why 580 expected win percentage for cardinals? easy schedule?

Happy to have the graphs back and better than ever. Sweet!

Owowow. For a second I thought I'd stumbled on one of those damned columns about Why Baseball Doesn't Work In Denver or something.

Guess I missed the previous incarnation(s) of the graphs.

I'm probably just a huge dumbie, but could someone explain why the historical data doesn't match the probabilities on this report at the time? E.g., in April, the Red Sox's P was much, much higher than .2 on the BP playoff odds report. Is this backwards adjustment the same thing that causes the 1-day and 7-day deltas to be arguably wrong? A sort of, knowing what we know now, Boston's P was < 0.2 in April, not > 0.8 as we reported at the time?

Do you have Oakland's and Anaheim's graphs reversed? Preseason, Anaheim was considered a near lock to make the playoffs but your graph shows them with a low score. I know they got off to a slow start, but it doesn't look right

The divisional graphs are great. I understand them without a caption.

However, what is plotted on the narrow graph below each of the divisional graphs?

A thought. The wild card position has changed dramatically this year. Winning a wild card slot is no longer virtually equivalent to winning a division, hence wouldn't it make more sense to equate winning one of the 2 wild card slots with winning half of a divisional title? From this perspective, the 1 day "wild card" game is equivalent to two teams ending the 162 game season with identical records and thus having to play a 163rd game to make the playoffs. Once the 163rd game is decided, the playoffs begins with 4 teams per league.

With this perspective, the total value of winning THE wild card playoff slot is 100%, not 200% as it is labeled now. And the probability of a team making the 4-team playoffs is more accurately calculated. The value of winning the divisional title is better reflected in the calculations. (The Rays chances of making the real playoffs are much worse.)

It might be an idea worth pursing. If explained properly, then it would be helpful.

File the Dodgers under "sad but true." It kind of stinks to be a rational fan sometimes.

Something's odd with the deltas again (as of 9/14). The Braves are shown as having a one-day delta of 59.7% but they've been above 98% for weeks now. There are other oddities, including huge one-day spikes for the Orioles and A's.

Does a 100% chance mean just that? In other words, even though the Braves have not clinched a spot by virtue of the number of games ahead of the Dodgers, is it impossible to find a combination of results moving forward that would result in the Braves not being one of the WC teams? That seems unlikely at this point, but I would think the number would be 99.9 rather than 100.

I've been thinking about streaks and how they might impact playoff odds. Has anyone at BP (or anywhere else) done the work on the frequency of winning streaks? I've seen the fact that 'more work needs to be done' referenced a few times. It doesn't seem that hard. Do streaks happen more or less often than probability would suggest they do if a non-streak-considering binomial model were used?

I raise it because it seems that 20%-chance events are happening more than 20% of the time. This year, BAL, CWS and DET are all going to see 20%-chance events happen over the last couple weeks. Only LAD and PIT are going to have 20%-chance events NOT happen. (Maybe OAK and LAA ... depending on where you draw the line.)

And last year, we all know that STL, STL, BOS and TAM all had less-than-20%-chance events happen.

So have we not accounted for streakiness? Teams get hot? Teams fall apart?

Something very wrong in this iteration. Tampa Bay is eliminated but you have them at 49.7% for the Wild Card. Texas is in and if they win tonight they're the West champs, you have them at 0% for the division champs and 50.3% for the wild card. Hope you can correct soon.

Brewers and Cards also not correct.

Yup, while I'd love the A's to have a 100% chance of winning the division, methinks you're having a little fun with us this morning.