Sorry but no. Feel free to mod it away (it's in defines) and observe the result though.
I know the result quite well. The problem is in the defines I only see a way to globally lower the "stubbornness". Is there a way to mod it so only the winner (as defined by positive war score) gets the length of war modifier?
The LoW modifier is a kludge. The most ideal would be to use other factors to increase or lower AI peace resistance, but as a stopgap applying it only to the winning side (and maybe bumping up something else for the losing side) would be interesting to play with.
You appear to be tacitly implying that things would be worse if the LoW modifier was less severe. I entirely acknowledge that you have better things to do than engage in lengthy discussion of the topic, so a one sentence summary of how the results would be worse would be entirely sufficient, since I'm sure you've got some observational reports.
I doubt he wants to because it reveals some bad implications. If you lower this modifier, the AI will declare wars, fight briefly, then just peace out for nothing at all, even if it is much stronger than its target. Basically, the AI has virtually no *effective* means of determining whether it's winning or losing other than war score, so that "length of war" is simply functioning as a war timer, a fake stopper put in to make sure the AI stays in a war long enough to get something out of the peace deal.
Of course, that isn't the reason he gave us in the other thread, instead asserting that wars without the modifier wouldn't carry enough risk (as if risk and the extent you fight somehow shouldn't scale on the kind of demands you want :/). Even if the "this lets the AI do anything ever because we don't have a better model for it right now" reason above isn't the only one, the idea that LoW is materially impacting player risk is, in the vast majority of cases, not credible.
There might be other reasons as well.