Haven’t seen any official announcement yet and no idea how it is supposed to work (and if it’s even a forecast at all instead of just a glitch in the Tibber api), but for the day after tomorrow the prices in VRM look pretty funny 
Anyone else seeing this?
Yes. I saw it too and wondered what was going on. Nevertheless, I observed over the last few days, especially the days leading up to Christmas, how positively the price forecast had impacted the function. The one or two periods with extremely low prices were particularly well utilized. I assume the developers are still working on this aspect. The weekend forecast has been running for quite some time now and, in my opinion, is working well. Therefore, I actually assumed they already had a basis for the missing weekdays. Apparently not. The weekday forecast data was far too consistent, while the current data fluctuates wildly. It remains to be seen.
A similar phenomenon occurred again today, in anticipation of tomorrow’s holiday.
It seems the forecasting functions are really a bit erratic at the moment. And not just the price forecast. Otherwise, I’d have to assume my solar panels would be running at night in winter. So either they’re working hard on it and it just needs some fine-tuning, or some excess alcohol got into the system over the New Year period 
It’s truly remarkable what’s been happening here for the past few days. Please don’t misunderstand, I think it’s fantastic that construction is taking place here, and I have high hopes for future developments and more detailed forecasts. I currently have the time to observe these developments and intervene if necessary, so everything is fine. But I would really like to understand what’s going on. Is it a new type of algorithm that needs to learn, similar to the 28-day learning phase when commissioning a new DESS, or is it a malfunction? It would be great if Victron could comment on this, even without going into specifics.
