Gah. This giant thread, and NOBODY has actually looked at the statistics that everyone's arguing about, Agassi's win percentage when returning first serves?
Sheesh, people, the data's right there. You want to know how good Agassi's first serve return is? LOOK AT THE PERCENTAGES HE WINS WHEN RETURNING FIRST SERVES. ATP has great data on this.
This isn't a thread about whether Agassi is a good returner or not. He is. It's about, specifically, whether his FIRST SERVE RETURN is as good as indicated by the occasional flashy winner he hits off of it - is the tradeoff between hitting some great returns at the cost of getting aced a lot worth it. It's not about whether he can beat all those big servers - he can. It's whether he beats them by crushing their second serve return and being ok on their first, or by both. Quotes that say "AGASSI IS A GREAT RETURNER" or stats that say that aren't relevant, because the question is which PART of his return is great - first serve return, or second serve return, or both.
Compare Agassi to, say, Goran Ivanisevic. Look at Ivanisevic's win percentage on first serves and win percentage on second serves.
Next, compare someone else to Goran Ivanisevic. Say, pick someone else elite - Pete Sampras, who is not as good a returner as Agassi, but is certainly as deadly of a player overall. Compare Sampras to Ivanisevic, look at win percentage on first serves and win percentage on second serves.
Now, I certainly expect that Andre would have a better win percentage on returning second serves. But the relevant question is
1) Does he have a better win percentage on returning first serves
2) And, if so, how much better is it?
Here's the data. Agassi and Ivanisevic have played six full matches against each other (not counting the 3-0 RET match). In those matches, Goran Ivanisevic won 86, 86, 87, 86, 88, and 78 percent of his first serve points, respectively. He won 39, 58, 30, 42, 57, 45 percent of his second serve points, respectively.
Sampras and Ivanisevic played a total of 18 matches against each other. Of those, the ATP had statistics for 12 of them. It those, Ivanisevic won 76, 87, 94, 89, 77, 82, 87, 81, 77, 80, 83, and 84 percent of his first serve points. He also won 51, 34, 51, 47, 52, 55, 52, 25, 40, 44, 56, and 33 percent of his second serve points, respectively.
I think this data supports the idea for Agassi being a great second serve returner, but not as great of a first serve returner against the big servers. You can pretty much read off Agassi's wins off of the second serve win percent - the three matches of that list that Agassi won off Ivanisevic were ones where he absolutely CRUSHED the second serve return - Ivanisevic could manage win percentages of a measly 39, 30, 42 on his second serve in those matches! Whereas, in nearly all of the matches between the two (except the very first one, back in 1991), Ivanisevic consistently would win between 86 and 88 percent of his first serve points, regardless of whether he would win or lose the match.
Sampras, not known as a "great" returner (though I'm sure he was, he had to be good at all parts of the game, just not as exceptional at it as Agassi was), got a better win percentage off of Ivanisevic's first serve (plenty of below-86 matches.)
This isn't quite a fair comparison, though, because Sampras had a better overall record against Ivanisevic - 12-6, of which one was a walkover so it's more like 12-5, and only one of those wins coming post-1993. If you look at the second serve win points, Agassi also did a tad bit worse on second serve return than Sampras did, on average! That's a bit weird to me.
Matches against Boris Becker are inconclusive, In the AA vs BB matches that ATPtennis.com had statistics for, AA went 7-1, letting BB win an average[1] of 72 percent of his first serve points and 41 percent of his second serve points. Pete Sampras, who went 11-7 against BB in the matches ATPtennis had statistic for, let BB win 78% of his first serve points overall and 47% of his second serve points.
Agassi was a more effective returner [2] - he was 6% more effective on both first and second serves, which would indicate that he's equally good at returning first and second serves compared to others in his field.
Against Richard Krajicek, another big server of the 90s:
Agassi: overall record 3-3 [3]. Percent of first serves won by RK: 81. Percent of second serves won by RK: 54.
Sampras: overall record 5-4 (of those with stats). Percent of first serves won by RK: 82. Percent of second serves won by RK: 51.
In this matchup, Sampras was a better 2nd serve returner and a slightly worse first serve returner, even though he had a worse record against RK overall? WTH?
OK, I'm not doing any more, I think I've come to my conclusion. Andre Agassi was as good at returning first serves as he was at returning second serves - the appearance that he got aced a lot was probably made up for by his aggressiveness on the returns which he DID get a racquet on. When comparing him to another player of his generation, I don't see a marked gap between 1st and 2nd serve return statistics that would indicate that his return prowess came ONLY from his second serve return stats. This effect may still be there for individual players, depending on individual matchups, but it's not looking like it's an "Agassi vs any big servers" effect, if it's ever there.
(Of course, these are only three players. Of these, Ivanisevic indicated that the gap is there, and RK and BB indicated that there is no gap. If someone wants to do more players, I'd certainly be curious to hear - Rusedski? Is there a lefty effect? There's far more big servers than these. But I've looked at enough to conclude that the effect that the OP was talking about, if there, is not very big, and certainly not obvious, and I don't want to look at any more, I've got better things to do. Oh! And all this is comparing Agassi to Sampras, but that's also just one comparison - if someone wants to do a comparison to a good return-blocking player of Agassi's generation, that would be informative as well.)
[1] (Note about my averages - I'm taking the averages per match, and then averaging those to get an "overall" average. I realize that's not the right way of doing it and I should be weighting it by total points. If someone else wants to do that work, go ahead, I don't feel like it.)
[2] Yes, I realize it's not a measure of how good he is as a returner, it's how good he is overall on returning first and second serves, so it takes into account his ground game and passing shots as well. It's the best proxy measure I can think of, though.
[3] I am excluding the match where RK was just coming back in a brief comeback attempt after not being on tour for two and a half years. Not even sure RK would count as a big server the way he was playing in that match