What are the characteristics of that form to which I pay particular attention when building a schedule?
Earlier (ancient?) posts started toward describing this.
https://makingbaseballschedules.blogspot.com/2009/09/schedule-parameters-introduction.html
https://makingbaseballschedules.blogspot.com/2009/10/modern-mlb-schedule.html
There will be things from there rehashed here, particuarly from that MLB post. I am aiming to highlight and emphasize some of these things and others I consider in the process.
The biggest factor is always how to lay out who plays who and when. I will not really go into that here as I expect it to be a focus of many other posts. This is more about the details around that main structure.
Length
This means more than simply number of games - the amount of time in which to fit those games is very important for design. Recently MLB extended the season length by a half-week such that it spans 52 half-weeks instead of 51. That is not counting the All-Star break which adds one to these numbers. So that means a few more offdays per season per teams.
It had been 162 games to 182 days, with 3-4 day All-Star break meaning 16-17 other offdays across 25 and a half weeks, which is very roughly 2 offdays every 3 weeks. Now there are 186 days meaning very roughly 4 offdays every 5 weeks. That may not seem like much difference, but it is a significant bump up at least closer to one offday per week.
For proportionally shorter schedules I will tend to try to keep the frequency of offdays the same. So for 108 games that would mean 2/3rds of everything for 162. Or rather, about 2/3rds of everything - typically there will need to be some sort of rounding and compromise to get close and fit series into the usual half-week scheme.
Offday distribution
I like to spread out the offdays such that there is a likely to be a few games variance among teams in how many games they have played. E.g., I like seeing the standings when my team has played 100 games and other teams have played a couple more or a couple less. It is almost always simpler not to spread out things like that, so no problem if someone wants perfect consistency in offdays among teams. But I like art in the variety and having it match what you see in MLB (and other leagues like the NBA and NHL).
The MLB requirement of no more than 20 straight days without an offday is made easier to satisfy with the increased number of offdays. I generally use that requirement as a rule in my construction unless something is specified that precludes it.
The primary way to establish the offday distribution is through the distribution of the series of different links. For 162 games in 52 half-weeks, that could be 46 series of 3 games and 6 series of 4 games. Some teams could be set to have more of their 4-game series earlier (and thus have a relatively higher number of games played at any given point), some could be set to have more of their 4-game series later (and thus have a relatively lower number of games played at any given point).
One method I use in deciding the fine-grained distribution of offdays is by trying to minimize the offdays within homestands, which means offdays are more likely to fall on travel days. E.g., if a team has a homestand of two 3-game series in a particular week, the weekend series is Fri-Sun, and the early-week series could be either Mon-Wed or Tue-Thu. So should the offday be Mon or Thu? Our example home team's early-week opponent could have a 4-game series against its next opponent on that weekend, meaning it is playing Thu which would force the early week series to be Mon-Wed. Otherwise I would set that series to Tue-Thu and the offday to Mon on the logic that it would be "nicer" for travel purposes for that home team to have the offday that week between the series the previous weekend it played on the road and the start of the homestand rather than within the homestand.
I was somewhat surprised to see that MLB does not seem to have any notable preference against offdays within homestands. They do make a point of often have "getaway-day" games being during the day rather than at night. More than anything, I guess it means my preference to put the offdays between places is not universal. Maybe offdays within homestands are actually quite nice and worth making travel a little tighter.
For OOTP, I do not believe travel is a real consideration - i.e., players are not differently affected depending on where their games the previous day were. So it is all cosmetic anyway and allows me to add that spice of variety where I can.
Series lengths
As of 2019, about 2/3rds of MLB series are 3 games in length. By the same token about 2/3rds of games are in 3-game series. That is a clear majority but definitely not everything. The rest of the series are split about 2-to-1 between 4-game series and 2-game series. Because 4-game series are twice as long, there are about 4 times as many games in 4-game series as in 2-game.
Unless told otherwise I will stick with only 2-, 3-, and 4-game series. And I try to avoid 2-game series. They are small and mean more moving parts to try to piece together. But like as MLB uses them, they are good for trying to get finer balance in games between opponents and with home/away splits. For a team there can be two 2-game series Mon-Tue and Wed-Thu plus a Fri-Sun weekend series. Such a 3-series week with two series in one of the half-weeks can frequently be handy to make a layout fit within the appropriate length.
