Turmoil this last week and bad results for the experiment so am closing this one down and starting a new one.

Total over the 3 weeks was 9 losers and 6 winners. I did like the implied volatility bands for targets and stops but, as ever, the direction prediction did not work out often enough. The COT data just was not reactionary enough on a weekly time scale with large speculators staying in positions for months on end. I had seen the same problem with retail trader sentiment, they also often stay long or short for months, for example after the Trump election win the S+P went on a 6 month up trend and retail stayed short for the entire time but weekly bars zig-zagged.

So for the next experiment I'm going to keep the IV bands but try something I've wanted to try for a long time for direction. Just use the direction of the previous week.

Will post next week's predictions later.