Nice post and visualization! You might be interested in a different but related thought experiment from Richard Chappell. 

My perspective on this (or more generally on the question of whether the future is likely to involve realizing a large fraction of the possible value it could have, whatever form it turns out "value" takes) is perhaps a bit more hopeful. In my view, the question only makes sense if we are moral realists. If there are no objective facts about morality, then I don't see why we should care whether our own preferences or someone else's win out. Furthermore, I think worrying about these questions is probably pointless unless two other things are true: that we have some way of discovering moral facts and that those discoveries have some way of influencing our actions. Unless those two are somehow true, we have no reason to think our efforts can in expectation increase the amount of value realized in the world. 

So far this is a somewhat pessimistic take, but I'm optimistic in a world where all three of these conditions are true, which in some sense is (IMO) the only world where this conversation makes any sense to have. In that world, we should expect that increasing the amount of things like intelligence, time to devote to research/reflection, and focus on studying moral questions in expectation leads toward getting closer to the true morality. Welfareans (or more broadly whatever target will produce the most true value) may indeed get enough advocates just by virtue of society making more progress on these moral questions. As an example, society today includes lots of advocates for groups like women, LGBT people, people of color etc., when historically the only advocates were a "tiny subset of crazy people." But of course moral progress is at best an extremely messy and incremental - factory-farmed animals are the victims of lots of people being either indifferent or wanting to maximize something other than welfare (profit, tasty food for humans etc.), and the impact of animal advocates has not been sufficient to prevent a massive explosion of suffering. 

Still, on net I lean towards thinking that given the opportunity for study and reflection (and given the three conditions described above), we can be optimistic that we will drive toward the things that matter. Therefore, focusing on the efforts to prevent existential catastrophe or value lock-in may be among the best things we could do to ensure that we're not leaving a huge fraction of the possible value of the future on the table. That may be easier said than done, since preventing value lock-in in practice means preventing people with maximizing ideologies from successfully carrying out that maximization at least for some period of time. But that makes me hopeful that existing EA efforts may not be too far off the mark. 

Manifesting

Do you have views on Tradewater as an offset provider? Their claim is that they can offset at $19/ton, and Giving Green seemed to think that was credible a couple years ago. 

Great post!

Mill was working as a colonial administrator in the British East India Company at this point in his life, right? Could there have been a role for cognitive dissonance in driving his depression? 

You may be interested to read some of MacAskill's older writing on the subject https://www.lesswrong.com/posts/FCiMtrsM8mcmBtfTR/?commentId=9abk4EJXMtj72pcQu

Gotcha, that makes sense! Even if producers slaughter at a lower weight, I think the number of chicken-days of life per kg of meat shouldn't change much relative to what goes into the WFP analysis. So I don't think that producers slaughtering earlier changes the quantity of time spent suffering very significantly, just whether it's distributed among fewer longer-lived chickens or more shorter-lived chickens. 

Agreed, this post seems like it goes way against standard forum norms if this is correct

I think it's worth noting here that (if I'm understanding it right) the alternative breeds recommended by the better chicken commitment are slower-growing but don't have a lower max weight. And the welfare footprint project numbers on pain durations already account for the longer time to reach full weight.