If you have followed my ramblings here I have pointed out there is a rather large but rarely consider assumption made with this curve fit. That would be "normal" for global land surface temperature would be about 8.3 C degrees.Since "today" which in that chart is 2011 and the chart has a temperature of about 9.9C, land response to Volcanic and CO2 changes in forcing is 1.5C degrees above "normal". The BEST curve doesn't consider anything but CO2, Volcanic aerosols and land temperature.
A couple of years later BEST produces a combined land and sea temperature anomaly product. No absolute temperatures this time and no repeat of the simple curve fit.
NOAA/NASA produce a sea temperature product that has ~18.4 C and the "present" temperature of the oceans. If you use a 70:30 sea:land ratio you can combine the two and the average absolute surface temperature would be 15.85 C degrees. I left two significant digits only because that seems to be in vogue today. 15.85 C with about 0.85 C of anthropogenic warming would imply that "normal" is about 15 C degrees. Temperature anomaly has a very tight uncertainty range, but absolute temperature, not so much. However, we have published data that would imply we have a very accurate estimate of "global" absolute temperature.
The CMIP5 model mean for the lowest forcing estimate provides an absolute temperature estimate of about 288K (15C) degrees. If you remember your greenhouse gas basics, 288K is "normal" doubling the amount of CO2 in the atmosphere should increase "normal" by 1.5 to 4.5 C. Since today's CO2 concentration is about 400 ppmv and "normal" is about 285 ppmv we are 400/285=1.40 we are 40% into a doubling or CO2 or about 50% into the forcing equivalent of a doubling due to the natural log relationship. So if current temperatures are effectively "normal" +0.85 C and should be "normal" plus 0.75 to 2.25 C you are very close to a tie as far as evidence of CO2 caused warming goes.
You can easily build a case that past Volcanic forcing cause reduces temperatures which take longer than expected to recover from or you can change "normal" so you can claim the is more warming than indicated. Drop "normal" enough and you can hit any sensitivity you like.
If you like the BEST simple fit, then "normal" would be close to the depths of whatever Little Ice Age depression there was. Then you can pick which paleo reconstructionist you want on your team.
With Michael Mann-o-Matic, king of the frozen North tree rings you your side, it is all CO2 all the time. With Oppo and the tropical ocean devotees on you side, today is about "normal" just like the early days of AGW simple GHE explanation indicated.
So it looks to me like a debate on, "There is no physical evidence that CO2 has caused warming", could be flipped to "There is physical evidence CO2 prevented cooling." So the debate would just keep on keeping on.
One thing you should not though. On the Oppo et al. 2009 overlay the "current" and pre-LIA temperature was about 28.5 C degrees. Temperatures greater than 28C in the tropical oceans produce deep convection which has a negative feedback ration on the order of 1.2, meaning SW reflection is about 20% greater than LW retention.
Since the Tropical SST and "Global" temperature anomaly correlate extremely well, that would be "evidence" that roughly 15 C is "normal" and that tropical cloud forcing tends to regulate temperatures in the "normal" range. So it would be a pick you facts type of debate.