12 replies to this topic

[mabrowndog]

As BtP noted in his Angels series preview thread, this marks the final West Coast trip for the Sox -- and it's only 6 games in two cities, with Seattle to follow Anaheim. As hard as that may be to believe, this is just one benefit of playing Oakland on the road (including the Tokyo games) at the start of the season. By the time this swing is completed, Boston will have already played 56 of its 81 road games.

At first glance, it seems their primary division rivals might have much tougher rows to hoe. I've linked each team's schedule to their bolded name below:

Red Sox
They'll return next weekend to start a 9-game homestand (NYY, LAA, OAK) that takes them through August 3. They'll then hit the road for games in KC and Chicago before heading home to play the Rangers and Jays. Next, road series in BAL, TOR and The Toilet before finishing August at Fenway against the White Sox. Note that there's an off-day between each of their road series in August.

September begins with the Sox finishing their homestand vs BAL. Following an off day, they're in Texas for just three games before returning home to face the Rays and Jays. Their final road trip takes them to TOR and TB, and they finish up at home against the Indians and Yankees.

Tampa Bay
The Rays come out of the break at home vs TOR and OAK, then travel to KC and TOR before an off-day on 7/31 after 13 straight games.

They start off August in the Trop vs DET and CLE before starting a trip to SEA, OAK and TEX. The final game of the Rangers series is Sunday night on ESPN, and they'll have to fly straight home to host the Angels in a three-game set without an off-day. Then it's off to Chicago for three before finishing August at home vs TOR and BAL.

September looks absolutely brutal, with 27 games in 26 days, only 10 of them at home. They've got just one lousy off-day, and that comes on the road between series at Fenway and the Toilet. It's also their toughest month of the season in terms of schedule strength. They start out at home vs the Yanks, then embark on a 10-day swing to TOR, NY and BOS. Next, it's back home for the Sox and Twins. The season closes with 4 in Baltimore (including a twi-night doubleheader on Tuesday 9/23) and 4 more in Detroit.

Yankees
13 of their first 16 will be at home. After hosting OAK and MIN, they hit Fenway for three. The Sox finale is a Sunday night ESPN game, and they open up a homestand the very next evening vs the O's and Angels. Next comes a 10-day trip to Texas, Anaheim and the Twin Cities. A 3-game home set with the Royals is sandwiched by a pair of off-days, then they head to TOR and BAL to close out August.

September starts with a four-city road trip. What once was an off-day on 9/1 will now be a Monday afternoon make-up game in Detroit. Then it's off to Tampa, Seattle and Anaheim. A day off on 9/11 precedes a 10-game homestand vs TB, CWS and BAL. Their final off-day is 9/22, and they finish the season in Toronto and Boston.

----------------------

So, do the Sox really have the scheduling edge the rest of the way? Let's take a look:



Log5, using Bill James formula to calculate Expected Winning Percentage, is explained here. Note that it only compares the winning percentages of the two teams, but does not compensate for home field advantage.

What we see right away is that Tampa Bay has the most games to play, but their schedule strength is pretty soft -- especially for their 37 road games. September, with that lone off-day, will be a critical test for a young team that's short on pennant race experience. In addition to a physically taxing slate, the Rays' Log5 against September opponents drops from .573 to .554.

Meanwhile all but 6 of the Yankees' remaining games are against teams playing .479 ball or better at the break, and 36 of them are against the AL's five best teams (LAA, BOS, TB, CWS, MIN). Their only respites are series vs KC and SEA. The Log5 method has them playing sub-.500 ball the rest of the way. As I note below, that should be taken with a grain of salt, but there's no denying that they're going to be playing the cream of the crop.

As for the Sox, their slate appears tougher than I first thought when perusing their schedule. But those six well-placed off-days in August and September could prove crucial in getting rest for the pitching staff should they hit any extended bumps in the road. After this week, they'll have no more West Coast flights. Ditto for doubleheaders and make-up games. Those factors, and the absence of late ESPN telecasts that lead into next-day road games, ought to ease the stresses of travel.

Obviously Log5 is fluid and using it to project games several weeks down the road is more for fun than predictive value, so I'll try to update the chart every week or two and repaint the picture. But right now I'm feeling pretty damned confident about this team's outlook relative to the competition.

