Even with a partial deployment I guess the drag would have a significant impact on forward velocity.
Mmm. My gut feeling tells me that that might be a bad thing actually. If you have a partially deployed chute slowing you down, it may (will?) slow you down below stall speed so you lose wing lift. But at the same time the chute is not fully deployed so it may not fully cushion the vertical speed. Additionally, it will take time for the aircraft to dangle backwards, vertically below the chute so initially there may be an increase of vertical speed. In an EFATO scenario, that might just happen to be the point where you hit the ground.
In addition to this, somebody mentioned that if you're sitting more or less upright, and are properly strapped in, your body can sustain 50G in the horizontal, but only 15G or so in the vertical, because your spine is relatively weak in compression. I haven't done the sums (and they would depend on the size of the horizontal and vertical crumple zone anyway) but based on this it may well be that crashing horizontally into something at stall speed is more survivable than crashing vertically into something under a partially inflated chute. (But I'll take a fully inflated chute over an unrecovered spiral dive any day!)
So my gut feeling tells me that using the chute in an EFATO situation, when you are below the altitude for full deployment, might actually worsen the situation. Compared to gliding the plane to a more-or-less suitable landing spot.
But I'm not a Cirrus pilot or owner, and all I know about the CAPS system is from what I read on here. I would certainly hope that the Cirrus type-specific training that is on offer (and, I understand, is mandated by insurance companies) would go into the deployment of CAPS, including when NOT to deploy it. Just like turning back to the airfield or not, in an EFATO situation.
(Edited: It's PPRuNe that feels the need to explain the E-F-A-T-O abbreviation every time, not me.
)