I've been building this one for a while and it's finally running live on sim, so I thought I'd drop it here and see what you guys make of it. The idea is pretty simple. Price pushes out of the session value area, fails, and closes back inside. I short that close and target the point of control, with the stop beyond the extreme of the rejection. It'll only take the trade if the level it's fading lines up with something else — prior day or overnight VAH/VAL, weekly levels, a high-volume node edge. Diagram below if that's easier to look at than read. It backtested well enough that I wanted to see it on live data rather than keep tuning it, so that's where it is now. Forward test, sim only, no conclusions yet. Which is really why I'm posting. I'd love a couple of people to run it alongside me. More charts and more sessions means I get to an honest answer faster than I would on my own, and I'd rather find out it doesn't work in six weeks than six months. Things I'd genuinely like opinions on: 1) Does the confluence requirement earn its place, or did it just survive because it tested well? That one nags at me. 2) What's a fair cost assumption when the POC target is sometimes only a few points away? 3) How many trades would you want to see before you'd believe anything either way? Also worth a mention since it's a nice example of the plumbing mattering as much as the logic: I found a bug in it this week where the dashboard table was being re-merged on every tick, which Pine treats as a runtime error. The script was dying part-way through execution and the symptom showed up as wrong alert values — nowhere near the dashboard code that caused it. The fix was deleting one keyword. Finding it was not a one-keyword job. Pine's attached. It's set up for MES on the 1 minute. Run it on sim, and if you do, please shout if you see anything I've missed.