Edited by mabrowndog, 23 July 2008 - 07:42 PM.

[biollante]

I think the days off will be key for the Red Sox. This could be the scheduling part that gives them the edge more than who they are playing.
I feel good about the second half - but see Manny thread about punishment - I don't want any distractions and this could be the type of "issue" that the media will drum beat us and the team to death with.

[Fratboy]

Obviously Log5 is fluid and using it to project games several weeks down the road is more for fun than predictive value, so I'll try to update the chart every week or two and repaint the picture. But right now I'm feeling pretty damned confident about this team's outlook relative to the competition.

Thanks for running this - I was planning to do this over the weekend but you saved me some time.

My only question about this - and I realize this may be noise - is if your calculations are taking into account the Sox and Rays' performance on the road this year. Given their sub 0.500 average on the road so far and 0.513 SOS for their road opponents, I have some difficulty believing they're gonna put up a 0.575 against them the rest of the way. Likewise for Tampa Bay.

I agree that I think it's gonna be close, but taking into account peripherals and whatnot, I think the Sox win the division by three games, something like 95-67 to 92-70, with the Yankees continuing to be irrelevant.

[mabrowndog]

Thanks for running this - I was planning to do this over the weekend but you saved me some time.

My only question about this - and I realize this may be noise - is if your calculations are taking into account the Sox and Rays' performance on the road this year. Given their sub 0.500 average on the road so far and 0.513 SOS for their road opponents, I have some difficulty believing they're gonna put up a 0.575 against them the rest of the way. Likewise for Tampa Bay.

Nope, I just presented the data generated by James' formula, along with the standard disclaimers. As defined by James, Log5 doesn't factor in a team's home or road record -- just the overall winning percentages of the two teams being compared.

That .575 road projection for the Sox is derived from their current overall .588 percentage and their upcoming road opponents' collective overall mark of .513.

[behindthepen]

coolstandings.com is also a good way to look at this. Their "smart" standings uses RS and RA to simulate the seasons going forward, so it basically has teams winning at their pythag rates for the remainder of the season, which obviously favors the Sox.

Right now their predicting 94 wins, 90 for TB and 87 for ny. Also 93 wins for LAAoA. The question mark for me is whether or not the hankees will do better than league average scoring going forward, or whether they can get that 4.69 RPG up.

[TomRicardo]

The Red Sox have the easiest schedule of any AL down the stretch. Ridiculously easy. No more West Coast trips. Really it is hard to see Tampa Bay surviving the meatgrinder of September they have. With Ortiz coming back and Lugo disappearing, the Red Sox are probably going to run away with the division.

[Fratboy]

The Red Sox have the easiest schedule of any AL down the stretch. Ridiculously easy. No more West Coast trips. Really it is hard to see Tampa Bay surviving the meatgrinder of September they have. With Ortiz coming back and Lugo disappearing, the Red Sox are probably going to run away with the division.

I'm not sure how you define "easiest," but I'd certainly go with "least taxing" for your reasons of no more west coast trips, more offdays than any other team in MLB, and no doubleheaders courtesy of the good fortune they've had with weather.

In terms of strength of schedule going forward, the Sox have one of the tougher schedules in the league:

SEA 0.535
NYY 0.534
TOR 0.534
BAL 0.529
BOS 0.520
KCR 0.516
CLE 0.515
TEX 0.514
DET 0.512
TBR 0.510
CHW 0.509
OAK 0.506
LAA 0.503
MIN 0.491

I'll hold to my earlier prediction that the margin in the AL East comes down to 3 games or fewer.

[5belongstoGeorge]

Which variable, less taxing or strength of opponent, has a larger effect on wins and losses? How would you quantify a less taxing schedule and separate that effect from other factors such as road/home day/night and so on?

It seems to me that strength of opponent is the more dominant variable of the two, although a short flight and more days off can certainly put a team in a better position to win.

[yep]

Which variable, less taxing or strength of opponent, has a larger effect on wins and losses? How would you quantify a less taxing schedule and separate that effect from other factors such as road/home day/night and so on?

It seems to me that strength of opponent is the more dominant variable of the two, although a short flight and more days off can certainly put a team in a better position to win.