Bottom line - given no other guidance, and if it works out well enough, I aim for mostly 3-game series and some 4-game series. That fits very well with my tools geared toward expecting two series per week.
All-Star break
I understand the desire to put the All-Star Game as precisely as possibly at midseason to bisect into a true first half and second half. But I have preferred the style of MLB putting it something like 55-60% of the way through the 162-game season. That translates to about 90 games into 162.
In my mind that gives more games to determine who "deserves" to be the All-Star Game from first half performance. For shorter seasons I will tend toward an even higher proportion of games being played before the break. E.g., for an 82-game season, depending on how things fall, I would be inclined to put the All-Star break 2/3rds or even perhaps a bit more into the season, like after about 55 of 82 games. Different sports, I know, but the NHL and NBA both currently have 82-game schedules. This season the NHL had its ASG at roughly the 60% completed point, and the NBA at about the 70% completed point. That is relevant insofar as it sort of primes me to expect an ASG to be past the midpoint in terms of games played.
That preference is purely cosmetic in the end. It can be a hassle to move an All-Star break in an already-built schedule, but if it is specified in the beginning then setting it anywhere (*gasp*, like even before the season midpoint) is usually no problem at all.
If the All-Star break is a half-week of 3 or 4 days and the other layout parameters allow it, I will put the break in an early half-week with the All-Star Game itself on Tuesday. I think most people are very comfortable with the MLB convention. Again the game can be set for whenever, but a main real-life consideration of the early-week All-Star break is that it means not making teams sacrifice weekend games.
Weekend series
As a general rule attendance is better on weekends, which I believe is reflected in OOTP as well. This is easy to confirm (or not), but I want to say I have seen way back in the day and would assume it would still be true. In adding a half-week to its season, MLB did not tack it on at the end to run the season across an extra Mon-Wed, it added at the beginning such that the season starts across another weekend.
Something the MLB schedules are good at doing is balancing the number of home series across all teams. With the current 52 half-weeks, there are 25 that are early-week and 27 on weekends (the 2 extra weekends are because of starting on a weekend and that the All-Star break falls early-week). Every team has an as-even-as-possible split with either 13 or 14 of those weekends being played at home.
Such balance is something I am making a point to achieve in my re-involvement in schedule making. I may not be quite as strict (e.g., maybe between 12 and 15 is good enough), but generally I want to get as much balance as reasonable. It can take something between a little extra care to a very special layout to make sure things work out fairly evenly. And if it is too difficult to make work evenly, I will abandon it.
Homestands and roadtrips
This may be the holy grail. MLB schedules are great at making the vast majority of homestands and roadtrips be either 2-series or 3-series in length. The 2019 schedule has no homestands nor roadtrips more than 11 games in length, and there are few cases of only a single series home or away between series of the other type. This is again cosmetic in OOTP, but it feels much more realistic not to have super-long homestands/roadtrips and not to frequently play only one series home or away at a time.
It is really hard, and I expect to talk more about it over time as I work at it more. I do usually (though not always) set a hard limit of 4 series (or 14 games) consecutively home or away, and I try to minimize the single-series cases while satisfying other criteria. But there still is likely to be a quarter or even more of series being single-series homestands or roadtrips. For 52 series, 25% is 13 instances of a single series home or away. I want to do better, but again, it is really hard, and the nature of the matchups (i.e., few different opponents with many games against each, or many different opponents with few games against each) affect the layout which affects how homestands and roadtrips can work out.
Doubleheaders
MLB has abandoned regularly scheduling them, and I do not use them unless explicitly told to do so. The biggest factor for me is that my building tools are not at all well-suited toward handling a team playing multiple games on a day. Some systematic things, like everybody always plays a doubleheader on Sundays, are not that bad for me to deal with.
Gametimes
I generally do 7:05pm for everything. That is just easy. I am trying to get my tools better able to provide variety if it would be wanted. An easy thing would be more Sunday afternoon games. My first significant goal is getting toward making more of the afternoon "getaway day" games - if both (or either?) of the teams are playing somewhere else the next day, have them play earlier in the afternoon rather than the evening. Not really meaningful in the game, but something that may be nice to make it feel more realistic.
One issue with going specific, particularly team-specific, with game times is how the game will randomize the schedule year-to-year. You may have a "Cubs" team set to play a lot of day games, but the next year the "Reds" team could take over that slot with a lot of day games, and the "Cubs" could move into what was the "Pirates" slot which had few day games.
No comments:
Post a Comment