bad because its popular, but the FX futures have been known to not plot and plot on the wrong day, etc. I have pics that show the plot on the wrong day and making the adjustment on that day. My workaround is to make my own time level drawing. And the FX products aren't necessarily stitched together on the same day. To determine what is active in the continuous mode, go to the trade page and you will see which one is the active contract.
to summarize, I think derivatives price gets fuzzy over time and the best approach is to develop a set of mechanics that is simple to use.
The following user says Thank You to Leon of Pizza for this post: