When to shoot; and when not to!?

Just shoot, the photos are essentially free anyway.  Culling really doesn't take that much time if you something like Pixinsight Blink.

The only real problem is that clouds can cause your guidance system to lose the guide star.  On a few nights I've shot unguided with short exposure time to avoid that problem.

Unless you get perfect skies regularly, I would suggest you snatch whatever time you can. And practice does make perfect, there are so many ways to go wrong, that it is easy to do if you wait weeks between getting out. I try to focus on one or two targets at a time, to get to know them. Only with some bad nights on a target do you realize what is good. I stack with DSS and generally find sky background, FWHM, number of stars detected are relatively consistent from night to night, if you get rid of the bad images, which is very easy to do once you know what is good and what is bad for that target. Remember that seeing is a function of aperture, so for smaller apertures seeing is better than what might be posted. And on cloudy covered nights, the sky between the clouds can still be excellent, you just need to be picky with your images.
"Clear-ish" skies!
Rick

Scott Homstead:
My issue is more "when to post and when not to post"

haha—this IS the real question!

When in a cloudless night a bull elk is running towards your scope, then you shoot. 😎
Since I do not own any filters, I gave up imaging when the sky is sad sparing myself a few disappointments.
Cheers, Robert

For instance, this is NG.

Yep, I pretty much shoot when I can, getting rid of any bad subs. I recently built my observatory which has made imaging in sporadic weather much easier, literally roll the roof on if necessary and resume when possible in the night (no more tearing the gear down and putting it up and PA etc etc. It makes a hell of a difference).

I will look into automation in the summer using cloud cover as a metric for controlling the roof, but until then - I am as I believe we all are an opportunist. I am literally working on 3 projects and they each have several nights data already.

CS
Brad

I also shoot whenever possible! Cloudless nights are very rare in Northern Ireland. Sometimes even if it is cloudy but the moon is still somewhat visible I will just leave the scope tracking on the moon and keep an eye on the laptop to see if the clouds clear at all. If they do I will have a go at imaging something at least. As others have said already, it is good practice and helps to iron out problems that could crop up at a more critical time when the skies are at their best. I find my big long lynxastro usb cable handy for this so I can monitor the skies "remotely" from indoors.

As for discarding unusable subs I generally just make the thumbnails large on windows explorer and delete all the obviously bad ones. Then I find that the deep sky stacker score works well enough in helping me to sift out the ones that i didn't notice. I just click the check all above threshold box on deep sky stacker and then put in a minimum score that seems to be reasonable whilst still making the most of the data.

I am relatively new to this hobby and I am constantly trying to improve my pictures so maybe some day I will become a bit more meticulous with my work flow in this area :-) Sometimes I consider whether I should be discarding more subs. Does anyone know what is best practice when it comes to this? Would you prefer to use more "average" quality subs or less "good" quality subs?

Just to David's comment on using DSS stacker low scores to eliminate subs, I would not recommend depending on it. If everything else is constant (seeing, tracking, wind, focus) scores actually decrease when the sky gets darker (or if you target is higher in the sky). I think the brighter sky brings fainter stars above whatever star threshold you set. If seeing is worse or focus goes off a little, then FWHM increases which does decrease score. But if your stars are really dancing around due to poor seeing, or your focus is really bad (donuts), or you have wind vibrations or poor tracking in one direction then FWHM can actually go down, which increases the score. So if it is getting darker and my target is going up in the sky, I expect scores to trend down. On that trend line I look for scores that are off that trend line, or FWHM that are off the line, up or down, or number of stars suddenly goes down (dew or clouds). I inspect those in DSS with the magnifier to quickly confirm the problem prior to deleting. Of course, if you are seeing scores like 500 consistently, obviously if you suddenly see 0 or 100 you can delete without further thought.

In my workflow, I typically register all my images, click on the data line for any one of the subs in that tab,  then use the copy to clipboard, which copies all the information for all the subs in that tab. Images on different tabs need to be copied separately. I copy directly into an excel template I created, which then graphs for me FWHM, stars, background, score (and also x and y values to see what my tracking is doing), makes it easy to see the outliers that need to be inspected and potentially deleted. A real time saver if you have lots of subs.

In theory one can also see everything plotted as you make your subs using DSS live. However, on my laptop, when I am running acquisitions it tends to crash. So I prefer to use the full version later as I am working up my data.

Hope my experience (throwing out good data), helps others to not throw out good data with the bad.
Clear skies all!
Rick

I have had ONE clear night in 2021, so right now I will take what I can get. However, I have to babysit my gear as it sits by the sidewalk and will get stolen if I leave it - I do have alarms on it but … I am not going to get there in time to do much

Given this, I do try to pick the best day if I have 2-3 predicted clear days in a row. If it's a single good day and not too many meetings, I try to go for it

Without a GOTO mount, multi-night projects are tough, so I try to capture 4-6 hours in a single session if possible

I think everyone should "automate"as much as they possibly can in order to save set up time.  Some have full rigs on rollers, others have tripods permanently placed, some have complete set ups ready to go in a garage.
I have 2 telescopes that are each out fitted permanently with dew heaters, cables and reducer/flatteners, so I don't have to put them together every clear night.
I now have the luxury of living out of town with a fenced yard so I have a pier and rig permanently mounted and covered 24/7, but I used to have just a tripod set up on cement pads that never moved.
I am now polar aligned and ready to shoot in around 5 minutes mostly because I still manually focus with a Bahtinov mask.
I shoot all the time.  I shoot in between rain storms.  I shoot during the full moon.  I Polar Align through clouds just to practice.  My images are not always good, but I don't want to miss an unpredicted clearing.  Chris Sullivan is correct

scott

I image every single time there is a chance, even through thin clouds (that's UK for you). As I have everything set-up in the back garden that is relatively easy (still around half-hour wastage). And I have two rigs so I can pick and chose or run them concurrently.

I also shoot whenever I can. Living near Chicago we can go months without clear nights and the "good" nights often have poor seeing or transparency.
I usually set up my gear and watch it for a while then eventually go to sleep.  I use Sequence Generator Pro for imaging and with broken clouds it can end your session with a terminal error if you don't have your settings right.  I found this out the hard way, but you can tell it to "retry" repeatedly and it will recover the sequence when the sky is clear again. With it set up correctly, I've been able to sleep through partly cloudy nights and wake up to a directory full of images. Culling through them I can see the clouds approaching, overtaking then clearing out again. There's definite down-time when the clouds are around but at least it's been able to get some decent images and recover my imaging session most of the time, nothing’s 100%. 

Like others have also said, even on poor nights, it's good to keep your skills up to date, especially if clear nights are rare. It's easy to forget a lot of details when you haven't set up for months then you waste precious time reacquainting yourself during a genuinely good opportunity. On the marginal nights I’m on the fence though and I learned to tamp down my expectations, especially those nights with a thin layer of clouds where the sky looks clear but images don’t pick up fine details and guiding isn’t very good.