Fangraphs: AL East looks like toughest division again

[Frobby]

[T]he AL East and the AL West, here, appear the strongest. Then there’s a gap, followed by all three NL divisions, which are practically tied. The separation between the best and the worst NL divisions is 0.4 wins. Then, bringing up the rear, there’s the AL Central.

* * *

As for the AL East — again, I have 13 years of preseason projections, and this is the twelfth time the AL East has projected as the toughest group. Back in 2006, it projected as the second-toughest. The current streak is up to 11 consecutive seasons, and the division has worked out as actually the strongest six times. On average, over the whole window, the AL East has projected as easily the best division, and it’s also been easily the best division. At some point, the group will get weaker, just because it has to, but it’s going to take a team or two giving up. It’s going to take a couple rebuilds, because right now, you could argue that any of the five teams could feasibly make the postseason. And for a while the projections have liked the Orioles too *little.*

* * *

All in all, maybe this is best for the Indians. If not them, it’s best for the Tigers, whose playoff shot is far more uncertain. In turn, that would make this most unfortunate for AL wild-card contenders in the other two groups. When a team like the White Sox elects to tear down, they’re not the only team affected. And you should remember they’re far from finished tearing.

http://www.fangraphs.com/blogs/the-al-east-looks-like-the-toughest-division-again/

[25 Nuggets]

 

 

As for the AL East — again, I have 13 years of preseason projections, and this is the twelfth time the AL East has projected as the toughest group. Back in 2006, it projected as the second-toughest. The current streak is up to 11 consecutive seasons, and the division has worked out as actually the strongest six times. On average, over the whole window, the AL East has projected as easily the best division, and it’s also been easily the best division.

 

All the more reason to appreciate the run the Orioles have had recently.

 

[Spy Fox]

I'm actually surprised the AL East has "only" been the winningest division six times in 12 years. If the schedule weren't so weighted towards the division games, causing nearly half of each division's games to be played to an even .500, I bet the AL East would have come out on top even more often. 

[Frobby]

 

I'm actually surprised the AL East has "only" been the winningest division six times in 12 years. If the schedule weren't so weighted towards the division games, causing nearly half of each division's games to be played to an even .500, I bet the AL East would have come out on top even more often. 

That doesn't make much logical sense to me.    Regardless of how many divisional games are played, the division that wins the most out-of-division games will have the best overall record (at least, when there are an equal number of teams in each division).

[Spy Fox]

 

That doesn't make much logical sense to me.    Regardless of how many divisional games are played, the division that wins the most out-of-division games will have the best overall record (at least, when there are an equal number of teams in each division).

It's true that whichever division wins the most out-of-division games would have the best overall record, since every division will have the same number of wins from divisional games. It's also true that the more games are played within the division the more it mathematically disadvantages the teams that play in better divisions and advantages the teams that play in worse divisions, in terms of year-end win totals. The more out-of-division games there are the more chances the better divisions have to accumulate improved win totals compared to the other divisions. 

Take last year's AL East. The teams had a 423-387 (.522) record overall with a 190-190 (.500) record against itself and a 233-197 (.542) record against all other divisions. That was with 76 out of 162 games for each team played within the division and therefore 86 of the 162 games played against everyone else. You could do (.542 x 86) + (.500 x 76) to get back to the AL East's average record of 84.6 wins and a .522 winning percentage. If you assume that the winning percentages stayed the same, but the schedule was scaled back to 68 games within the division (17 games against each divisional opponent instead of the current 19) and 94 games against everyone else, then your better win rate against the non-division teams gets weighted more heavily and instead you get (.542 x 94) + (.500 x 68) = a very slightly higher 84.9 wins and a .524 winning percentage.

So, it wouldn't move the needle too much unless there was a major schedule shift. But playing more divisional games does mathematically disadvantage the teams in the stronger divisions when compared to the win totals of other divisions. 

[BohKnowsBmore]

1 hour ago, 25 Nuggets said:

 

All the more reason to appreciate the run the Orioles have had recently.

 

"It has been a very tough division going on two decades.  Except for that one year the Orioles won it... It was complete trash that season!"

[BohKnowsBmore]

24 minutes ago, Spy Fox said:

It's true that whichever division wins the most out-of-division games would have the best overall record, since every division will have the same number of wins from divisional games. It's also true that the more games are played within the division the more it mathematically disadvantages the teams that play in better divisions and advantages the teams that play in worse divisions, in terms of year-end win totals. The more out-of-division games there are the more chances the better divisions have to accumulate improved win totals compared to the other divisions. 

Take last year's AL East. The teams had a 423-387 (.522) record overall with a 190-190 (.500) record against itself and a 233-197 (.542) record against all other divisions. That was with 76 out of 162 games for each team played within the division and therefore 86 of the 162 games played against everyone else. You could do (.542 x 86) + (.500 x 76) to get back to the AL East's average record of 84.6 wins and a .522 winning percentage. If you assume that the winning percentages stayed the same, but the schedule was scaled back to 68 games within the division (17 games against each divisional opponent instead of the current 19) and 94 games against everyone else, then your better win rate against the non-division teams gets weighted more heavily and instead you get (.542 x 94) + (.500 x 68) = a very slightly higher 84.9 wins and a .524 winning percentage.

So, it wouldn't move the needle too much unless there was a major schedule shift. But playing more divisional games does mathematically disadvantage the teams in the stronger divisions when compared to the win totals of other divisions. 

The results are the same, just makes for less spread out differences between divisions.  Said another way, all it does is bring the cluster of total division records closer to .500.

[Frobby]

20 minutes ago, BohKnowsBmore said:

"It has been a very tough division going on two decades.  Except for that one year the Orioles won it... It was complete trash that season!"

Not complete trash, but it was kind of a down year for the AL East (411-399).   Nobody other than the Orioles won more than 84 games.

[BohKnowsBmore]

1 minute ago, Frobby said:

Not complete trash, but it was kind of a down year for the AL East (411-399).   Nobody other than the Orioles won more than 84 games.

I was being somewhat tongue in cheek.  

[Frobby]

5 minutes ago, BohKnowsBmore said:

I was being somewhat tongue in cheek.  

Yes, I figured.    But it's true that it wasn't the best year for the division.  

[Spy Fox]

34 minutes ago, BohKnowsBmore said:

The results are the same, just makes for less spread out differences between divisions.  Said another way, all it does is bring the cluster of total division records closer to .500.

You're right. I had something mixed up in how I was thinking about it. Balancing the schedule would make the best divisions better, but it wouldn't mathematically change who the best division was. Oops!