I'm curious why you feel you can interpret his data better than he can, since he clearly disagrees with you. Are you assuming that he is ignoring this bias himself, despite him describing it clearly in his graphs and his results? Did you factor in any of his mentioned caveats, such as the "divergence problem" of recently-declining tree-ring sensitivity (accounted for in Fig.2 B)?
I'm sorry, but I'm not interpreting anything when I observe that Fig 3 shows the Had and CRU instrumental records hitting 0.8C by 2000 and NONE of the reconstructed lines sitting higher than 0.2C. Nor am I interpreting anything in observing that the EIV reconstruction exceeds 0.2C multiple times historically. That is a statement of fact about the graph.
Here's the extent of my own 'interpretation' of Mann's graph. Imagine that the HAD and CRU(red and grey) lines are removed from the graph. The reconstructions you are left with show the temp since 1900AD rising, but still lower than 1400, 1000, 800 and close to 600 and 400.
Yes, I too looked at the data, and unless you're just slapping an arbitrarily-large boost onto the proxy data, I don't see anywhere that it supports your own conclusion. E.g. the CPS Land proxy slightly overestimates the instrumental record (in the period of overlap of Fig.3),
The under estimation problem SHOULDN'T exist in Fig 3. As Mann stated, the systematic underestimation was when calibrated to early years and then projected onto later years. In Fig 3 the calibration will have been to the entire instrumental record. My observation above though stands, the reconstructions since 1900 ARE NOT higher than the reconstructions themselves back in 1400, 1000, 800.
. The EIV proxies barely overlap the instrumental record at all, but you could perhaps assume about a 0.2C underestimation where they do (though this is not well supported by Fig.2 C/D)
No, assuming an understimation in Fig 3 would be wrong, As noted above that only occurred for early calibration and Fig 3 was not calibrated that way. The early calibration was for separate validation testing as Mann performed for Fig 2. As for Fig 2 c/d supporting a bias with the EIV method, Mann had this to say: Interestingly, the problem is greatly diminished (although not absent—particularly in the older networks where a decline is observed after 1980) with the EIV method, whether or not tree-ring data are used
It seems to me that you're taking an out-of-context quote about finding a bias, making an unsupported assumption as to how big that bias must be, then reinterpreting his conclusion to suit yourself
At the risk of repeating myself beyond all reason, I'm assuming NOTHING about the size of any bias. I am observing the fact that the reconstructions themselves as the represent the warming since 1900 do NOT show unrivalled temperatures, but instead reflect a current warming that was matched by the reconstructions multiple times earlier. It is only when including the instrumental record that the comparison becomes anomalous. That is a statement of fact with no interpretation, bias or nuance involved. My sole reason for noting the early calibration late verification bias observed by Mann was that it is a suggestive explanation for this difference between reconstructed and instrumental. The observation stands regardless.