Well, a true and neutral assessment of stats should come ahead of cricket history, cricketers and cricket books, though these factors should also be considered. On the other hand, a flawed assessment of stats means next to nothing. It all depends on how objectively the stats are analyzed.
Exactly. There might be a hundred million people in India writing tomorrow that Dhoni is the best batsman of all time, and someone who knows about cricket, Shane Warne, just put Tim May as the best finger spinner in his list.
I love to hear and read people's views, including books, history, analysis by respected historians, but in the end it has to be backed up. And Warne is not the exception. A guy many consider to be one of the foremost, if not
the most knowledgeable guy around (Benaud), and he didn't have a single West Indian bowler in his all time xi, his 2nd all time xi and his 3rd all time xi. The people he picked weren't bad bowlers, and he certainly made a convincing case for each of them. But 99% of others, if asked to provide THREE all time XIs, would certainly have at least one west indian fast bowler, no?
Tony Greig neglected to have Sir Jack Hobbs in his English all time XI. Thirty years from now, historians will quote Tony Greig and explaining why he is the font of all knowledge while the guys who look at stats will bang their heads.
Stats, properly interpreted, and taken in context with other stats of the same era, are the great equalizer. I said O'Reilly had a relatively high S/R. But that doesn't mean I can compare that S/R directly with Warne's. Or Trumper's average directly with Lara. That's not looking at them the right way.