According to the spreadsheet, it was closed 10 days before it was opened, but I think that was a matter of entering the wrong month for the exit. Nevertheless, I still have some confusion about whether the official numbers are based on the actual system exit or Matt's discretionary exit.

The trades are triggered off ES, but yeah we trade MES.

James, I am a bit unclear on this difference. My own definition of trending means that DMI indicator is above 25. The distinction between impulsive and corrective in EW speak is related but not exactly pertinent, as we have seen last month and this month (the price could be trending and still be corrective). The price above resistance or below support like on Friday or today is suggestive but not indicative either if the ATR is small. Any comments?

Next bar open would be with entries on the open of the following trading day. We're currently finding a mix as to whether or not performance is better or worse using this approach, for subsequent entries we're actually generating limit orders, but the initial entry is just always at the open of the subsequent bar.

Our current approach is the best guess trade 20 - 15 minutes before the close, with confirmation at the close, the message I put out at around 3:40 EST saying the STrap would go to the wire was because the trade was turning off and on every few seconds, price needed to be below about 399.39, we obviously got there at the close, but leading into the close it wasn't clear we would.

We've played around some with artificial trading days, ie: excluding the last 30 minutes or rolling that data into the next day but performance has been worse than doing next bar open, and it doesn't address a core complaint of some of our users, ie: having to be around at the market close to execute trades. I suspect most swing traders will prefer the next bar open approach, while a minority of day traders will be annoyed by it.