Those are interesting facts you point out; I hadn't noticed those. Although in whatever little opportunities Tendulkar got against McGrath, he did fine apart from lacking a dominant series.
Tendulkar in Donald's presence was weaker. I don't give too much weight to against Akram because it never looked to me like Akram was going to put Tendulkar in much trouble whenever they met including many limited over games.
Apologize for briefly changing the topic back from Warne. Just to clarify, I didn't notice these facts because I did some stat-digging to put Tendulkar down. I was enamored with quality fast bowling after having watched cricket in late 1970s and especially 1980s. It is the main reason I follow the sport. The only time I really watched batsmen like Lara or Tendulkar or Waugh was when they came up against a great fast bowling attack. That's why I know on top of my head because I watched very little of Tendulkar simply because (comparatively) he rarely came up against great fast bowling. And whenever he came up, while he never ever had a massive series failure like for example Dravid did in Australia in 1999, I never recalled him having a stand out series either.
There is one more thing I need to tell you because you bring this up constantly about how a batsman was never troubled by a great fast bowler therefore it wouldn't matter. That's not true. Effect of a great fast bowler (with good supporting attack) on a batsman need not always be direct. It is not necessary that the bowler dismisses the batsman on majority of the occasions.
McGrath dismissed Lara 15 times, but it was 15 times in 46 innings! In fact, McGrath got Tendulkar with the same frequency as well (only 6 times in 18 innings). Even Donald got Tendulkar only 5 times in 20+ innings. So if you look at the whole picture, dismissals - even against bowlers that trouble batsmen - are not always as frequent as they are perceived to be.
However there is simply no denying the impact that McGrath's presence had on both Lara's and Tendulkar's figures whenever they came up against him. It is not a mere coincidence that average of Tendulkar collapses from 70s in McGrath's absence to 30s in McGrath's presence. Similar collapse is seen for Lara & other Indian batsmen too. It need not be that McGrath gets them each and every time. They can even get out cheaply without facing a single delivery from McGrath. Aussie attack can run through or trouble batsmen at the other end, or McGrath can stifle scoring so much that batsmen are forced to take more risks against other bowlers. There are several factors that come into play.
Converse question also needs to be asked here. How well, Lara or Kallis or many other batsmen of 90's/00's era, would have done in Australia if they faced "McGrath-less" Aussie attacks on 30+ occasions, and if they had to face McGrath just 6 times (that too only in Adelaide, Sydney & Melbourne)? I am certain they would have made merry just like Indian batsmen did. Don't get me wrong, I am not trying to belittle Indian batsmen here. Batsmen just do their best against whatever attack/conditions they face.
Lara averages nearly 70 against Australia without McGrath. I don't see any reason why he would not have done extremely well in the imagined scenario.
Quality of opposition obviously matters as I said. But as you said, 6/30 innings isn't a lot at all and certainly not enough of a sample size to draw a conclusion from the numbers alone. Fwiw average in X country can be misleading too. Border played 3 tests in SA in 1992 and averaged in the 30s. Does it mean anything? Not at all. It's the same with average vs x bowler unless we're talking something like Anderson vs Warner where they face off regularly. What makes it even more flawed is that a major part of batting is playing out the tough bowler and getting runs at the other end. Also I recall Lara getting dropped very early on in that 213 off McGrath. In a 6 innings sample size that can blow up the numbers significantly. It's just a crap stat in general that tells you nothing.
McGrath made the Australian attack what it was though. There's absolutely no question about that.
Have no reason to disagree with all the facts you've put forward re: Tendulkar v Lara though. It's just the particular average v x bowler stat I dislike.
Tendulkar played 18 innings against McGrath overall and averages in 30s.
6 innings were specifically against McGrath in Australia.
18 is still a decent number (although still much less than Lara's 46 innings with McGrath in the opposite side).
In any case, my fundamental point is Tendulkar didn't face great attacks anywhere as often as some of his contemporaries did. He never faced peak Wasim-Waqar, played just 1 series (6 innings) against Ambrose-Walsh, played very little against McGrath. This is a fact.
To my mind this has to be factored when stats of Tendulkar are compared with those of his contemporaries. I was attempting to bring this out when I was making "average vs bowler" stat. Yes, I agree with you that this stat could be even more misleading because of very small sample size. But I don't know how else to can you make the comparisons genuinely equal.
Aussie attacks/conditions Tendulkar frequently faced in his career cannot be equated with the Aussie attacks/conditions that Lara faced in his career, even though their eras completely overlap.