clock menu more-arrow no yes

Filed under:

Fun With Probability & How to Build a Roster

New, 11 comments

Something we always used to hear back in the day of good Mariner teams was that this was a team "built to win in the regular season," that it didn't have what it takes to succeed in the playoffs. And it was difficult to argue otherwise - after all, between 2000-2003, the M's finished 393-255 during the season but went 9-10 in the playoffs, bowing out twice to New York in the ALCS. The Yankees were the right kind of team, made up of 5% carbon and 95% mystique; they knew how to win in the national spotlight, whereas those timid Mariners fell apart when it counted. And that was that.

Seattle's not alone, though. Last night, we saw another team "built for the regular season" drop out of the playoffs, as the Cardinals - winners of 205 games in the last two years - could never really get off the ground against Houston's strong pitching. Despite Albert Pujols' otherworldly talent and David Eckstein's divine constitution, St. Louis is perilously close to being labeled as a team that can't succeed under pressure.

Now, there are two popular ways of looking at this:

Bastardized Stathead: The playoffs are a crapshoot. Even if you get there, the odds are always against your winning the Series.
Joe Morgan: Good teams that lose early in October are chokers.

Unfortunately, neither is really correct. While it'd certainly be nice if someone could explain the concept of "statistical probability" to Mr. Morgan, the playoffs aren't really all about luck, either - it is possible for a team to improve its odds of winning it all by building its roster for the specific purpose of succeeding in October from Day 1. And it doesn't require that you sacrifice any regular season victories, either.

Consider two teams, Team A and Team B.

Team A:

  1. Nine position players worth 75 runs each
  2. A bench worth 75 runs
  3. Five starters who allow 95 runs in 200 innings
  4. A bullpen that allows 190 runs
Team B:
  1. Nine position players worth 75 runs each
  2. A bench worth 75 runs
  3. Three starters who allow 75 runs in 200 innings
  4. Two starters who allow 125 runs in 200 innings
  5. A bullpen that allows 190 runs
Each team's Pythagorean record in the regular season would be 90-72 (750 runs score, 665 runs against). For the sake of simplicity, let's say that this is good enough to get them to the playoffs, and that it's also the average record of the eight remaining teams.

Here's where the fun happens. Having made it to the postseason, each team shuffles its rotation by dropping the fifth starter and divying up his innings among the top three (as is usually the case). Team A keeps the same Pythagorean winning percentage (.555), since its fifth starter wasn't any worse than the other four, but Team B sees its expected winning percentage jump from .555 to .590, since it's suddenly giving more innings to better pitchers.

Operating under the assumption that everything else (lineup, bench, bullpen) stays the same, the following are each team's respective odds of winning each round:

Division Series:

  1. Team A: 50%
  2. Team B: 56.6%
League Championship Series:
  1. Team A: 25%
  2. Team B: 32.6%
World Series:
  1. Team A: 12.5%
  2. Team B: 18.9%
(Team B's odds of winning each game calculated with a little help from this article by Tom Tippett).

This isn't really any sort of scientific breakthrough; generally, you'd prefer to go into the playoffs with a topheavy rotation instead of a deep one, because the depth doesn't do you any good in a short series. Curt Schilling and Pedro Martinez played a big part in getting Boston its title a year ago, just as Schilling and Randy Johnson had done in Arizona in 2001. On the other side of the coin, deep rotations like Seattle's in 2000/2001 or St. Louis' in either of these past two years haven't really done their teams much good, as they lacked the shutdown stoppers to match zeroes with the other teams' aces.

It should be noted that, while a team stands its best chance of winning the Series by frontloading its rotation, it's risky to try the same maneuver with the lineup or the bullpen - you can't hide bad hitters or relievers as easily as you can a bad fifth starter. Given two lineups that score 800 runs, one with nine above-average hitters and one with one or two total sluggers and a bunch of flotsam, you want to go with the former, as the latter is vulnerable if the slugger(s) hit a slump at the wrong time (see Guerrero, Vladimir with Anaheim about a week ago). You get a little more leniency with the bullpen, since stud relievers tend to eat up a ton of innings in the playoffs, but you're still going to have your middle arms and setup men throwing important pitches at some point during the series, so it's not really wise to throw caution to the wind and throw some randomly-generated AAA LOOGY into the fire.

It's fitting, then, that Houston has made it this far, given that they're the perfect example of a pitching staff that's built to win in the playoffs - three top starters and an untouchable closer, with a bullpen underbelly that won't embarrass itself if it needs to soak up a few innings in between. Chicago's also a good example, although a lot of their apparent strength on the mound is really strength with the gloves. Still, the White Sox' starting rotation boasts three of the top 11 ERA's in the AL, and that says something.

So, where do the Mariners stand? They're not really in that bad of a position, if you take this approach. Their bullpen next year is going to be in the upper third of the league, they have Felix at the front of the rotation, and the lineup is a few probable improvements (Beltre and Reed being the two most notable) away from being a consistent run-scoring threat without many black holes. Throw in a Burnett, Millwood, or Matsuzaka, and you've got a team with all the right ingredients to make a lot of noise in October. It's not necessarily a deep team, so just making it to the playoffs will be a considerable challenge, but should they add a talented starter and stay relatively healthy over the course of the season, Seattle's going to have good reason to be optimistic.