To work out a specific example because I think you're going to need it, with 20% moonshots.

.2 - first try, spent 20 total.

.8*.2 - second try, spent 40 total.

.8*.8*.2 - third try, spent 60 total.

.8*.8*.8*.2 - fourth try, spent 80 total.

.8*.8*.8*.8*.2 - fifth try, spent 100 total.

.8*.8*.8*.8*.8*.2 - sixth try, spent 120 total.

.8*.8*.8*.8*.8*.8*.2 - seventh try, spent 140 total.

.8*.8*.8*.8*.8*.8*.8*.2 - eighth try, spent 160 total.

...

Expected value spent:

.2*20 + .8*.2*40 + .8*.8*.2*60 + .8*.8*.8*.2 + 80 + ...

After 1 try - 20% chance moon - 20 spent

After 2 tries - 20% chance moon - 20 spent. 16% chance moon - 40 spent. 64% no moon - 40 spent

EV - 36% chance moon - 36 spent

After 3 tries - 20% moon - 20 spent. 16% moon - 40 spent. 12.8% moon - 60 spent. 51.2% no moon - 60 spent.

EV - 48.8% chance moon - 48.8 spent

No matter how many 20% moonshots you take, your overall expected amount spent to get a moon is equal to your current odds of having hit a moon. The same will pan out no matter what your chunk is sized. In terms of getting a moon it doesn't matter what size shots you take.

What does matter, however, is losses you may take on your end to kill off the incoming moonshots. These are guaranteed to be lower if you break up the incoming attacks.