Playing Off of Postseason Odds


The Twin Cities are abuzz with Twins playoff chatter, the sort of September confidence that can only come from three straight division championships. Truly, the Twins have been able to coast for long stretches, especially in 2004, while still finding a way to sneak into the playoffs. The last couple of years, they got hot at the right time, just as other teams (particularly Chicago) collapsed. Through a gaggle of lackluster performances, though, Minnesota finds itself in a more difficult situation in 2005, needing to catch and pass three teams in a little over a month to seize the Wild Card for the first time.

Using a complicated simulation metric, Baseball Prospectus estimates that it will take between 91 and 92 wins to claim the AL WC, giving the Indians the best shot to do so. To win 92 games, the Twins would have to finish a red hot 26-12 over their last 38 games, a stretch even if they continue their hot play from the last couple of weeks. Although the Indians, Yankees and A's are deadlocked atop the standings (with Cleveland percentage points behind the other two), the Twins need to overcome a 2.5 game deficit, but hope that all three teams play below them the rest of the way. It's probably safe to say that a 26-12 finish would be enough to ward off all three other teams, but that's also most likely a pipe dream.

Specifically, the playoff odds for the AL WC contenders look like this:

Indians- 40.6% probability of winning the WC
Yankees- 21.6%
A's- 7.6%
Twins- 6.0%

The Division leaders also have a large chunk of WC probability as well, because each of them holds a large enough cushion of wins that they would have a good shot at the playoffs even if they lost the division. No other teams have more than a 1% chance.

The odds are calculated based on a simulation of the team's actual remaining schedule by using the team's third order winning percentage and running one million Monte Carlo Simulations of the rest of the season. In other words, they take the team's component stats, extract from that a model of performance, and match it up against the teams they have left to play. Oakland's percentage is lower because they have won more games than their level of performance would project. Similarly, Cleveland is still sleeping off the hangover from their nightmarish start where they played way under their Pythaganport projections. Thus, the records for both teams should continue to look more and more like the actual performance.

Since the Twins have been pretty much luck neutral this year (as opposed to the last three years in which they were consistently win lucky), they should naturally come close to catching the A's according to this metric. Not surprisingly, the Twins component strengths are their strong pitching buoyed by a low BB rate while they suffer from too little taking and raking (OBP and SLG) at the plate. Of course, these standings don't really tell us anything we don't already know: if the Twins keep hitting like they have all season, they will probably not make the playoffs. What keeps us tuning in to FSN North is the possibility that Lew has really found his swing, Morneau will continue slowly coming around, and Stewart keeps it up with the multi-hit games. If there really is some intangible quality that has helped them over the hump by turning it up to 11 in September the last three years, now is the time to start using it, because the tangible skills aren't getting it done.

As for the National League, the Astros have a decisive lead with a 39% probability, due in large part to the unbalanced schedule giving them the Pirates and Reds while the NL East (all between 6% and 15% probability) tear each other apart.

One interesting sidenote from the adjusted standings: The White Sox are still 3rd in the AL Central in 3rd order wins, meaning they have played to the level of the third best team in the division. They also project to finish 20-19 the rest of the way.