I made a thread about this a while ago, but one thing I find people want to compare apples to oranges in the per clutch, per egg debate. Also you need to watch how you word what you say and most of what we say is incomplete anyway so it is left up to interpretation of what is meant. I mean saying *insert punnet square results* is per clutch is very incomplete and can be interpreted different ways. I used a more simple example with just a 4 egg lesser normal pairing....
Say I have a 4 egg clutch
L = Lesser
N = Normal
0 Lesser - 1 way
NNNN
1 Lesser - 4 ways
LNNN
NLNN
NNLN
NNNL
2 Lesser - 6 ways (the 50/50...)
LLNN
LNLN
LNNL
NLLN
NLNL
NNLL
3 Lesser - 4 ways
LLLN
LLNL
LNLL
NLLL
4 Lesser - 1 way
LLLL
To everyone saying one egg doesn't effect the outcome of the next egg, you have gone out of the clutch aspect (apples) and now into individual eggs (oranges). Just take the clutch as a whole and leave it that way (now we are comparing apples to apples). Now what is the most likely outcome? it is going to be that punnett square for the clutch. That does not in anyway change that anything can happen, as shown above. If you know the clutch size, you can go farther and say chances are X I will get Y. Like in my above example your chances at getting 0 or 4 lessers is 6%, 1 or 3 lessers is 25% and 2 lessers is 38%. but we don't know the clutch size most of the time. Is it wrong to say per clutch in most cases? It would be depending on the wording, which most of the time is incomplete. Per clutch however is a more complicated aspect than per egg. As shown it is not as simple as punnett square results, but the punnett square results are the most likely outcomes, which is what I believe most people are getting at anyways, again all interpretation. So why do we debate this again?
Honestly I think it becomes an English debate rather than anything science or math related.