For the '08 Sox it appears the only variable that really counts for anything is whether they are playing at home or on the road.

I don't know the answer, but it wouldn't be that hard to look at actual RS and RA numbers on opposite-coast away trips compared with projections based on overall pythag win % weighted appropriately for home and road records and see whether there is a generic penalty for faraway travel. E.g. take a sample of teams from one coast or the other, use their overall road numbers to calculate pythag win %, then look at opposite coast teams they have visited and tally up their home pythag %. Use those two numbers to figure out the expected win % and compare that to the pythag win % generated by actual RS and RA numbers on bicoastal road trips. A meaningful sample size would probably have to include a lot of different teams over different years, but nothing too complicated-- the data entry would be the hardest part, and if you don't like pythag % you could use pretty much any other metric the same way (eg team A road OPS against vs team B home OPS or whatever).

Edited by yep, 23 July 2008 - 11:58 AM.

[behindthepen]

I think that given the H/R splits for the Sox, getting key games at home (6 vs. NY, 3 vs. LAAoA, 3 vs. TB) is a pretty huge factor in their favor. Both TB and ny have lots of road games vs. contenders.

BOS
vs. teams over .500
home 21
away 13

vs. teams under .500
home 13
away 13

TB
over .500
home 17
away 19

under
home 9
away 18

NY
over
home 15
away 23 (6 each vs. BOS and LAAoA)

under
home 12
away 12

[Barbara]

Right now we are at 102 games, TB & NY at 99. Somewhere along the line having those extra 3 days off has got to help us. I hope.

[TomRicardo]

For the '08 Sox it appears the only variable that really counts for anything is whether they are.

No the Red Sox have been absolutely putrid outside the eastern time zone (6-11) and long road trips (more than 6 games) at 14-23 as opposed to 7-6 on short road trips. Every single losing streak (more than one loss in a row) except 2 have been on the road (we lost two at home in a row to St. Louis and we lost two to LA at home before losing three more in Tampa). We have been terribly unlucky in Tampa losing all six by a margin of 9 runs.

This plays better when it is home and not on a long trip.

[yep]

No the Red Sox have been absolutely putrid outside the eastern time zone (6-11) and long road trips (more than 6 games) at 14-23 as opposed to 7-6 on short road trips. Every single losing streak (more than one loss in a row) except 2 have been on the road (we lost two at home in a row to St. Louis and we lost two to LA at home before losing three more in Tampa). We have been terribly unlucky in Tampa losing all six by a margin of 9 runs.

This plays better when it is home and not on a long trip.

Uh, not quite sure how the sentence you quoted got truncated, so I'm not sure whether you were responding to my actual post or to the nonsensical line that you actually cited, but I was in agreement that the team has been vastly better at home.

As for the better performance in "short road trips," I'm guessing that you mean mid-april in CLE and NYY (2 games each, 3-1) mid-june against CIN and PHL (3 games each, 4-2) and the 3-game set in Tampa where they got swept. Is there a real conclusion that can be drawn from this set, more valid than their excellent performance on tuesdays? This is a real question, not rhetorical.

Overall on the road they have scored 237 and allowed 243, and should be about a .480 winning percentage, about 4 games better on the road than they have been.

I know that SSS and normal variance arguments get tiresome when the season is half over, but it looks to me like they've been unusually unlucky on the road. Looking at the schedule, there is no obvious and repeating decline over the course of long trips, like where they consistently fade after the first six games on the road. Out of the 4 road trips lasting more than six games so far, in half of them they won more games in the second half of the road trip than they did in the first half. So it seems the "long road trip" effect, if this season's record reveals one, would have to be kicking in the moment they leave home on a long trip and not, for instance, the product of exhaustion after four or five days away from home.

I'm not contesting that life on the road can and usually does make it harder to win games, but I'm not seeing any obvious symptoms of schedule-induced losses, other than maybe the return from Japan. It looks like this team has been performing at almost a .500 clip on the road in terms of how well they score and prevent runs, but that they just don't have the wins to show it yet. The good news is that regression to the mean works both ways, so there is every reason to think that identical performance will yeild better results going forward.

Edited by yep, 23 July 2008 - 10:20 PM.