Nate, i'll start off by saying that you are a brave individual. You've laid your heart, soul, system and room along with all of its' non-linearities out for all to see and criticize. Kudo's to you for having the guts to do that. Having said that, i think that most folks would be utterly stunned if they saw the response of their system when obtained from testing like you did. I know i was when i first had access to upscale test equipment. Even just listening to the image shift when playing the Cardas sweep tone told me that something was VERY wrong. The funny thing is that moving over just one foot to either side and playing that same frequency sweep will produce DRASTICALLY different results. As mentioned above, that's why averaging of results becomes important.
In order for one of these devices to work and be properly interpreted, you have to take GOBS of readings. Something that most folks don't understand is that, if the speaker isn't very linear to begin with, you can't compensate for that with room treatments. That's why i've stressed speaker designs that take into account real world room acoustics with typical in-room placements. Most speakers, especially a lot of higher priced "audiophile approved" designs don't do this.
Other than that i don't know if you took any nearfield measurements of your speakers, but you should do that if / when you get a chance. This will tell you how much of what you are measuring is room based and how much is speaker based. As you get further away from the speaker, the room becomes a bigger factor. This is why some folks absolutely love and "preach the gospel" of nearfield listening i.e. less room interaction.
Other than that, your in-room response from appr 125 Hz up to appr 15 KHz varies +4/-5 dB's. This is actually not that bad compared to some that i've seen but could obviously use some help. Once again though, i don't know if you are fighting speaker problems, room problems or a combo of the two ( most likely ).
As to bass response, most rooms look absolutely horrible to say the least. Poor speaker design along with poor acoustics make for VERY "non-linear" response. For a peak at what the measured response is of someone else's system that has taken a lot of time in setting things up, take a gander at Joe's measurements regarding his "Critical Q subwoofer"
installation. As a side note, i would encourage those that are interested in learning about woofer design and / or the "sealed vs ported" debate to give this entire article a read. It is quite educational and deals with real world situations and the design goals / trade-offs that we have to deal with.
Back on topic now, Joe's system actually measures quite good, yet it is something like +6/-5 dB's from 20 Hz to appr 15 KHz. It should also be noted that Joe chose to use a very revealing method to test his system, so it would look much "flatter" if he had used the standard 1/3 octave method that many others use.
As such, Nate has nothing to be ashamed or embarrassed about. While his system looks "rough", part of that has to do with the size of the graph used. If we were viewing a smaller graph length-wise with a narrower vertical gradient scale, the in-room response could have been made to look "much better". Obviously, his system isn't perfect and needs work, but the scaling of this graph somewhat exaggerates the results compared to what we are used to looking at.
Having said that, I have a lot of respect for anyone that would have both the guts and humility to post potentially "embarrassing" measurements ( only to those that aren't well versed in actual in-room response measurements ) and then ask for help in resolving those problems.
Not to single him out, and i know that he won't take this the wrong way, but even with the calibre and cost of gear that he's using, one can see how easy it is to end up with a "non-linear" system. On top of that, it is also easy to see how one could have a very different "sonic conclusions" when making comparisons from one system to another. That is, each system is going to produce unique deviations in frequency response ( and associated observations in timbre and transient response as they are all inter-related ). This is why i and many others strive to promote "neutrality" in both electronics and speakers, as a neutral response removes as much of the individual deviations out of the equation from system to system as is possible. Obviously, minimizing the deviations from a "flat" response can be a LOT of work and very frustrating to say the least.
With that in mind, my suggestion is to NOT make any brash decisions until you can make some very thorough tests and analyze the results. That is, IF you ( or anyone else in a similar situation ) aren't happy with what you are hearing. Obviously, some folks are content to listen to what they have and think that it is "as good as it gets" without really knowing what they are dealing with whereas others strive to achieve the best results possible. It would appear that Rooze and Nate are folks that fall into the latter category. As mentioned though, when one first takes steps like this, the "shock to the system" that you get as results may be enough to scare one away from seaking the "sonic grail of neutrality".
If you were to "quit" now and simply enjoy your system, i would understand. At the same time, i know that once someone sees this type of information about their system and know what is going on, it is a hard thing to forget about. As such, i would advise others to NOT take the steps that Nate did UNLESS you have the heart, soul and courage of a warrior. I say "warrior" as you'll have one helluva fight in front of you once you see what you're really dealing with. Sean