Obviously the bullpen hasn’t been great. But applying a rate developed over 486 games and expecting it to be repeated in a stretch of 30 games, especially early in the season with short starts and the use of “openers”, seems to warrant a “small sample size” alert.
I also looked at those 8 games and 2 of the 8 “bullpen” losses were charged to Sproat and Patrick when they replaced openers. In the other 6, it looks like only 3 of the games included blown saves. Those are situations when the relievers give up winning runs in tied games.
I also note that the number of runs scored by the Brewers in the 8 games in which the bullpen was charged with losses are: 2, 2, 6, 7, 5, 3, 3, 2. If the Brewers had a perfect bullpen they would have a couple more wins, but when you score 2 or 3 runs you’re going to lose more than you win.