Strength of Schedule: Do the Standings Lie?

As we approach the quarter mark of the season, many people are starting to
assemble "surprises and disappointments" lists. Some of these
lists reflect a lack of careful analysis–such as any comments about
Dante Bichette‘s inept performance–but another widely-ignored
problem is whether the quarter of the season just played will be
representative of the season as a whole. In many cases there’s reason to
think that it’s not.

As an example, let’s look at a couple of teams in the NL East. In its
pre-season predictions, the BP staff regarded both the Phillies and the
Expos as middle-of-the-pack teams. Yet after Monday’s games, the Phillies
had the worst record in the National League while Les Expos were only a
game out of the wild-card spot. It would be easy to come to the conclusion
that we had goofed badly on these teams.

But a closer look at the schedule reveals some interesting facts. Through
Monday, Philadelphia had played 36 games, with 21 of them against the
Braves, Cardinals, Reds and Diamondbacks, the four teams who would make the
playoffs if the season had ended May 15. They had yet to play any of the
Pirates, Cubs, Brewers, Rockies or Padres, teams which had combined for a
record 31 games under .500. By contrast, Montreal has played those five
teams a remarkable 23 times in 35 games, while not playing a single game
against the four playoff-position teams.

Now, there is a chicken-and-egg question here. Do the four top teams have a
great record because they’ve been beating up on bad teams like the
Phillies, or are the Phillies scuffling because they’ve had the bad luck to
play the top teams a lot? Clay Davenport took a look at the strength of
schedule for all 30 teams and used won-loss records from both 1999 and 2000
to try and escape this problem. The methodology is fairly simple: the
strength of each team is determined by dividing their wins by their losses,
and then the difficulty of a team’s schedule is determined from those
numbers. The results are given here:

The numbers were run twice, with the 1999 column using the final standings
from last season and the 2000 column using the early results from this one.
The larger the number in the table, the tougher the team’s schedule.

             1999   2000                    1999   2000
Anaheim      1055   1041     Arizona         890    767
Baltimore    1054   1132     Atlanta         866    830
Boston        964    902     Chicago        1161   1184
Chicago       948    929     Cincinnati      982    875
Cleveland    1049   1101     Colorado       1170   1355
Detroit       980   1178     Florida        1083    978
Kansas City  1000   1033     Houston         892    845
Minnesota    1022   1061     Los Angeles    1312   1378
New York      982    906     Milwaukee      1110   1232
Oakland      1043   1085     Montreal        865    774
Seattle      1054   1133     New York        894    902
Tampa Bay    1019   1088     Philadelphia   1313   1573
Texas        1063   1168     Pittsburgh     1154   1167
Toronto      1060   1101     St Louis        968    838
                             San Diego      1306   1480
                             San Francisco  1108   1402

The numbers confirm the eyeball analysis. Using both 1999 and 2000 numbers
the Phillies came out with the toughest schedule in the majors while the
Expos came out with either the easiest or second-easiest schedule.

While it would be silly to suggest, based on these numbers, that the
Phillies could claw their way back into playoff contention or that the
Expos will wind up at the bottom of the heap, they do indicate that the
results so far should be taken with a grain of salt, and that our
pre-season predictions of mediocrity may very well wind up looking pretty
accurate.

That grain of salt should also be taken when looking at individual players
on those teams as well. Vladimir Guerrero has been a popular pick
for NL MVP so far, but what will happen when his free swinging ways (six
unintentional walks in 150 plate appearances) run into Greg Maddux,
Tom Glavine, Randy Johnson and the other elite pitchers he
has yet to face? Scott Rolen‘s .278 batting average and .360 on-base
percentage might seem somewhat disappointing compared to his 1997 and 1998
numbers, but he has yet to get a chance to fatten up the numbers on the
detritus that constitutes the pitching staffs of the Cubs and Brewers.

NL East

The Braves have had it easy so far, but none of their close competitors
have had it that much worse. So while they probably won’t keep up the
110-win pace, they can be pretty confident that the schedule so far is a
fair sample. Bad news for Mets fans: there’s reason to think things may get
worse given that their schedule has been fairly easy so far.

NL Central

The top teams have had a somewhat easy time so far, but again there’s not
much sign of competition. The Pirates have had a harder road, so they could
sneak up as the season moves along.

NL West

Out West, Diamondbacks fans shouldn’t start budgeting for those playoff
tickets yet. While they do have a substantial lead, they’ve done it with
the help of a schedule that’s among the three easiest in the majors.
Lurking behind them are the Dodgers, Giants and Rockies, all of whom have
managed to assemble a record of .500 or better despite very difficult
schedules. With that many teams lurking in the weeds, the odds are good
that someone will make a charge once their schedule eases up.

AL East

Those who argue that the Blue Jays will yet make their presence known in
the race will find some evidence to support them here. They’re only
four-and-a-half games out despite one of the toughest schedules in the
league, while the Red Sox and Yankees have played two of the easiest.

AL Central

The White Sox have had a substantially easier time of it than the two teams
breathing right down their necks, which is bad news for the South Siders.
At least for this year, it’s still reasonable to expect the Indians to
eventually take over the division.

AL West

Everyone seems to be clustered in the middle of the pack, which is only
fitting for a division where 2 1/2 games separate the top from the bottom
and no one is that far from .500.

Looking at the results across the majors, perhaps the most startling fact
is that there is a much smaller spread in the AL numbers than there is in
the NL numbers. There may be a simple logistical reason for that though.
Since the AL has had 14 teams since 1977, they’ve had plenty of time to
develop a reasonably balanced schedule that avoids some of the quirks of
the NL schedule, such as teams finishing with one team for the year before
ever playing other teams.

And just think, if Bud Selig gets his way and realignment goes through,
there could be even more confusion at this time next year.