POS vs DPOR stake weight balance?

This was brought back up in the following Gridcoin GitHub issue: https://github.com/gridcoin/Gridcoin-Research/issues/106

When considering how to improve total participating stake weight, we should evaluate the POS:DPOR stake weight balance (huppdiwupp's posts are worth reading) : https://cryptocointalk.com/topic/49141-dpor-stake-weight-calculations/

Link to hupdiwup's initial (not final proposal) post & following quote:

"If you have a low magnitude the probability to stake would be greater if you just ran an investor wallet. If you have a high magnitude the probability to stake is greater than running an investor node.
This was a little bit of a surprise to me. I thought that you would always have at least the probability to stake of an investor and the work you do for boinc crunching increases this probability. So that despite the imapct of "owed" the ratio between investor and cruncher would be constant, given a magnitude."

Proposal 1 conclusion:

"Now, if you are owed nothing the probability to stake is the same as if you were an investor. If owed is positive your probability goes up, if owed is negative it goes down. The impact of the variable owed is dependent on your magnitude, so it boosts small magnitude nodes more than large magnitude nodes.
The problem with this is that you could create a lot of gridcoin+boinc nodes with a small magnitude, close the wallets and then build up probability.

The way it currently works you can also build up probability by crunching with the wallet closed. But since small magnitudes and small balances are being penalized it is more expensive. This is at the cost of newbies not staking blocks or having a newbie boost.

This is basically the same problem as with PoSv1 vs PoSv2, whether to use coinage or coin. So maybe we should eliminate the owed-variable from the equation."

Proposal 2 conclusion::

"The advantage of this would be that you cannot build up probability to stake. Regardless how long your wallet has been offline or what you are "owed" the probability stays the same.

But this would come down to the same discussion we had with the vote weight. What should the ratio between magnitude and balance be.
If ideally all coins would stake all the time it would be no problem if the weight was equal (p=1). But if only 20% of the total coin supply stake at one time it could be too much.

I don't know if the total coin supply is easily accessible. If it is not it could be set as a constant but then the importance of Magnitude would slowly decay as more coins are created.

If there is a way to determin how many coins stake at a single time, we could use that metric instead of total coin supply. This would mean that you would have to have 51% coins staking plus 51% magnitude staking for an attack at any point in time.