I'm going to say it because few will: Today was the low point for Bard, and if he can stay intact mentally, he is primed to improve from here and get back on track. If this was too much psychologically for him to handle, he might not and need a change as a crutch.
I thought his comments after the start were telling (although I think of him as more racehorse and less jockey, and care more what he does and is told to do than what he thinks at this point):
“It’s that maybe we tried to turn me into a starter rather than just taking the same pitcher I was out of the ‘pen and moving that guy into the rotation. That’s probably what should have been done. It’s partially my fault. It’s all my fault. It’s a matter of what I had success doing in the past.”
I have said since spring that all the talk of pacing himself, throwing more changeups, etc was premature and too large an adjustment to make at once, and he needed to establish his fastball, let it fly like when he was a reliever, use his slider to put away, and gradually learn to stretch out and pace, and his role model had to be Verlander and not Lackey

.
Even in his opening start in Toronto as opposed to this one, when he pitched really well overall with bad BABIP on grounders and bad relief/inherited results, he still threw too many breaking balls and too many breaking ball strikes, and didn't establish the fastball effectively.
But what I have seen, is that for all his struggles in the zone, inability to strike people out, diminishing velocity, he was able to do what Clay was not able to do: allow few enough runs to give his team an eyes closed reasonable chance to win.
@Tor: 3 runs in 5 innings, left in too long and Thomas got to pitch to a righty, etc.
TB: 1 run in 6.2, but should have been 0 runs in 6.2 and pulled two walks earlier, we can all agree
@CWS: 3 runs (2 earned) in 7.0
OAK: 1 run in 5 innings, then left in too long to give up three doubles and a hit by pitch in a span of four batters
@KC: 3 runs in 7 innings (the left in too long as obvious as the TB game walking a couple before Albers allowed the homer)
CLE: 1 run in 6 innings
PHI: 5 runs in 5 IP, his first stinker, but only three hits and settled down after the big first inning to give his team a chance.
@BAL: 2 runs in 5.1 IP
DET: 2 runs in 5.1 IP
@TOR: COMPLETE MELTDOWN!!!!
So with a quicker hook in some of his other starts (not monday morning quarterback, but situations where the game threads were calling for pitchforks and torches), he has in 10 starts:
Gone 5 IP nine out of ten times
8 of 10: Given up 3 or fewer runs through five or more innings
1 of 10: Had a winnable bad start (Lackey/Clay victory)
1 of 10: Stinker, Didn't have it from the first pitch and had a stinker
Compared with a more casual glance through others:
Lester: 2 stinkers where he gave up 7ER and failed to go 5, one winnable bad start where he gave up 5 in through five
Beckett: 2 stinkers where he gave up 7ER and failed to go 5
Buchholz: Only one stinker where he failed to go 5, but tons of winnable bad starts (including a streak of 5 straight with 5 ER), including three times where he gave up 5 before finishing his fifth inning
Doubront: 1 stinker where he gave up 5ER and failed to go 5, the rest inefficient awesomeness giving us a chance every time
Dice-K: in 2011 2 stinkers in 7 starts, 2010 3 stinkers in 25 starts, 2009 2 stinkers in 12 starts.
Lackey: 4 stinkers in 2011 plus lots more pretty close where he just managed to be left in longer
Cook: 4 stinkers in 2011, 6 in 2010 including 3 in a row at one point
So, all this put together along with the fact that Doubront CANNOT make every start this year without having an insane innings jump for a promising 24 year old and will need to be skipped/DLed/shutdown rather than just given the ball, and I just flat out don't see how we can pull the plug on the Bard experiment.
If now is the time to skip Bard, let Dice-K start, let him be DL'd or work out of the pen for a game or two while he works on something mechanically, before being put back in the rotation, fine. But saying "nah, it isn't working" when 8 of 10 starts it has worked acceptably, and he clearly has had some approach issues with his transition which he seemed to finally acknowledge today, along with the normal mechanical/repetition learning curve we would expect, just seems very very silly.
Today he didn't have it, and that happens to almost every pitcher ever at some point. For some it is flat hittable 9-10ER meltdowns, for some it is can't throw a strike meltdowns, but this happens and one game very very rarely is an indication of a pitcher's ability (Ankiel excepted

). I would love to see him back at it and see how he steps up mentally and physically and deals with the adversity, and I would love to see the team and coaching staff rally around him and show confidence in him, but if it needs to be 15 days because "something wasn't quite right...we want to let him work on some things in the pen and build up his velocity again", that's fine.
To instead pull the plug at this point and insert him in a role (RHRP) where he hadn't performed well the last month he was in it, and where we have a glut of people including two good ones in Pawtucket, and another very good one getting closer to coming off the DL just seems pretty silly. Coming into today he was on a string of 8 of 9 time pitching effectively for at least five innings. That alone give him more leash for me and it would require a string of failure a lot longer than half an hour for me to change my mind.
I understand the issues with peripherals and velocity, but he has been able to be effective pitching about as badly as we possibly could imagine. Unless you think he is doomed to get worse and incapable of getting better, I ride that out and don't Joba him.