An elementary baseball question

[Jens]

I know very little about the sport, sorry to say, but I was wondering something about the World Series (not really sport related, more about business). They don't know how many games are going to be in the series, but I assume they have to reserve the stadium, security, logistics, sell tickets, etc., etc., for the whole series. Isn't it an enormous loss if the series finishes quickly? I'd think it would encourage the owners to put pressure on the players to keep the series going as long as possible.

[hhEb09'1]

I think those concerns have led to agreements where the players are paid only for four games--but of course they're paid well. It's not just baseball, I'm sure, basketball and hockey probably have similar rules.

But what would the owners do? Try to convince their own players to lose a World Series game? Seems like short-sighted economic policy.

Unless they collude with other owners.

[Nick Theodorakis]

There are also benefits (financial or otherwise) to winning the World Series, so it would not be in a player's best interest to try to stretch out the series and risk losing it.

Using the data listed here, I counted the distribution of the number of games needed to win the series. Over the last 25 years, the distribution is:

4 games: 7 series
5 games: 5
6 games: 6
7 games: 7

If there was a pressure to stretch it out, I would have suspected the distribution to skew toward the high end.

If you go back farther, the number of series going to 7 games does seem to be little overrepresented, but I don't know of that is statistically significant:

last 50 years:

4 games: 10 series
5 games: 10
6 games: 11
7 games: 19

last 75 years:

4 games: 14 series
5 games: 15
6 games: 17
7 games: 29

Nick

[Tog]

No idea on baseball, but this is slightly related. We have a minor league hockey team. A couple of years ago, our hotel got to house the visiting team for the second round of the the playoffs; games 3, 4, and 5 of a best of 7.

The reason we got to house them was that the normal hotel, the one that sponsors the team, gave away all the rooms that they should have reserved. Presumably because they didn't expect the team to make it out of the first round. The following year, there was a new hotel housing the visiting teams, with a radio add that also a slight slam at the old one that played off of the name.

I imagine that it's in the best interest of the hotels, security, and venue to plan on the game going 7 and then taking a loss on those nights if the series ends before 7.

[Chuck]

I suspect that too many people would have to know about it for it to be kept a secret year after year.

[jfribrg]

Remember too that the winners get a substantial prize fund, so prolonging the series runs the risk that the other team will get it. The owners and TV networks may have an incentive to prolong it, but not the players. Remember also, that the next year, the World Series champions tend to draw more spectators both home and away, which is an incentive to win the series as quick as possible. My feeling is that the managers have a bigger incentive to throw a game ( there are plenty of gamblers who would gladly pay millions for a guarenteed result), than do the players and owners. A manager can influence the game by leaving a struggling pitcher in longer or by replacing an effective pitcher. The players don't have to be in on it.

[Fazor]

A manager can influence the game by leaving a struggling pitcher in longer or by replacing an effective pitcher. The players don't have to be in on it.

True. But a manager that loses World Series' is a manager that loses his job. I'm not saying you're automatically out for losing a World Series; but job security isn't earned by losing.

Besides, managers are often as competitive as the players. They want to win too.

[DonM435]

If they were forced to go ahead and play out the seven games even after one team had clinched the series victory, you'd really see some anemic attendance figures for those meaningless games.

[Chuck]

They could make it most total runs wins the series instead of most games won, but the final game or two might still be essentially meaningless if one team is way ahead.

[Metricyard]

I would think, as far as the players are concerned, is to win as quickly as possible. They are payed on yearly, or multi year contracts. It doesn't matter to them if they play four or seven games, they still receive the same amount of money. Of course winning is always better than losing.

Not sure how the stadiums handle it. I would assume they would love to have more games played, because, well, they would make more money.

[aurora]

Typically, teams sell playoff tickets to their season ticket holders (first, then the general public) and in baseball they often require people to buy the entire set -- tickets for every possible playoff game -- for every series (division, league, world series) and assuming that each series goes the maximum possible.

They hold the money, and weeks later when and if the game is not played, then they get around to refunding the money.

At least, that is how it worked one year when I got playoff tickets.

Edited to add: for season ticket holders, this is done by every team that is still in contention with a week or two left in the regular season.

