Open Thread on Strength of Schedule

[Elster88]

Only two National League teams are on pace to win 90 or more games, the Mets and the Cards. And the Cards just barely. 90 wins is a .556 winning percentage, the Cards are at .558.

[Guest Giant Squidlike Creature]

Bump.

[Guest Rotblatt]

Victory Lap or Pit Stop?

In the 31 days between the start of September and October 1, we will play 30 games--a brutal stretch. Fortunately, most of it is against the plodding mediocrity of the NL and our 15.5 game lead will allow us to rest our starters liberally. While this should be little more than a victory lap for the Amazin's, there are a few compelling story lines to follow:

1. Clinch. If we clinch 18 games from now, it will put us in Pittsburgh on Saturday, September 16. The earliest we could clinch would be at home on September 6--against Atlanta, which brings us to #2.

2. Exorcise. At the start of the season, schedule-watchers like myself pointed to the final 6 games against Atlanta as the crux of our season. With the pennant firmly in our grasp, that is clearly not the case, but we still have a chance to host an event Mets fans have long waited for--the mathematical elimination of the Braves from the postseason. If we get mindbogglingly, extraordinarily lucky, we could both clinch and eliminate the Braves on September 6. Demon, I cast thee out!

3. Spoil. Florida is still in the running for the wild card. With 7 games against them in September, we could go a long way towards knocking them out.

4. Heal. Petey, Glavine, El Duque and Cliff need to get healthy and tune up for October. Our success in the post season depends on it.

5. Reach. With some inspired play, we can reach 100 wins, which would mark just the third time in Mets history that we've hit 100 or more wins--and the first time since 1988.

Hit List for August 28.

September/October Schedule

Mets: 3rd, .581 W%

Houston (20th, .487) 3 games A
Atlanta (17th, .496), 3 games H
LA (9th, .532), 4 games H
Florida (21st, .485), 3 games A
=blue]Pittsburgh (28th, .405), 3 games A
Florida 2, 4 games H
=blue]Washington (26th, .435), 4 games H
Atlanta 2, 3 games A
=blue]Washington 2, 3 games A

We face 2 crappy teams and one decent team over this 31 day stretch. Anything less than 15 would be an embarassment, but since we're probably going to rest our regulars, uh, regularly, I'm going to project only 16 wins. Anything more than that is gravy--with any luck, enough gravy to float us to 100 W's.

[dinosaur jesus]

It would be the fourth time with 100 wins, not the third.

.500 ball for the rest of the season would be okay with me--I'd like more, but I could live with that. It would at least guarantee the best record in the league.

[Guest Rotblatt]

Oops! Can't believe I forgot about '69, DJ!

So in BP's latest Standard Postseason Odds Report, the Mets finally hit a 100% chance of making the post season, which means that the Mets, in 1,000,000 simulations, made the post season every single time.

In the PECOTA & ELO adjusted reports, we failed to make the post season a couple of times, leaving us with 99.99998% shot in each (more or less).

Each report predicts we'll win either a little under or a little over 100 wins.

[Guest Yancy Street Gang]

When we know the playoff matchups, will Baseball Prospectus simulate a million playoff series? Or do they only do this exercise for the regular season?

[Guest Rotblatt]

]When we know the playoff matchups, will Baseball Prospectus simulate a million playoff series? Or do they only do this exercise for the regular season?

They might run the scenarios--I'm not positive. It's such a small sample size that if they do run them, they probably have lots of disclaimers.

They took up Bobby V's challenge last year, simulating the Chiba Lotte Marines against the White Sox, using the Davenport translations for Japanese players. That was pretty fun, even though the good guys lost, 4-1 (still a better showing than Houston, though!).

[Frayed Knot]

The 'Playoff Odds Report' and the mythical USA v Japan series that they did last year are two very different things. The first is simply a mathmatical exercise based on a defined schedule of games and the teams' results to this point. The other was purely speculation along the lines of a sophisticated Stratomatic approach.

Figuring the playoffs here could work the same way as the in-season report once you've got the matchup set - but it would change so much from day to day that there doesn't seem to be a lot of point to it.

btw, last night's win assured us of a winning season.
I'm sure y'all were worried about that.

[Guest Rotblatt]

Good point, Frayed! I was trying to suggest that BP will probably do SOMETHING to project odds of winning, but I couldn't remember what they'd done last year . . .

Do you have any recollection? I seem to remember them giving odds, maybe, but I could be wrong, and if they DID, I've no idea of what methodology they used.

[Frayed Knot]

I dunno, it barely seems worth it to me.
The benefit of the in-season report is that it takes the almost endless possibilities of the reg season - with it's dozens of games remaining, the differing schedules, the varying RS/RA ratios - and boils that down to general odds.

But in a small sample like a lose-and-go-home series all that stuff becomes meaningless. Each game is more-or-less a 50/50 proposition so it doesn't take a lot to realize that going from a 1-1 tie to a 2-1 lead sends your odds soaring.
Then, by the time you get near the end, you can pretty much figure out your own. If you're up 2-1 say in a best-of-5 there's only 3 possibilities left:
* Lose then Win
* Lose then Lose
* Win
On two of those 3 you advance and on one you don't ... ergo your odds are more or less .66667

You can attempt to account for those particular games by figuring home field edges or starting pitcher matchups - but by that point you're way into the speculation realm and away from actual odds.

[Guest Yancy Street Gang]

For a playoff series, it might make sense for them to run the simulation at the beginning of the series. Once it's underway I don't see the point of redoing them every day.