In the game thread yesterday there was some interest in seeing some plots from Gagne from 2007 with a Boston and Texas split and then his 2008 stuff.  I mentioned that this year he was relying on his change a lot more in 2007 and maybe it wasn't as good of a pitch as it was before.  There was some talk about that being the problem in Boston as well.  So here is the plot of his fastball and change from 2007 while in Texas.

via baseball.bornbybits.com

I know it says curve but it means change I just forgot to change that when I ported the code over from my curveball study.  In any case, you can see how his release point looks pretty reasonable and the change mirrors the fastball well before dropping late.  The two tick marks are when the ball has traveled .075 seconds which is what Adair calls the information gathering time in his book.  Now here is the Boston data.

via baseball.bornbybits.com

Wow that looks very similar to me.  It appears like his change really didn't change so either his problems were with another pitch or mental or maybe just plain dumb luck.  Here is the 2008 data however.

via baseball.bornbybits.com

Hey I finally got the change up labeled correctly,  Anyway, yikes!  Gange's change is now crossing over his fastball and isn't dropping much at all compared to his fastball.  Again this data is uncorrected but I think there is a real problem here.  While the speed difference is still nearly 9 MPHs but if the batters can recognize the difference in speed they really don't have to lower the bat much.  I have no idea what needs to be altered in his delivery to fix this but it is pretty clear to me that something really needs to be fixed.  The crazy thing is he is throwing that crappy change so much here in 2008 compared to 2007 .