Playoff Momentum—Mo' Money or Moe Howard? Slowly I turn….
by Mike Carminati
With the White Sox up two games to none in the World Series, the series moves to Houston tomorrow with the Atsros' backs almost against the proverbial walls. They are approaching the wall like Chris Burke circuitously tracking a fly ball at the fence in left. A loss in game three would be pretty big given that other then the Red Sox in the ALCS last season, no team has ever come back from 3-0.
One way to look at is that the Astros winning one in Chicago would have put the Sox at a severe disadvantage heading to Houston. So by narrowly winning each game, Chicago has just held their homefield advantage. One could say that the goal for Houston is to win at home and just take one of four in Chicago. They lost in two of those four attempts already but still have two remaining.
So is being down 2-0 a big disadvantage?
Unluckily, that is not readily apparent from the data at hand. Luckily, we had a day off and I was able to load the data. I took a look at every playoff series from the 1903 World Series through the 2005 League Championship Series.
There have been 64 best-of-seven series that started out with one team owning a two games to none lead. Of those 51 were won by the team that led by two games. That's a .797 winning percentage. Of those 51 series wins, 21 were sweeps, 15 were went five games, 8 went six games, and 7 went a full seven games. Of the 13 series loses, four went six games (meaning that the trailing team swept the four last games), and 8 went seven.
But how does homefield factor into those numbers? Of all 2-0 series, 43 of the leading teams were playing those games at home and 21 were on the road. Of the 43 home teams, 33 went on to win the series (77%). Of the 21 road teams, 18 went on to win the series (86%). So three-quarters of the teams in the White Sox's position went on to win the series.
OK, maybe there's nothing earth-shattering there, but then again it was just an excuse to take a look at the postseason game data and, more specifically, at the idea of momentum, which is bandied about whenever a trailing team wins a playoff game.
So I will now pose the question, is there any such thing as momentum in the postseason. Discuss
If a team wins a given playoff game, what are the odds that they will win the next? The answer is that those teams win 52.4% of the time (477-433). That's a bit better than a 50-50 coin toss. Maybe momentum does come into play, eh?
Well, you might see a hole in this logic. What is one team totally dominates the other? If there's a sweep, then momentum has very little to do with what's going on since the momentum never shifts.
So I filtered out all sweeps and re-examined the data. In non-sweeps, the team that won the previous game will win the next only 45.2% of the time (357-433). So much for momentum.
But maybe that's a bit unfair. Teams may trade a game or two at the beginning of the series, but when the series is on the line perhaps that is when good ol' momentum kicks in.
I took a look at all "brink" games, i.e., games that brought the victor to within one game of winning the series. Perhaps when a team wins after one of these "brink" games, like the Cardinals did on the Pujols home run after falling to a 3-1 deficit, perhaps that is when momentum rides in like the cavalry.
First, teams that win a brink game win the series 54.3% of the time (119-100). But we are concerned with the next game, when the trailing team has their collective backs against the collective wall collectively. Those teams won 49 out of 100 times (49%). That seems about as close to a coin toss as one can get. And teams that lost after the brink game still went on to win the series 59% of the time (59-41).
OK, but there are different situations that we are lumping all together here. A team that falls behind 3-2 can much more easily turn around and win in seven games than a team that trails 3-0.
So what happens when a team is trailing by one game in a series and is one game away from being losing the series? That is, they need to win the final two games of the series. If the trailing team wins the game after the brink game, thereby tying the series, how likely are they to then pray at the altar of momentum, that is, win the final game and the series?
Guess what? It's exactly even (24-24). Again, where's the momentum?
Using the Pujols scenarios from before, how likely was it that the Cardinals would be able to win the series. It seemed almost expected that they would, but of course, as we all knew they collapsed a World Series closer.
Teams that are trailing a given series by two games and have lost the brink game only to win the following game still lose the series 64.4% of the time (29-16). Momentum is fading faster than a Brad Lidge save opportunity.
Lastly, when a team is down by three games and then won the next game, they have won the series just once out of seven tries (the Red Sox in the 2004 ALCS).
So what does it all mean? It means that the next time that Tim McCarver prattles on about the momentum shifting from game to game, we will all know that it just aint so. Teams in close series trade wins, but so do heads and tails in coin tosses. There's no evidence that momentum exists except when Lidge is on the mound.