It depend a little on your current timing and where you stand relative to your end time, but in general taking the two clears at 1.8x worth of points is superior to the 1.0x over that 5 hour span. Let’s map it out, assuming a 1am end time:
Option 1: Clear 3p (x2), 8p, + 11p (x5). In terms of base points, that’s 1.8, 0.93, 2.65 = 5.38 (this is approximate because you can refresh during the 2hour time and sneak a few extra points in to the nodes)
Option 2: Clear 3p (x1), 8p, 11p (x5). This option is 1.0 + 1.0 + 2.65 = 4.65. Those two hours idle cost around 3/4 of a refresh.
Unmentioned Option 3: Clear 3p (x2), 9p, 12p (x5): 1.8 + 1.0 + 3.0 = 5.8
When you return in 5 hours, it’s more optimal to wait for the 6th hour and full refresh, unless you’re near the end of the slice. Otherwise if you resume every 3 hours from the 5 hour mark, you’ll only be clearing the nodes at ~93% of value repeatedly, so you’d be losing 0.07 per clear for every refresh where you work from now sub-optimal clear time.
Obviously this also assumes you can clear the entire sub in an hour, which in an event like Prodigal’s last sub, isn’t a practical thing given the volume of the nodes. So for option 3 there, if you start at 11pm, you’re back to 1.8 + 1.0 + 2.65 = 5.45. Still more optimal than option 1, but the difference is pretty negligible.
Assuming you’re clearly optimally for top placement awards, rubberbanding will be negligible in both cases, missing one refresh is doubtful to trigger the RB effects. Even if it does kick in, they won’t be enough to make up 3/4 of a refresh.
Shorter version: Other than the end game, any time you clear the nodes an additional time, you should consider that your clear of the following refresh, and stay on the projected optimal clear time from when you originally started.
