Four days and nights at Fenway
 
For the past four days and nights, I've spent most of my waking hours getting to Fenway Park, sitting in Fenway Park, and getting home from Fenway Park. And I've spent some normally-non-waking hours doing those things, too.

With everything that has happened, and with the vast stretches of time over which those events occurred, a lot of thoughts have passed through my mind. Here are a few of them.

Before the Fenway portion of the series began, I was thinking that it could still get interesting. If Boston could take two of three at home, they'd force a game six in New York. Pedro Martinez would be the starter for that game, and if he could find a way to win, the series could go seven. The Yankees would still be favored since Boston would have to use most of its bullets just to get that far, but game seven is game seven, and you never know.

After New York broke open game three, I started thinking about the fact that no baseball team has ever come back from three games down in a seven-game series. It's not impossible, of course. It just hasn't happened yet.

Before game four, someone asked me what I thought about the Sox chances to win four straight. I thought about it for a few seconds and came back with an estimate of 2%.

If you assume that each game between two evenly-matched teams is a coin flip, the probability of a four-game winning streak is 1 in 16, or about 6.3%. That's assuming the teams are evenly matched, the odds of winning each game are the same, and the results of one game do not affect the odds for future games.

Of course, we don't know whether this simple model reflects reality. History tells us that 25 teams have fallen behind three games to none. None of them has ever come back to win, and only two have even forced a game six. Perhaps those series didn't involve evenly-matched teams. Perhaps losing three straight reduces your chances to win subsequent games by wearing out your pitching staff or by creating a psychological barrier that cannot be overcome. Or maybe the model just doesn't apply.

My view was that Boston's chances were lower than 6.3% because, with Schilling injured, the teams were no longer evenly matched and Boston's bullpen was in terrible shape. So I arbitrarily cut it by two-thirds when I came up with that 2% figure.

Before game five, I was asked the same question. Using similar logic, I put the chances at 6%. This time, my starting point was 1 in 8, or 12.5%, for a three-game streak. I reduced that number by half because Schilling was presumed to be out of action and the Yankees would have home field for two of the remaining three games. It wasn't clear whether Boston's bullpen was in any worse shape than New York's, so I didn't include any carry-over effects in my estimate.

The first two innings of game five made me think that Pedro Martinez was "pitching to contact", as my friend Gil Lebreton once described a start by Kenny Rogers of the Rangers. It made sense. The bullpen was depleted, and Pedro knew that he had to go deep in the game if the team would have any chance. He seemed to be throwing more fastballs and daring the Yankees to hit them, presumably hoping those batted balls wouldn't do too much damage.

He got away with it in the first inning when Hideki Matsui's long drive was caught at the bullpen fence for the third out. He didn't get away with it in the second inning when Bernie Williams hit the first pitch into the right field stands. But he did end that frame with a 2-1 lead and a pitch count of 22. The first time through the batting order, the Yankees saw only 2.4 pitches per batter, about 40% below the norm for them.

Even though the results were pretty good so far, it was clear that Pedro didn't have the electric stuff he showed in his two previous starts. I never felt that he was on top of his game, and I assumed he was a little tired.

So I wasn't surprised when things began to change in the top of the third inning. Pedro threw ball one to the first four hitters he faced, and while he got out of the inning without allowing a run, he needed 26 pitches to do it. Meanwhile, Mike Mussina was settling down after a very long first inning (two runs and 33 pitches), and their pitch counts began to converge.

By the sixth inning, I was thinking that Pedro would be done by the end of the seventh. He didn't make it that far.

The sixth inning began innocently enough, with Williams flying out to left and Posada reaching on a chopper that barely made it over Pedro's glove and then died in the grass behind the mound. Ruben Sierra followed with a single, but when Tony Clark struck out to bring Miguel Cairo to the plate with two out, it seemed as if Martinez would escape again.

But Martinez hit Cairo with a pitch, bringing Derek Jeter to the plate with the bases loaded. When a future Hall of Famer makes a bad mistake to a #9 hitter in such a critical situation, you have to believe he's running out of gas. Sure enough, Jeter lined pitch number 100 for a three-run double that gave the Yankees a 4-2 lead, adding to an already outstanding postseason resume for the New York shortstop.

