You might want to see this academic paper specifically arguing against the net-negative view (empirically calling these assuptions into question). I critiqued it in my PhD thesis (which I really should put online somewhere...), and argued for a more agnostic view. You can read it below:
Heather Browning and Walter Veit (2023) provide several reasons to be skeptical of [the net negative view]: first, most deaths will be short, relative to the length of the animal’s life; second, for the most intense deaths (most obviously, predation) a “shock response” may numb the pain that is felt; third, stressors experienced during the animals’ lives (such as an awareness of proximate predators, or a disease response) are “not sufficient to assume negative experience without additional evidence linking this to welfare”; fourth, wild animals’ lives will also be filled with many positive experiences, often paired with any bad affects (e.g. the pleasure of eating and the pain of hunger); and finally, the baseline, purely ordinary moments of life might be experienced positively, since it is advantageous if animals value their lives and seek to keep living, and especially exploring and discovering.
Browning and Veit’s argument presents a compelling case for agnosticism regarding the balance of positive and negative experiences in the lives of wild animals: without really knowing how intense the bad and good experiences of wild animals are, we cannot say whether their lives are bad overall. But their optimism that wild animals will generally have positive lives is less convincing. Their arguments also provide reasons for discounting the badness of experiences that we normally consider terrible for humans, such as severe injury or violent death, the constant threat of danger or starvation. The fact that these experiences might (respectively) be partially numbed, brief, or are paired with experiences that are good, do little to dampen their horror, especially for those who die because of them at a young age. While we do not know how other animals will experience these states, it is at least reasonable to suppose that they involve a similar level of suffering. Secondly, while it is certainly possible that purely ordinary experiences of animals are positive, there is no evolutionary necessity for them to be so, since animals are already driven to keep living by negative experiences, such as hunger and fear. Lastly, the positive experiences which Browning and Veit point to, such as the joy of discovery, are equally paired with negative ones, such as boredom. Consequently, we still have strong reasons to think that wild animals suffer greatly, and a strong case for agnosticism about whether their lives are worth living. We can also say that, even if their lives are generally worth living, they are no more so than are human lives in times of famine or war; in fact, we should expect the lives of wild animals to be worse, since their mortality rates are higher. This is not altogether denied by Browning and Veit, who acknowledge that “whether or not it dominates, there is a large amount of suffering in nature” (Ibid.).
You might want to see this academic paper specifically arguing against the net-negative view (empirically calling these assuptions into question). I critiqued it in my PhD thesis (which I really should put online somewhere...), and argued for a more agnostic view. You can read it below: