Well usually in an ECL kit, one of the reagents is the substrate and the other one is hydrogen peroxide, which activates the substrate for breakdown using the HRP on your secondary antibody (Ab) so I genuinely doubt what you are seeing is the result of ratio difference. In order to make that conclusion you need to run your samples twice (on two separate membranes) and do a 1:1 and 2:1 ECL and if you observed such a difference then you are on firmer grounds.
I suspect what has happened in the above experiment is that you have done a 1:1 (with 1Ab and 2Ab not distributed equally on the membrane) then after perhaps not a very thorough wash, you have done another 1Ab and 2Ab incubation using the a 2:1 mix. I suspect what has happened is that in your 1:1 the left hand side has not received the Ab mixtures very well and when you did the 2:1 since the right hand still Ab attached to it (if you didn't strip the membrane) so the left hand side was "forced" to get the Abs hence again the Ab mixture and ECL not equally bound. It is obvious since one image is opposite the other so its the Ab incubation step that has gone wrong. Otherwise if you have done the second ECL incubation mixture straight after the first and didn't do a wash step, the same can result if the membrane did not receive the initial ECL mixture equally since you can get a blocking effect by the old ECL mixture and perhaps what has happened is that the H2O2 or the substrate was present in excess in the second incubation step hence amplifying the signal in the sections of the membrane that did not receive the ECL mixture very well in the first place, relative to the rest of the membrane, which still has ECL liquid on it from the last experiment (again if you haven't washed the membrane well as ECL can be active for up to 24 hours in some kits but after a few hours the activity usually reduces but can still be observed)
What you need to do is after ECL, wash the membrane with TBST for a few minutes followed by TBS, leave the membrane in the sun to dry so that the Ab slowly denature (although that doesn't always happen specially with high affinity Abs such as anti-tubulin and actin). After at least 24 hours, strip the membrane from Ab and redo the Ab incubation. The best thing to do is to do all the incubation steps in a sealed bag then make sure the membrane receives the ECL mixture equally and do the visualisation. I would say however since the membrane is unequally "contaminated" with the Abs then the best thing to do is to redo the experimental run and re-run the gel. Its a pain but you need to get to the button of this!