It could have been a lot worse. You just know that Boston manager Terry Francona wanted to take Pedro out of the game there, but who was there to turn to? As it happened, Martinez hit Alex Rodriguez with a pitch, walked Gary Sheffield to load the bases, ran the count to 3-1 on Matsui, then watched as Matsui hit the next pitch right on the screws but right at Trot Nixon for the third out.

Starting with the Cairo atbat, Pedro threw 16 pitches, and only three of them were strikes. One was Jeter's double. One was a 2-0 pitch that Sheffield was probably taking all the way. And the third produced the Matsui line drive, a laser beam that would have won the series if it hadn't been hit at someone.

Little did we know that we still had eight-plus innings of baseball in front of us.

Mussina got the Sox in order in the bottom of the sixth, but this may have been the turning point in the game. Nixon lined out to center for the first out, but Mussina needed seven pitches to get the job done. Next, the switch-hitting Jason Varitek -- who stood in the right-handed batter's box against Mussina because he was only 4 for 48 lifetime against the Yankee right-hander -- fouled off four two-strike pitches before grounding to third. Bill Mueller flied out to end the inning, but not before pushing Mussina's pitch count over the century mark.

When Mussina was lifted after giving up a leadoff double to open the bottom of the seventh, I was very surprised to see Joe Torre go to his pen so quickly. Given that Mussina had only thrown 95 pitches in game one and was working with an extra day of rest, and that Tom Gordon and Mariano Rivera had each thrown two innings only 20 hours before, I thought Mussina might be good for 120 pitches.

And so it became a battle of the bullpens, and what a battle it was.

I wasn't surprised when Boston was able to tie the score in the 8th. The Red Sox were the league's most prolific run-scorers in 2004, they were playing in a hitter's park, and the Yankee relievers were tired. I was surprised that an equally-fatigued duo of Mike Timlin and Keith Foulke were able to hold the deficit at two runs against the league's second-best lineup.

With the game tied, there are plenty of opportunities to do some second-guessing on the tactical moves made by the two managers. I'm not going to do that, because I don't believe it's appropriate. I am, however, going to do some first-guessing.

I'm going to ask you to believe me when I say that each of these thoughts were formed before I knew how things turned out. I don't claim that my approach would have fared any better than the moves made by Torre and Francona, because there's no way to know that. Which is precisely why I don't believe in second-guessing.

When Jeter came to the plate in the top of the 8th with Cairo on second with nobody out, I was thinking bunt. A three-run lead in that situation would be huge, Jeter's an accomplished bunter, and I thought A-Rod and Sheffield had a very good chance to cash in that run with a tired Mike Timlin on the mound. Jeter got the bunt down, but A-Rod struck out, Sheffield walked, and Foulke came on to retire Matsui. But I blame the A-Rod strikeout, not the decision to bunt, for the failure to score.

In the bottom of the 8th, Ortiz made it 4-3 with a leadoff homer and Kevin Millar drew a walk. Francona made the obvious move and brought in Dave Roberts to pinch run, just as he had done so successfully in the wee hours of the morning. (Yes, game four ended earlier the same day.)

In the previous game, Roberts stole second, but this time I was thinking that you don't try to steal. With the bullpen in tatters, I think you have to play for the win right here, right now, and that means trying to score twice, not once. Gordon pays a lot of attention to Roberts, and perhaps because of that, yields a single that puts runners on the corners with nobody out. Exit Gordon. Enter Rivera. Varitek hits a sac fly, and the game is tied.

Doug Mientkiewicz comes on for Millar defensively, and wouldn't you know it, the Gold Glover makes all three outs in the ninth inning. (OK, OK, they were all towering popups that anyone could have caught, but we're not going to let that get in the way of a good story, are we?) Sandwiched between those outs were a Sierra walk and a double by Tony Clark on a ball that skipped into the right-field stands by the narrowest of margins. If that ball doesn't leave the field, it's almost certainly 5-4 Yankees.