[SeanF]

If you go back farther, the number of series going to 7 games does seem to be little overrepresented, but I don't know of that is statistically significant:

Define "overrepresented"? It seems to me that if baseball is at all successful in "balancing" the talent, the World Series should skew towards the full seven games. A four-game series would suggest one team was pretty heavily overmatched.

They could make it most total runs wins the series instead of most games won, but the final game or two might still be essentially meaningless if one team is way ahead.

Now you sound like the people who want to get rid of the Electoral College.

[SeanF]

Edited to add: for season ticket holders, this is done by every team that is still in contention with a week or two left in the regular season.

Yes, and for season ticket holders, there's a pretty strong push to have the extra money just deducted off of next year's tickets...

[DonM435]

If two teams are evenly matched, you'd expect a 7th game 5/16 th of the time, or 31.25%. (Based upon [nCr formula with n=6 and r=3] multiplied by [0.5 to the 6th power].)

[mugaliens]

Because winning means far more income than the cost of renting the stadium for days not played, the owners would never encourage their teams to loose just to drag out the series beyond the minimum possible four games.

Now the promoters, on the other hand...

There is an incentive to drag it out, though, as both clubs get a portion of the gate receipts. However, the winning ball club gets a larger proportion, so depending on how much larger a proportion they get, that could be a disincentive.

[Jens]

If they were forced to go ahead and play out the seven games even after one team had clinched the series victory, you'd really see some anemic attendance figures for those meaningless games.

I guess that's probably true. There are sports where that happens, though. In Sumo wrestling, the one sport that I do watch pretty faithfully, there are 15-day tournaments, and it's not unusual for somebody to win on day 14 or even day 15. But people still watch the last day. I think it would be different with baseball though. (The reason people in Sumo will still go is because the two top wrestlers fight on the last day, regardless of who has won. So even if the tournament is decided, you still get to see a good fight.)

Even in baseball, during the regular seasons there are games after the championship is decided. I suppose the attendance isn't very good.

[Jens]

And thanks everybody for clearing that up. My motivation was really that it seemed like a logistical nightmare to me, getting all the people ready for work and then deciding at the last minute to cancel the next day's game. Though I suppose that's how things work in that kind of world. If a hurricane shows up the same thing would happen.

[DonM435]

I understand your concern. Many teams go through all of that several times per season anyway, when there are weather postponements and hasty reschedulings. If there's a tie for first place at the end of the season proper in any of the major league baseball divisions, they may have to schedule a playoff game for the very next day.

The NFL Super Bowl has always been a "Best of One" proposition, and so doesn't have to worry about "if necessary" games.

[pghnative]

Iven in baseball, during the regular seasons there are games after the championship is decided

Actually, no.

Perhaps you meant that during the regular seasons there are games played after the playoff races have been decided. (e.g., my hometown team (see avatar) is usually eliminated from the playoffs by a week or two after mid-season).

[hhEb09'1]

If two teams are evenly matched, you'd expect a 7th game 5/16 th of the time, or 31.25%. (Based upon [nCr formula with n=6 and r=3] multiplied by [0.5 to the 6th power].)

Does everybody else get 7 games (5/16), 6 games (5/16), 5 games (4/16), and 4 games (2/16)? I did it in a similar way, but the equality of 6 & 7 makes sense--given a sixth game, half the time you'd get a seventh, and half the time not. Each team has (1/16) chance of going four and won, so 4 games being (2/16) makes sense too.

[SeanF]

Does everybody else get 7 games (5/16), 6 games (5/16), 5 games (4/16), and 4 games (2/16)? I did it in a similar way, but the equality of 6 & 7 makes sense--given a sixth game, half the time you'd get a seventh, and half the time not. Each team has (1/16) chance of going four and won, so 4 games being (2/16) makes sense too.

And if the 2/16, 5/16, and 5/16 all make sense, then the remaining option must be 4/16 also.

[Donnie B.]

There was one world series that many suspect was extended deliberately. It was in 1912, between the Boston Red Sox and New York Giants.

It was an unusual series in several ways. For one, it went eight games, because one game ended in a tie. And that's what led to the situation that some find suspicious.

Game 2 was held in Boston, and was tied in the 11th inning when it was called due to darkness. Since the game wasn't played to a conclusion, the baseball commission refused to pay the players their usual share of the gate receipts.

