I’ve been using adapt forecast usually towards the end of the week (the red dot shows up early) to account for days where I rode xtra (usually more often than not), or those days when I stopped short on a workout (need a rocker plate for indoor rides…they are killing me :-)), but recently I went to adapt and it’s actually starting to flatten the curve, and actually bring my fitness down from 3-stars to 2. So I’ve left it and kept from applying the updates.
Is there a reason this is occurring?
Post your Program dialog box configuration and slider entries on the Settings tab.
Looks like you’ve recently reached your max hours/week or Availability is limiting TL moving forward.
If you ride more than the forecast, a flattening will happen long before your target event date which appears to be 4 months from now.
Target TL must be greater than Current TL to continue trending up.
One way to fix this condition is to change to Goal (if currently set to Event) and raise Target Watts while watching the Target TL change to a higher number.
For example, if Current TL is ~68 keep increasing the Focus Power watts until Target TL is 80 or whatever load you can reasonably commit to between now and your target date. Then recast the plan which will change the Start date to today. You can test this scenario and view the updated plan chart without saving the changes.
This assumes your max hours/week and Availability support the increase in TL.
Without more details/info, I’m guessing that you’ve nearly achieved the TL target(s) for your goal/event. You might consider re-forecasting your event or goal with more XSS or higher goal power, respectively.
What I ended up doing is removing the forecast and restating the goal - along the lines of what ridgerider2 shared. It did take a few re-runs, but then things made sense again just as I was about to screen shot and re-share what everything was.
For a matter of practice - should I be running adapt forecast when the red dot appears, or do it more along a weekly basis?
Depends, I suppose. On how close you’re able to stay to the XSS recommendations. The less you deviate from the ‘planned’ XSS, the less the need for adapting. As an example, if you’re recommended a rest/recovery day and you go smash a KOM in the middle of a 100km ride, then you’re likely going to need to adapt the rest of the week 
Not sure if it’s the ‘best’ way, but what I’ve been doing is to plan out & pin a couple key sessions (weekly group ride & weekend long ride) into my week, ‘Adapt’ my forecast around those key sessions and then plan the remaining days by their XSS recommendations. I usually do this each week on Sun, my weekly rest day. YMMV