In the bottom of the 9th, Damon legged out a broken-bat grounder for a leadoff single. I'm thinking bunt here, mainly because Roberts barely managed to steal second off the same pitcher-catcher tandem in game four, and Damon doesn't have as much speed as Roberts. I'm also thinking bunt because Orlando Cabrera has a good chance of executing the play successfully, thereby removing the possibility of a double-play grounder with Ramirez and Ortiz due up and giving the team two shots to single in the winning run. And if Torre chooses to walk Ramirez to set up the double play, I'm comfortable putting the game in the hands of Ortiz and Nixon. As it turned out, Damon was caught stealing and the Sox went quietly.

In the top of the 11th, Francona surprised me twice. First, he lifted Bronson Arroyo despite a strong 10th inning, preferring to have Mike Myers face Matsui. Myers had walked Matsui on four pitches the night before (or was it earlier that morning?), but this time he got the job done.

The second move was one you hardly ever see, lifting one lefty (Myers) for another (Alan Embree) with a series of switch-hitters and right-handed batters due up. I like both moves, though, since Arroyo may be needed for game six and/or seven, and Myers had thrown 42 pitches in game three. If Myers was only able to face one batter, this was the time to use him.

Actually, that reminds me of something else I wanted to mention. In the 6th inning, when Pedro was out of gas, Myers was up in the bullpen, presumably to face Matsui if it got that far. Sure enough, Matsui did come to the plate, and I was very surprised when Francona left Pedro out there. The situation reminded me a lot of the 8th inning in 2003 when Alan Embree was ready and Grady Little chose to let Pedro face Matsui.

Why warm up Myers if you're not going to use him? All I can think of is that when Myers started warming, Francona couldn't have known that the bases would be loaded for Matsui. When it happened that way, he may have thought back to that four-pitch walk and decided he had a better chance with Pedro.

As you know, last year Matsui drilled a double down the right field line, setting the stage for the Posada blooper that tied the game. This year Matsui hit the ball just as hard but to a spot where Nixon could make the catch for the final out. It is indeed a game of inches.

Managers are vilified when their judgment calls don't work out. They don't often get enough credit when they do. We'll never know whether it was "correct" to leave Pedro in, but we can't prove that it was wrong, and it did work.

In the bottom of the 11th, Mueller led off with a single, bringing Mark Bellhorn to the plate. I turned to my friend and said, "they'll probably bunt here, but I wouldn't do it." Even though Bellhorn was batting only .100 in the postseason and seemed like a natural candidate for a sacrifice, my feeling was that you're facing Paul Quantrill, who posted a 7.09 ERA in the second half and gave up a game-winning homer earlier the same day, and you can beat him without getting fancy. Bellhorn had already hit the ball hard a couple of times in the game and was entirely capable of working a walk that would move the winning run into scoring position with nobody out.

Bellhorn bunted foul twice and then singled, moving Mueller to second. There's no way to know whether he would have singled had he not been asked to bunt first, but I still took this as validating my feeling about this situation.

Bellhorn's single brought Damon to the plate in another obvious bunt situation. Again, I would have refrained from asked Damon to sacrifice. With a struggling pitcher on the mound, I'd rather take three shots at driving in the winning run, and Damon has enough speed to stay out of the double play. If he hits the ball on the ground, chances are you'll have a runner on third with one out anyway.

Damon did bunt, badly, popping up the first pitch. Quantrill tweaked his knee on the play, whereupon Esteban Loaiza came on to end the inning on a double-play ball.

Moving ahead to the top of the 13th, we're into our sixth hour of play and Tim Wakefield is now pitching for Boston. Which Wakefield are we going to see, the one who pitched brilliantly in two starts in last year's ALCS, or the one who gave up the Aaron Boone series-clinching homer in game seven?

Turns out we're seeing a guy with a knuckleball that's moving all over the place, so much so that Matsui reaches on a third strike that gets away from Varitek. Then Matsui reaches second when Varitek misses another one. And after Posada is intentionally walked, Matsui gets to third on yet another passed ball. The Sox aren't going to lose the series on a sequence of four passed balls, are they? That would be a new one.

I found myself wondering whether it was time to bring in Doug Mirabelli to replace Varitek. The advantages are that Mirabelli has been Wakefield's personal catcher for two years and, unlike Varitek, hasn't caught 25 innings in the past 26 hours. But can you risk bringing in a guy who has seen action in only one game in the last two weeks? Under those circumstances, would you be shocked if he muffed a pitch, too?

