Click on the link Cog! :-}
Quoth McIntyre: "There is a substantial difference in trends in the “raw” data, which should have been reported."
Raw data usually has errors in it. That's why you don't use it to say anything definitive, because if you use uncorrected, unadjusted, un-QA/QC-ed data, hardly anything you say about it is scientifically defensible. So Peterson, who has been working with this data for decades, is unlikely to do something so unscientific.
While some of what McIntyre has done has merit, here I think he's pontificating beyond his realm.