Obviously more than 60% but if you only spend around the minimum on players 18-30 (averaged at say $100k so say $1.3m) then you have around $7.8m to spend on your top 17. This is over $450k average.
This comes with a massive assumption which I think many people miss.
Paying players 18-30 an average of $100k each seems clear cut on the surface, but the assumption is that there is a sufficient supply of players worth about $100k. If there isn't, then if the max you want to pay is $100k you could end up paying this amount to a bunch of guys worth considerably less than $100k. There is a massive gulf between a player worth $100k and a player worth $50k, even if the $50k player is paid as much as the player worth $100k. To illustrate this point, the development list players can only be paid $60k max. That means they will probably be worth less than this. If they were worth more than $75k they would surely find their way into some club's top 30. Obviously there will be exceptions.
So the point is that you don't want 13 players in your top 30 worth $100k or less. This is getting into the Dane Aukafolau or Ray Stone territory. Sure these kids might have a big pre-season and play well enough in reserve grade (e.g. Reed Mahoney) but when almost half of your top 30 are filled with these guys you can guarantee the majority will disappoint, and worse, you will need to rely on them to play some first grade, some of them a dozen games or more.
And this is why clubs end up with 'too much depth' while the armchair experts piss and moan about how coaches and administrators don't know this obvious stuff. They do know, but they also know the less obvious stuff that the Glenneels of the world can't grasp.