I wimped out on this one, never really reaching a conclusion in my mind. ("Is that your final answer? Gasp! I don't know.") Francona, who doesn't have the luxury of waffling, chose to leave Varitek in the game. It worked. They emerged from the inning unscathed.

In the Boston half of the inning and in the top of the 14th, both teams were set down in order. And my mind, which was working on 3-1/2 hours of sleep, was starting to play tricks on me.

I was wondering who's available to pitch if the game goes another few innings. For a moment, I thought Boston might use Pedro out of the bullpen, because with the game closing in on six hours, it felt as if three or four days had passed since he last pitched. Swear to God, I actually forgot for a moment that Pedro had started this very game.

As the bottom of the 14th is getting underway, I'm wondering how on earth the Sox have managed to go scoreless for four innings with Felix Heredia (6.28 ERA in the regular season), Quantrill (he of the 7.09 second-half ERA), and Loaiza (5.70 regular-season ERA) on the hill. And I'm thinking that if they can't beat that crew, they don't deserve to win.

Well, they finally got to Loaiza in the bottom of the 14th, with Ortiz saving the day at the eleventh hour. Literally. The game ended on the stroke of 11:00, five hours and forty-nine minutes after it began. (The NLCS game, which started three hours after this one, was already in the 8th inning.)

A few stats from the three-game home stand:

- time of game: fifteen hours, eleven minutes

- pitches thrown: 1299, 636 by the Yankees, 663 by the Red Sox

- number of times a batter stepped out of the box to ask for time: 612 (yeah, I'll admit it, I'm just guessing at this one)

Oh, and by the way, the Red Sox "stole" both of these wins. In September of 2003, I wrote an article for ESPN.com that identified games in which a team emerged victorious despite being outproduced. Production was measured in total bases and walks. I called them stolen games, as in, "we stole that one."

In that article, I pointed out that (a) most teams were around .500 in stolen games, (b) the Red Sox had by far the worst win-loss record in stolen games, and (c) the Yankees won the 2003 season series from the Sox because they stole two games from Boston that year.

I ran the stolen-game numbers again this September. Lo and behold, the Red Sox were once again the biggest losers in these contests, having stolen only a couple of games and having given away something like ten of them.

But the Yankees produced more total bases and walks than the Red Sox in both game four and game five. While New York was stranding a boatload of runners, Boston was finding a way to manufacture the tying run against the best postseason closer in baseball history, not once, but twice in succession. Quite the role reversal, wouldn't you say?

Looking ahead, I can't give the Red Sox more than a 10% chance to finish off a miracle comeback. Visiting teams win about 46% of the time, so even if you assume these clubs are evenly matched and equally rested, there's still only a 21% chance (.46 * .46 = .2116) that the Sox would win twice.

But that's optimistic, I think. In game six, the Yankees are starting Jon Lieber, who has been pitching very well for several weeks and has dominated the Red Sox twice in that stretch. Meanwhile, Boston is going with Schilling, who a few days ago seemed more likely to be climbing onto a surgical bed than the mound in Yankee Stadium.

And isn't it interesting that the Yankees are now counting on Lieber to seal the deal? Last winter, all the talk was about Kevin Brown and Javier Vazquez. Brian Cashman must be feeling very good about signing Lieber when he was still a year away from being healthy enough to pitch. Sometimes an insurance policy can pay off in a big way.

As I write this, Boston and New York have played 50 games that count since the start of the 2003 season, and each team has won 25 of those games.

If, at the end of the ALCS, the number of games stands at 51, the Yankees will be up on Boston by one win and two World Series trips.

If that number is 52, baseball fans will have been treated to a dramatic seventh game for the second year in a row. If the teams are 26-26 at that point, New York will still be ahead by two World Series visits to none. But if it's 27-25 Boston, the Red Sox will have accomplished something that has never been done before.

And after all is said and done, whether that's tonight, tomorrow, or (in the event of a rainout) Thursday, there's still a World Series to play. Will either of these teams have anything left?

posted by Tom at 6:03 PM