Take a look at this chart:
HowSTAT! Player Batting Graph
Divide every number on it by two.
After doing this, consider that we might still think of this player as an all time great.
Consider also, that the number of other batsmen in history who have averaged more than this over the course of their careers (min 20 innings) totals 41. That means, were this hypothetical "half Bradman" to have played, he would still have an average in the top 50 of all time, in over 100 years of history.
There has been some discussion as to the concept of a "golden patch" that Bradman may have been in. What is not known is that Bradman actually did have a bone fide golden patch. Between 1 Jan 1930 and 31 Dec 1932 (that is, the calendar years 1930, 1931 and 1932), Bradman scored 11 centuries in 16 tests (20 innings) at an average of just shy of 130. This is high enough to suggest that this was a statistical anomaly in an otherwise "average" career averaging nearly 100.
Currently, there are 38 test match batsmen who have scored more runs than what Bradman did in his career. The least number of matches that a player who has scored more runs than Bradman has played was Wally Hammond, who had a career spanning 85 tests (140 innings). This is 33 tests, or 60 innings more than Bradman.
Now I am not saying any of this to belittle the other batsmen mentioned here. They are all very fine batsmen. There have only been 42 in the 100+ year history of cricket who have managed to maintain a career average of over 50 (and many of them are still playing and thus their averages may drop to below this mark). The 43rd highest averaging batsman of all time is Alistair Cook. Bradman scored twice as many runs as Alistair Cook does, on average before he got out. If I could pick two Alistair Cooks and play a side of 12 men or one other batsman, even one as good as Tendulkar and play a side of 11 men, I'd take the two Cooks any day of the week.
Sure, now maybe there is a valid argument that says people have grown stronger and faster over the last three generations. That is a valid point. But with Bradman we're not talking about stronger and faster as we would understand it. We're talking about Spiderman superiority here.
We could exhume the corpse of the great Don Bradman and have him play another 40 tests for Australia. Assuming that he would be timed out in all 78 innings (the last two would not need to be played because Ricky Ponting would make a double century in the 39th Innings which would give us a lone innings victory, and there would be a washout in one of the games), Bradman would still average more than most batsmen in history.
Let's forget, for a moment about Bradman's average. Let's, for a moment, forget about how many runs he made and look at what his contemporaries had to say about him. Actually let's forget that too, because to remind the gentle reader of what was so patently obvious to any individual who saw the man play would be rubbing the proverbial salt in the wound.
The real problem here stems entirely because Bradman is far too superhuman to be real. He was so much better than everyone else that he must have been a caricature. An invention by statisticians to prove a point to their students about outliers and statistical impossibilities. He was a phantom. Seen by many, his exploits recorded but entirely unbelievable by any rational being. If, instead, his average was 75 or even 80 then he would be undisputedly the best player of all time. Not only that, but he'd be within the realms of possibility. But with an average of 99.94, he is nothing more than an old wives tale. His exploits simply unfathomable to those who never saw him play.
So can we please go back to the Lara/Tendulkar holy wars? There is less than a standard deviation between those two players, which makes for far more interesting debate.