Cain got paid through 36, but at a rate that reflects his expected decline, 5 yr $80M assumes about 9-11 WAR over the life of the contract, reasonable for a guy who is probably 3 WAR his first season, 6-7 WAR his next three and maybe waived his fifth year. Had it been structured as a 4yr, $80M, it would have made the same sense, but a higher "cap hit" in the first four years.
I don't think it's collusion, rather, the rise of analytics meant teams have analyzed past FA contracts, and realized the inflated valuation of WAR by some teams was a mistake. I think teams are simply holding the line to the expected value of FAs and not engaging in wishful thinking.
So if teams figure Darvish is a 12-14 WAR player (7 WAR next two years, 5 WAR the following two years at 33-34 and maybe 0-2 WAR after that point), his market value is probably $90 - $125M (7.5 to 9M per WAR, i.e., those $10M per WAR projections are based on past stupidity) - so if he's looking for 6-7 years at $25M, he''ll be waiting a long time.
I think if his agent called Klentak and said he'll take 4yrs, $100M, he'd be signed tomorrow.