8/5/2013 Monthly Retail Sales (2130 EDT)
Forecast: 0.81B
Actual: 0.60B
Previous Revision: -0.16B to 0.51B
INDECISIVE
Started @ 0.8895
1st Peak @ 0.8903 – 2131 (1 min)
8 ticks

Reversal to 0.8886 – 2131 (1 min)
-17 ticks

Notes: Report fell mildly short of expectations, with a concurrent moderate downward revision to the previous report. We warned about the HPI report at the same time being potentially problematic as it was in May. Since it was positive, this caused offsetting news and indecision. It started on the R1 Mid Pivot and 200 SMA, then spiked long to the R1 Pivot and reversed back down to about 4 ticks above the LOD. With JOBB your order would have filled long at about 0.8898 with no slippage, then you would have been stopped out within 2 sec at 0.8891 with 2 ticks of slippage for a 7 tick loss. After the :31 bar, the market rallied for about 20 ticks to the R2 Mid Pivot in the next 30 min on the HPI news. Fortunately we raised the risk warning and gave warning about the HPI report, so this should have been a disqualifier for most.

5/6/2013 Monthly Retail Sales (2130 EDT)
Forecast: 0.20B
Actual: 0.31B
Previous Revision: +0.07B to -0.11B
INDECISIVE
Started @ 1.0213
1st Peak @ 1.0221 – 2131 (1 min)
8 ticks

Reversal to 1.0203 – 2131 (1 min)
-18 ticks

Notes: First, the reason AUS Trade Balance was indecisive was due to a positive Trade balance result and a concurrent extremely negative Quarterly HPI report that had the biggest offset from the forecast in several years. Normally the HPI report is a yawner and has little effect on the market. However, with a mildly positive Trade Balance with a normally larger share of influence coupled with an epically disappointing low influence report, they balanced each other and caused indecision. Here is a reconstruction of what happened:

1. The spike initially went long and filled your long entry with about 2-3 ticks of slippage at 1.0218 with the # of contracts you desired. 2. The spike reversed and headed short while the stop loss from your ATM strategy attached to the long entry was setting up during the brief overlap while the opposite short entry was still on the chart before the OCO function triggered. This caused both the stop and the short entry to fill with about 0-3 ticks of slippage at 1.0210. So your initial fill was stopped for about -8 ticks with the # of contracts you intended to trade and you had an additional # equal to your initial lot fill as short positions at 1.0210. 3. The 5 tick stop loss for the 1.0210 short entry tried to apply at 1.0215 as the market was rebounding long, but was disallowed by NinjaTrader as it was below the market price at that time. 4. Then the knife switch logic in JOBB saw that both entry orders were filled and attempted to flatten all open orders, resulting in closing out your additional short positions for an additional loss of about 9 ticks. So you should have lost about 17 total ticks on this report.

BOTTOM LINE:
This is the second time this scenario has happened (German IFO on 3/22/13 is the other occurrence) where it spikes one way and fills the entry, then reverses during the narrow window when both the stop loss and the opposite entry are on the chart together. The overlap is only about 3/10 of a second according to the programmer, but necessary to have no gap in protection. Fortunately this has only happened twice in about 15 months of JOBB trading over 500 reports. The AUS Trade Balance remains a safe report, but the concurrence of abnormal news caused an undesirable situation this time.

1/7/2013 Monthly Retail Sales (1430 HI time / 1930 EST)
Forecast: -2.21B
Actual: -2.64B
Previous Revision: -0.35B to -2.44B
DULL REACTION then DOWnward fan
Started @ 1.0460
1st Peak @ 1.0452 – 1431 (1 min)
8 ticks

Final Peal @ 1.0423 – 1751 (201 min)
37 ticks

Notes: Report moderately disappointed along with a sizable downward revision to the previous report. This caused a 8 tick initial spike on the :31 bar that crossed the 100 SMA then hit the R1 Pivot and 200 SMA and could go no lower. After the initial short spike, it stepped lower in a slow, but deliberate manner to manage 37 total ticks in nearly 3.5 hrs. It is not clear if the entire short move was due to the sentiment of the report as it usually has mild influence. If so, the reaction pattern may be shifting with this report. With JOBB your order would have filled short at about 1.0455, with 2 ticks of slippage, then you would have only been able to get 1-2 ticks unless you were patient and stayed in longer than 8 min. As this was a dull reaction, I moved my stop loss down to 1 tick above the 100 (1 tick loss) and closed out with 1 tick of profit. There are 2 reversals during the downward move that reclaimed 10-11 ticks, but nothing significant after the bottom was reached.

10/2/2012 Monthly Retail Sales (1530 HI time / 2130 EDT)
Forecast: -0.69B
Actual: -2.03B
Previous Revision: -0.97B to -1.53B
Spike WITH 2nd peak
Started @ 1.0178
1st Peak @ 1.0163 – 1531 (1 min)
15 ticks

2nd Peak @ 1.0157 – 1536 (6 min)
21 ticks

Reversal to 1.0172 – 1615 (45 min)
15 ticks

Notes: Report fell well short of the forecast, with a large downward previous report revision. This caused a relatively big and fast short spike for 15 ticks that would have filled your short JOBB entry at about 1.0172 with 1 tick of slippage, allowing for about 8 ticks to be captured as the bar settled at the maximum low just below the S1 Pivot. It then claimed 6 more ticks on the 2nd peak 5 min later before reversing for 15 ticks in the next 40 min as volume dried up. The S1 pivot acting as support and the mild bearish sentiment before the market likely prevented a bigger drop, but it was still a great breakout for JOBB.