27 August 2007

Realignment and scheduling

In the previous installment, I made an argument for why realignment and expansion should be done by the 2015 season. In the end it all comes down to balance. That means balance for both leagues, for all divisions, and also for scheduling reasons as well. It would stand to reason that having even numbers of teams throughout the table makes a certain symmetry possible.

Well, I had no idea how bad the scheduling system was until I started to really take a look at it. Anyone looking at a standings table with a rudimentary knowledge of how MLB games are played could easily ascribe a certain level of messiness to the task of scheduling. But when you dig deep the ugliness really starts to show.

In the
CBA there are only a few pages devoted to scheduling. Much of the language is meant to dissuade teams from scheduling splot or day-night doubleheaders, which means the language has been largely successful. Regular doubleheaders are not discouraged, but every team has financial reasons to keep games separate. (Yes, I dearly miss the Sunday A's-Indians double dips too.) There's little of interest until you get to the section regarding interleague play:
Each Club may be scheduled to play up to 18 Interleague games during each championship season.
"Up to 18?" That didn't make much sense to me. So I looked at this season's MLB standings with interleague records. Sure enough, most teams played 18 interleague games including all AL teams. Some NL teams, however, played only 15. In previous seasons, some played as few as 12. What gives?

Look at the tables above. There's little rhyme or reason to it. There is a sort of "make good" that occurs in following years but it's just a band-aid to the overall lameness of the system. The unbalanced leagues created this situation. I'm all for interleague play, but there's something seriously wrong with it if some teams play a full interleague schedule - 18 games - while others don't. Then you start getting into which league is more dominant (AL recently) and how many games against specific competition are being played with uneven splits (home/road/intraleague/interleague). These factors contribute to strength of schedule, so every effort should be made to make scheduling as equitable as possible. Unfortunately, as long as the 16-14 scenario continues to exist, the schedules will continue to be unbalanced.

Compare the A's and Giants' scheduling above with the next graphic. It's much simpler and straightforward, and only possible with an 8-division MLB.

In this iteration, the A's would play each of the other three AL West teams 16 times, 8 at home and 8 on the road. 16 isn't a common number for games played between two teams. Normally you'll see a team play a division foe 18-19 times per season, whereas the old pre-expansion balanced schedule had two division rivals play 12-14 times. 16 strikes a good balance between the two, but scheduling it can be a bit tricky. A series is 2-4 games long, so the likely formats would be 3-2-3/3-2-3 or 4-4/4-4 (home/road). A prototype April 2008 schedule would look like this:
  • LAA @ OAK, 4 games (Apr 7-10)
  • OAK @ SEA, 3 games (Apr 11-13)
  • OAK @ POR, 3 games (Apr 14-16)
  • SEA @ OAK, 3 games (Apr 18-20)
  • POR @ OAK, 3 games (Apr 21-23)
  • OAK @ LAA, 4 games (Apr 24-27)
That takes care of the first stretch of division games. SEA and POR have the 3-game sets, so they would also have 2-game home/road sets with the A's just before and after the All Star Break. The A's wouldn't see the Angels until September, when sequence above would be repeated the final 3 weeks of the season. Why spread things out this way? 7-day weeks make it difficult to fit completely uniform scheduling such as 3-game series unless every week has an off day. That format wouldn't work because MLB requires that a regular season be wrapped up in at most 183 days. Running counter to that is a rule that dictates that teams can play at most 20 days in a row without an off day.

When it comes to playing the other AL divisions, it's pretty simple there too. One division's teams (in this case the AL North) would play the A's 12 times, while the other two divisions' teams would face the A's 6 times apiece. There's an alternative below in which the A's would face two divisions' teams 9 times, but it's really all a matter of choice. The nice thing about formatting the schedule this way is that it's easy to segment the play. In-division play is generally reserved for April and September. The other divisions would generally get 2-week stretches. Interleague play would be kept around same timeframe as it currently is: 3 weeks from late May through the middle of June.

If you're interested in further studying the scheduling model, you can
download an Excel sample. It has one odd patch in that there's a Friday-Monday 4-game series in September, but everything else works reasonably well. And that's a draft I put together in only a few hours (my adding it delayed this post), so it could certainly use some refinement.