Later in the series, Boston had a lead (3 games to 2 with one tie) going into game 7. Their ace pitcher, Smoky Joe Wood, was shelled and the Giants won the game. That led to a Game 8 in Boston, which the Red Sox won.

The suspicion is that Boston threw game 7 in order to make up the lost gate they were denied for game 2. Some say the loss in game 7 had a different motive -- that Wood was personally miffed at management (I don't remember the exact reason) and tanked the game.

I've heard a few contemporary opinions that playoffs in various sports are stage-managed to maximize revenue. In this cynical age, I'm surprised we don't hear that accusation more often.

There's a theory that this very college football season, the officials in the Southeast Conference have made suspicious calls to favor Florida and Alabama, with the intent to manipulate the outcome of their games so that one of the two teams makes it to the national championship. There certainly have been a lot of highly questionable calls that benefited those two teams.

[Jens]

Perhaps you meant that during the regular seasons there are games played after the playoff races have been decided. (e.g., my hometown team (see avatar) is usually eliminated from the playoffs by a week or two after mid-season).

Yes, that's what I meant. For the playoff races. My terminology isn't very good, because I don't watch the game much, and when I do it's usually in Japanese.

[Chuck]

If each team had a 50% chance of winning each game, the average best of 7 playoff series would last 5.8125 games. If one team were better than the other the average would be lower. If the average were consistently higher then some explanation would be nice.

[Jens]

The home team advantage may play some role. Suppose that the teams were evenly matched and being the home team provided a decisive advantage. I think that in a best of seven games series, it's usually done 2-3-2? So that would mean that team A would win in game 7. If being the home team just gives a small advantage, though, I don't know what it would do to the statistics.

[mugaliens]

If each team had a 50% chance of winning each game, the average best of 7 playoff series would last 5.8125 games. If one team were better than the other the average would be lower. If the average were consistently higher then some explanation would be nice.

Since one team is usually better than the other, the average number of games played in the world series would be higher.

Going back to 1950, we have the following means:

Win: 4.000 (duh)

Loose: 1.8644

Games: 5.8644

Spread: 2.1356

It appears your 5.8125 figure is the mean number of games actually played since it was called the World Series, not a hypothetical statistic.

Instead of doing the math, I ran a monte carlo approach over 1,000 series, assuming two equal teams, and it produced a mean of 5.8005, but various runs produced ranges between 5.7350 and 5.8390.

The result is not statistically different (95% CI, 3% error) than either the actual number of games since 1950 or your 5.8125 statistic, we can conclude that the teams are statistically equivalent in terms of their likelihood of winning. This only stands to reason given the very large population from which players are chosen, the large number of teams which play, the highly stylized rules of play, the highly objective (usually) umpiring, and the often variable nature of play.

I'd still bet on the Yankees, particularly when, as Jens notes, they have a home team advantage...

[SeanF]

The home team advantage may play some role. Suppose that the teams were evenly matched and being the home team provided a decisive advantage. I think that in a best of seven games series, it's usually done 2-3-2? So that would mean that team A would win in game 7. If being the home team just gives a small advantage, though, I don't know what it would do to the statistics.

I remember reading somewhere that, statistically speaking, baseball has less of a home-field advantage than do the other major sports.

However - the last eighteen times that a World Series has gone seven games, the home team has won 11 of them (including all eight since 1980).

[aurora]

World series is a small sample size. Home field advantage is best computed based on thousands of games.

[pghnative]

Game 2 was held in Boston, and was tied in the 11th inning when it was called due to darkness. Since the game wasn't played to a conclusion, the baseball commission refused to pay the players their usual share of the gate receipts.

According to wiki the players could only receive a share of the gate from the first four games, so this argument doesn't seem valid.

(unclear on whether 1912 players received only three games worth (Games 1, 3, 4) or four (Games 1, 3, 4, 5), but seems the players had no stake in money from game 8.

[Chuck]

Since one team is usually better than the other, the average number of games played in the world series would be higher.

I don't think having one team being better than the other would mean more games played. In an extreme case with one overwhelming team, the average would be near 4. If there were a big home field advantage and the home team nearly always won then the average would be close to 7.