Is this consistent with one of the three 'old reports', or is it a whole new methodology?

I think it says something about the wonderful randomness of baseball that the biggest favorite to win a division still checks in at "only" 57.9%.

This thing consistently underrates the Phillies, year after year.

Mousing over the headings reveals no stat definitions. :-(

And if the Orioles are projected to have a .502 win %, how can they be projected to win 80.3 games? Are you rounding any anything above .0 UP?

Honestly surprised at how good the Giants look here, and that's coming from an overjoyed Market Street parade attendee.

Many of the questions regarding methodology or why the winning percentages and win totals do not match up are likely covered in Colin's debut article about the Playoff Odds:
http://www.baseballprospectus.com/article.php?articleid=13040

Surprised to see the "Wainwright less" Cardinals picked to win the NL Central. I'd place them third, behind Cincy & Milwaukee.

I was referring to the win total; however, in light of current situations with the team right now...I'll take it back.

What's the plan to update the depth charts (and I'm assuming this uses the projections in the depth charts)? The Reds depth chart still has Alonso as the primary LF and Carlos Fisher throwing more IP than Sam LeCure.

I really liked looking at the graphs and odds history that have been available in the past. Are those still hanging around anywhere or are there plans to bring them back?

Could you add current record to the table?

Anyone know where and when one can find the ELO adjusted playoff odds? I don't see them.

wow the Astros playoff chances are the same as Bluto's gpa in "Animal House." Wondering what team hit sub-0.1% in playoffs odds the fewest games into the season.

Just to be clear, expected win percentage is the winning percentage for all remaining games, not where you expect them to finish the season at, right?

It's the expected win percentage against a neutral schedule.

What was wrong with last year's system? I could understand it and I could take whatever stat I thought more appropriate. This is simply mysterious with no explanation. If we are here we are statheads so don't dumb it down!

jrmayne makes a great point here. I am not sure how much the actual stats (or third order stats) are used to project at certain points of the season. Clearly, the later into the season, the more the CURRENT stats mean, and the less the projected stats mean. somewhere, there is use for Bayes theorem, though I am not certain exactly how (not sure if BP is doing that) due to no real sample size used for projections.

look at the Twins. you originally project them to win 83 games (a little over .500). There second and third order percentage is less than .300 and we are over 1/4 into the season. What do you think their expected winning percentage is the rest of the year? I don't know your methods, that even with a healthy Mauer, I find it hard to believe they will play at a .500 clip. You assume a .499 clip, which tells me that you are either barely accounting for the current stats and/or attributing the fact that they are 14 games below .500 almost completely to Mauer.

Gardenhire is a great coach, but it seems to me, that with some version of Bayes theorem (say 3/4 weighting of projected stats and 1/4 weighting of current stats), the Twins projection should be closer to 70 wins.

GoodKingJohn - in the past, I'm pretty sure they have used some time-weighted combination of PECOTA projections and current team quality (3rd-order WPct, I think). The expected WPct currently used above, however, is 100% PECOTA/depth chart and 0% current stats. Of course, you're absolutely right that a better estimate of team strength would include this year's performance thus far; I believe they recently re-ran one of Rany Jazayerli's studies showing that by this point of the season current-year performance is, in fact, meaningful.

I'm not sure if or when the BP staff plans to use updated estimates of team strength, but in the meantime the playoff odds seem to be growing more and more meaningless by the day. Which is kind of a shame, since I've always enjoyed checking in on this report.

What I don't understand is why the "Expected Winning Percent" in the Playoff Odds Report table above doesn't seem to be updated in line with the corresponding values from the Team Depth Chart pages (which have been updated over the course of the season). Best I can tell, those numbers should match...

The Red Sox have won the past 2 days and their playoff odds have decreased. Is this because their true talent level is above that of the Rays and Yankees, so they are "expected" to win more games, especially when they were playing an inferior opponent, the Blue Jays?

Expected win % for the A's of .624 does not pass the smell test.

Please change "Expected Win Pct" to a name that is more accurate. It is not the expected winning percentage of the team; it's the true, schedule-neutral strength of the team.

This did not change from yesterday, though it says it was updated today.

If the odds for a team round to 100%, will the report show that or default to 99.9%

Does the current report reflect changes to the depth charts due to trade deadline action?

What are the sorting criteria for this report? The sorting for the projected last place teams in the NL look out of whack. (E.g., how the heck does Houston escape the cellar?)

Is SF over Arizona a mistake?

What explains the Phillies expected win percentage of .530? Are they that much worse than Atlanta? (.559)

The adjusted standings show 3rd order win pct for Phillies of .637, while Atlanta is at .544.

That SF/Arizona thing looks weird. Zona's up 3 in the loss column and the Giants have had more runs scored on them than they've scored this year. Take into consideration all of the DL time SF has now, and I don't think that looks right. Do we really believe that the Giants' second string is that good?

How do the cardinals beat the brewers yesterday without a 1 day change in their playoff odds? Clearly incorrect.

Did the Phillies dissolve their team?

I don't like your use of 100.0% odds when the Magic Number is still 1 or more. This is analogous to probabilities for T-tests that are sometimes reported as .000. We are all told in our stats training to interpret that as <.001 -- since there remains some finite chance that the difference is not significant.

So please max out your odds at 99.9%. Isn't that good enough?

The one day delta column reflects Saturday's games rather than Sunday's, even though the update time is Monday morning. I don't know if other columns have not been updated.

Sorry if this is a stupid question, but is there a way to find out what playoff odds were on a historical date (e.g., July 31st, 2011)?

How did TB win and their PS odds decrease, not only decrease, but decrease by more than the Angles who lost?

Does odds report include the possibility of the one-game playoff between Sox and Rays (thus projecting only one winner), or do the percentages represent the aggregate chance of the team either winning or tying for the division championship (allowing for two champions)?

I meant wild card champ, not division.

Red Sox 55 percent to the Rays 45 percent with two games left????? I don't think the computer is taking into account that while the Rays will be playing a Yankee team that really doesn't care beyond keeping their timing sharp and getting some backend pitchers some innings, the Red Sox will be facing an O's team for whom this series is their season.

Hopefully someone will soon be explaining why/how, tied at game 161, the Cardinals' and Braves' post-season chances are so enormously different (and, considering their oppositions, in the wrong direction!)

Why does Boston have a higher sim win but Tampa has a higher percentage to make playoffs? With just 1 game left this doesn't make much sense. Also the simulation include a tie breaker if played?

Are you guys going to run the simulation for the playoffs?

Harrumph:

Updated 5:23 am Pac, Oct 1 2011

and Boston has 48.8% chance of winning the wild care?

I'll repeat Jay Taylor's comment above. It would seem to be easier to simulate the playoffs and world series than the regular season.

To echo the above comments: why aren't you running simulations for the playoffs?