You can't compare Sampras with the others, his carrier was mostly during a time with 16 seed at slams. This is something not talked that often, but it changed the dynamics of slams more than anything imo.
You had more upsets in the early rounds, which explains why Sampras got so "few" top10 matches and why he lost more to non top 10 opponents. What you can take away from his record is that once Sampras went deep in a slam, found his groove, he was incredibly hard to beat. He has a incredible ratio in terms of QF played/GS won 14 out of 31
Maybe he got lucky a couple of times but that alone cannot explain 7 Wimb and 14 GS ..
Federer, Nadal and Djokovic benefited a lot from the 32 seed draw, it erased most of the danger of losing during the 1st week, there's almost no dangerous sleeper any more (also, the ranking system has changed towards something more stable). Yet their ratio is lower than Sampras because they hurt each other, which proves "the system works" (homogenisation, 32 seed), the better players meet more often because they get sweet, easy first rounds.
This being said, i like Federer and Djokovic records more than Nadal, because once again Nadal record is completely RG dependant. His record there is huge and at other slams it's really average (16w/14l). And, as noticed by uscwang, he lost more often to non top 10 players. Djokovic and Federer have good stats at the 3 others slams and are only average at RG where Nadal hurt both of them hard.
Personally i think it's some piece of evidence that the 32 seed draw helped Federer, Nadal and Djokovic getting the number of slams they have now and that in a 16 seed era, they would have less.
I appreciate where you are coming from, but I have to disagree
AELTC 2001 was the first grand slam to introduce the 32 seed draw, which is when Pete No 1 seed, played Roger No 15 seed, in the fourth round
We all know Roger won 7/5 in the fifth in that fourth round match
For Roger to be unseeded that year he would have had to have been No 35 in the world (Norman and Lapenti, who were seeded, pulled out, and their places were taken by players next highest ranked, who had entered the draw : Bjorkman and Levy, who became seeds 33 and 34, respectively, but having the advantage of being effectively seeds 31 and 32)
Now this is where it gets interesting : To get through to that fourth round meeting, Pete/Roger had identical routes :
1) They both won 3/0 in the first rnd
2) They both won 6/3 in the fifth in the second rnd
3) They both won 3/0 in the third rnd
The only difference is that Pete somewhat made a meal out of weaker opposition than those Roger faced
Pete was taken to 6/3 in the fifth by the English player, Barry Cowan (ranked 265 in the world)
Whereas Roger was taken to 6/3 in the fifth by Malisse, world no 53 (who at AELTC 2002 went on to the semis before going out in five)
Also, Roger beat no 33 seed Bjorkman in the third round, which in all likelihood was an excellent warm up for the Pete match...(And the other replacement seed, no 34 Levy, was drawn in Pete's third round section, but lost in the first round)
Going into that fourth round encounter,
All the pressure was on Pete, the No 1 seed
Roger had reached his 'allocated' position in the draw and anything else was a bonus, and he could play somewhat without pressure
Plus Roger had played the confidence boosting warm up match with Bjorkman, which can only have helped his match chances with Pete
For this reason I believe Roger went into that fourth round match with Pete, with an all round 55/45 advantage...
Which was ultimately born out by the result :
Roger winning 7/5 in the fifth (which then consequently ramped up more 'pressure' and 'expectation' on the 19 yr old in the quarters, resulting in Roger going out in a tight 4 setter)
Now, had the Roger/Pete match taken place in the quarter final, with both similarly getting through to that stage with identical developement, as in the case of their fourth round match, the 'expectation' and 'pressure' on the 19 yr old would have been higher than in their fourth round encounter, partly because Roger would have already 'overachieved'
before he even got onto the court for the Pete match-up
So I believe the very same match, as in their fourth round meeting, would have a shift in advantage bias 55/45 Pete...And Pete would have won that close match instead
Now, had the very same match taken place in the semi finals, with both going in with identical wins (so no great conditioning advantage, anywhere)
then the pre match pressure bias moves to advantage 60/40 Pete...And Pete being the ultimate big match 'mettle' player would win this in a tight four setter
Now, had the very same match taken place somehow in the finals, with both getting through with very similar routes, again as what happened in their real fourth round match
then the pressure bias indicator moves to advantage 70/30 Pete...Pete being the ultimate 'Rollerball' AELTC unbeaten King of AELTC Finals, playing against a 19 yr old Roger
before Roger is ultimately 'ready' to lift the cup, would result in a 3/0 victory for 'Rollerball' Pete
I think the same happens if you go 'down' a peg...
Had Roger/Pete, played that very same match in rnd 3/rnd 2/ rnd 1, something critically important is missing from Roger's 'confidence' resume....
And that is the helpful and experience driving wins over Malisse, in five sets, in rnd 2 (giving Roger his first five set win at AELTC)
As well as rnd 1 giving Roger his first ever match win at AELTC, 3/0 over Rochus
And critically importantly - the 'feeding-in' match win over no 33 seed Bjorkman, would have been missing for Roger
So Pete would go into these rnds 1/2/3 matches with Roger, with a 60/40 pre advantage and would have probably won any of these in 4 or 5 sets
So, it's interesting to note that AELTC 2001; was the very first slam where Roger was seeded, but ultimately he played Pete in just the 'right' round
Any earlier (or later) would have yielded no advantage for Roger
So you can see, Roger caught Pete, just 'right', and took advantage
But even if Pete had won, it would not have counted towards his top ten wins because Roger was ranked 15 in the world at the time
And in this instance, Roger getting to play the number 33 seed, a round before Pete, definitely helped Roger
From 1989 to 2000, men’s seeds were upset 263 times in the first two rounds of slams. Only 51 of those losses were to players in the top 32
Which means 80.61% of those 'upsets' would have occurred even with a 32-seed format, and presumably, some of the remaining 51 matches would still have resulted in upsets
And this was during the bulk of the Pete era