Does scheduling really make that much of a difference? It's hard to say. The nice thing about this type of schedule is its modularity. The components can be moved around so that they can fit in different ways. Are the Twins and Indians continually having trouble in their outdoor ballparks with a late winter in April? Move the in-division block back a week and put them on the road in the Sun Belt first. Need to guarantee that the BoSox and Yanks meet on Patriots' Day? No problem - stick with the normal configuration. Want to reduce the number of interleague games from 18 to 12? Sure thing, and now that makes 6 extra games available for your division rivals.

There's a better way of doing things on the horizon, and it comes with expansion and realignment. Some critics have pointed to dilution of talent, and I say "quantify it." Just as there are crappy 5th starters now, there were crappy 4th starters 30 years ago. Is the current Washington Nationals' roster appreciably worse than the '71 Washington Senators? Of all the controversies that have marked baseball, which is most responsible for the supposed decline in play: out-of-scale economics, juiced balls, bandboxes, performance enhancing drugs, the rise of basketball and football as major competitors for athletes, etc...? I would argue any of those issues is more responsible that a dilution of talent. Major league baseball belongs in as many places as can reasonably support it. Let's see it happen.


Anonymous said...

Please " splain " what this topic has to do with a blog about the Pacific Commons development/Cisco Field/ballpark village ?

Marine Layer said...

Zero. Next question.

Anonymous said...

FYI everyone, next Wednesday the Commonwealth Club is having a panel discussion on the future of the 49ers in San Francisco.

The Cactus Leaguer said...

You are right, the scheduling/realignment piece is incredibly complicated. I tried piecing it together in the past at and it is a bear. Even the very reasonable schedule you put together is problematic when you try to put it on a calendar... I don't think the number four game (or two game if you wanted to break them up) sets against intradivision opponents is workable.

I have been trying to fiddle with it again with mixed results... if I have a breakthrough, I'll post a link (but don't hold your breath for it). In addition to what you have, I am trying to add a couple of items into the mix:

1. 54 games against your intradivision opponents;
2. A certain number of interleague games against the same division (AL West v NL West, etc.) each year (without messing with the integrity of the "rivals" series);
3. Some interleague games against a different division, spaced out in a way that assures that every team will play in every ballpark at least once every 8 years (4 would be nice but too onerous).

Jeffrey said...

Actually I would argue realignment/expansion plays a role, albeit peripherally, in the future of the A's and thus Cisco Field.

In particular, if Cisco Field falls through, one of these expansion cities will be taken. If ti is Portland, that means there would have to be a new West Coast city to add to the pile as there is no way another American league team would be put back in the Bay Area.

Or maybe that is a stretch, but I sure like reading about it.

Anonymous said...

ML, great work. I'm glad you're looking at scheduling. Due to the schedule seeming severly changed and messed up over the last 3 years, I wasn't suprised to find out that a different company took over MLB scheduling about 3 years ago. This year was an all time worst, many two games series, including a trip by the A's to Boston during the last week bookended by days off. Also, how do the New York Mets qualify as an NL Central team? Playing a first place team on their field instead of the Cubs or Pirates is a crock. Frankly, as someone who likes to get to as many interleague games as possible, I feel personally robbed of not getting to see the Cubs in Oakland.

Thanks again for all the great work.

Marine Layer said...

TCL - why wouldn't the 4-4/4-4 situation work? Sure it's quirky but at least it's predictable and consistent. It'd be pretty easy to rotate any team's 4-4/4-4 opponent on an annual basis.

Does anyone remember life before the unbalanced schedules? I have a media guide from the '96 season. Care to guess the date of the first M's-A's game?

If you guessed July 1st, give yourself a gold star for you are correct.

MLB has had over a decade interleague play right. They'll always be hampered by the built-in inequity between the leagues. And no, last time I checked the Mets weren't in the NL Central.

BigMac1212 said...

Nice idea. I would of gone with San Antonio, but that's just me.

I would like to expand this to the minor leagues. Can you help?