Connors d. Krickstein 3-6, 7-6 (10-8 ), 1-6, 6-3, 7-6 (4), fourth round It lasted 4 hours 41 minutes. Connors was 39 that day – and only ten months younger than Rosewall had been at the 1974 U.S. Open. Krickstein was 24. Krickstein was not the youngest player that Connors faced at this U.S. Open. Krickstein was a year younger than both Paul Haarhuis and Patrick McEnroe, but he was 3 years older than Jim Courier. The Courier-Connors semifinal, with a disparity of 17 years 11 months, was like the 17 years and 10 months separating Connors and Rosewall. Krickstein had beaten Agassi in straights in the first round here. The following are my own stats unless otherwise noted. Connors hit 70 winners apart from service: 15 FH, 10 BH, 21 FHV, 12 BHV, 12 overheads. Krickstein hit 27 winners apart from service: 12 FH, 14 BH, and 1 overhead. We've done a few matches in which Connors had more winners from volleys/overheads than from ground strokes, but never by such a large margin, nearly 2 to 1. He had close to 1 volley/overhead winner per game. Meanwhile, Krickstein hit just 1 winner in the air, a smash in the fifth set. Connors had just two service return winners, one from each side, both off Krickstein’s second serve. If he didn’t have more it was largely because Krickstein did not serve-and-volley. And Connors had few passes: just 3 backhands. Krickstein likewise had no return winners, though he did have 16 passes (including 1 lob from each wing): 11 BH’s and 5 FH’s. Connors' winners by set: 11, 22, 5, 10, 22 Krickstein's winners by set: 1, 9, 4, 4, 9 The second and fifth sets both went to tiebreaks, and they both had 22 winners by Connors and 9 by Krickstein. All 4 of the winners that Krickstein hit in the third set were baseline-to-baseline forehand winners that Connors chose not to run after. Connors gave that set up to conserve his energy for the fourth. The Kansas City Star had Connors at 106 unforced errors for the match, and Krickstein at 44. Per CBS, after 24 games Connors had made 55 unforced errors, Krickstein 24. Connors served 9 aces, Krickstein 7. CBS had the same counts as mine, except that the announcers were silent about Connors' last ace. The ATP site puts each man at 8 aces. There were no cases of near-aces that might have prompted judgment calls or caused confusion, so it just looks like a mistake by the ATP. Whatever the case, I can't recall a match we've done in which Connors had so many aces. Connors served 7 double-faults (the same as CBS's running count), and Krickstein 2. The ATP site has the same numbers. Connors got a return error from Krickstein 18 times -- of which I counted 4 as service winners. Krickstein got a return error from Connors 36 times -- of which I counted none as service winners. So Krickstein got twice the number of return winners that Connors did -- partly because he served more points than Connors did (203 vs. 171, per the ATP). But his first serve was also better than Jimmy's, and he was hitting it more aggressively, with a lower first-serve percentage (per the ATP). Some ATP stats: Connors served at 66%, making 113 of 171 first serves. Krickstein served at 57%, making 116 of 203 first serves. Connors won 75 of 113 points on first serve (or 66%), and 30 of 58 on second serve (or 51%). Krickstein won 76 of 116 points on first serve (or 65%), and 51 of 87 on second serve (or 58%). Krickstein was broken less frequently: 5 times compared to 7 for Connors. He won 27 games to Connors' 24. And Krickstein won more points overall: 193 to 181, a considerable margin. Connors won 5 of 24 break points, Krickstein 7 of 16: meaning that Krickstein saved 19 breakers, compared to 9 saved by Connors. That is probably the reason that Krickstein had to serve more points than Connors: he extended his service games by saving so many break points. SERVICE WINNERS CBS looks like it's tracking service winners and adding them to its total winner counts. At 5-3 in the first set, CBS had Krickstein at 4 winners. By my count he had hit only an ace and a backhand pass. At 2-all in the fourth set, CBS had Connors at 50 winners, Krickstein at 24. I have Connors at 41 winners and 6 aces, for 47 winners overall -- falling short of the network's count by three. I had given him 2 service winners up to that point. Similarly, I had Krickstein at 16 winners and 6 aces, or 22 total, falling short of the CBS count by two (I had given him no service winners up to then). The Kansas City Star had Connors at 86 winners for the match, and Krickstein at 38. By my count Connors had only 79 clean winners/aces, Krickstein 34. So it looks like Connors was credited with 7 service winners, and Krickstein with 4. This method is not exact, of course. My count of the clean winners/aces may not be exactly the same as the available counts. I wouldn't be surprised, in a match of this length, if there's a discrepancy due to errors. So this method of deducing how many service winners are in the official stats should be taken as approximate. My own count of the service winners was low: 4 for Connors and none for Krickstein. NET STATS Per the Times, Connors won 88 of 137 approaches (64%), and Krickstein 10 of 20. Yet in another article, the Times reports that Connors came to net 107 times more than Krickstein, which would put him only at 127 approaches. I'd go with the first article, which actually reports the total number of approaches by Connors. And 127 approaches, if Connors won 88 of them, would give him a winning percentage of 69%, somewhat harder to believe than the 64% rate in the former numbers. With 7 games remaining, CBS had Connors winning 61 of only 94 approaches (65%), Krickstein 9 of 18.