I’ve looked at the numbers enough on this for it to be clear that it would not benefit the median player.
If all you want to do is smooth out the curve, not have this be a huge increase in the precursor drop rate, it would mean lowering the drop rate from the forge for the first several crafts to compensate for higher drop rates after you have been forging a while. In the long run, you would see about the same number of precursors, but they would show up at more regular intervals.
If you used a linear increase in the drop rate with each failure, starting at effectively zero, that would mean that the drop rate would be lower than it is now until you’ve had roughly 100 failures.
Also, cumulatively, it would mean waiting longer for a precursor, on average; at current numbers, you have a 50% chance of seeing a precursor before roughly your 111th exotic, but under the linearly increasing system the 50% chance doesn’t hit until roughly your 150th exotic.
The benefits are on the back end – at the moment, you have only an 85% chance to see a precursor by your 300th exotic, but under the linear system it would be closer to 94%.
The sort of increasing system you propose would be a big boon to dedicated precursor forgers; it would smooth out the luck and break up the very long losing streaks that pose the biggest risks. But for your average player, who doesn’t throw thousands of rares into the forge, it would make the forge even less rewarding; the odds of getting a lucky forge as a casual player are awful.
This doesn’t even get into the way it can pervert incentives – in this kind of system, you only get rewarded if you are really dedicated, instead of randomly, so just throwing 4 rares or exotics into the forge on a whim is awful in the proposed system; if you’re not doing to do it en masse, don’t even bother.
Of course, you could compensate for this somewhat by making the drop rate higher, overall – you’re excluding casual players getting lucky, after all, so the drop rate can probably go up a bit to keep output the same. But this is a net transfer, from random lucky casuals to dedicated forgers. Do we really want this to be a mechanic that drives the economy towards the rich getting richer?
I definitely appreciate how frustrating an independent, low percentage system can be. At the same time, I don’t think the casual player really grasps just how many resources need to be consumed by the forge to spit out a precursor. You need to feed the thing rares by the thousands, and evening out the odds isn’t going to change that.