MattS' Conditional Playoff Odds
Going into today, most systems had the Phillies at approximately a 99%+ chance of making the playoffs. Clearly, with a Phils loss, a Mets win, and a Brewers win, those odds had to go down. How much? Coolstandings.com is now saying 98.9%, still pretty good. BaseballProspectus.com, Sportsclubstats.com, and maybe even Tradesports.com will have estimates tomorrow morning.
Unsatisfied with the cold facts, I've decided to dig a little deeper and give myself a sense of the stochastic roller coaster ride I'm about to take!
Firstly, I'm saying 98.0% chance of making the playoffs. From there, I'll explain some breakdowns.
{EDIT: I've changed a few numbers due to some calculation errors I made}

Using the fact tradesports.com odds of certain teams winning baseball games are more or less accurate, I'm approximately what to expect from here out. I'll use the following guess:Odds that the Mets, Phillies, or Brewers beat a team other than the Cubs (Braves, Pirates, Nationals, Marlins) are approximately 70% on any given day. Odds that the Mets or Brewers beat the Cubs are about 60% on any given day. Obviously, this isn't exact, but it give me an idea of what's in store, but given those assumptions, the following things should be true:
Our odds of clinching on the following days are:
Thursday: 6.3%
Friday: 45.3%
Saturday: 32.4%
Sunday: 11.9%
Monday: 1.7%
Tuesday: 0.4%
Keep in mind that if the Phillies have clinched the playoffs by Sunday, they won't need to use Cole Hamels-- those odds are 84%. (I assume we wouldn't waste Cole on Sunday to win the division if we had a playoff spot wrapped up, but our odds of winning the division by Sunday are 51.3%.)
What about if you just want to know what we need to do to win?
Our odds of making the playoffs given that we have the following records is:
0-4: 60.2%
1-3: 79.2%
2-2: 96.8%
3-1: 100%
4-0: 100%
This makes you feel a good deal better, doesn't it?
For us to win on Thursday, it requires that we win tomorrow and that Brewers lose to the Pirates tomorrow and Thursday. That seems unlikely...6.3% about actually.
Going into Friday, we can expect to have the following possibilities:
6.3% pre-clinched
32.1% magic number of 1
43.8% magic number of 2
17.8% magic number of 3
Obviously, tomorrow will give us a better sense of these, but with the 32 possible outcomes of the five possible games that can happen between now and Thursday, this is how it breaks down.
Going into Saturday, we can expect:
51.63% pre-clinched
33.42% magic number of 1
13.11% magic number of 2
1.84% magic number of 3
Calculating this out for Sunday is a bit much right now, but maybe after some more games happen, I can narrow this down.
For those who are curious, our odds of winning the division by Sunday are approximately 76.4% and our odds of having a one-game playoff with the Mets on Monday (needed only if the Brewers have a record as good or better than the Phillies and Mets) are about 2.0%, so our odds of winning the division are approximately 77.4%.
Let's hope that this gets easier to calculate after tomorrow...a win would probably help that cause!
0 recs |
5 comments
Comments
I would really prefer to win tonight (well duh…)
If the phils lose, they open up the possibility to the mets that the teams will be tied going into their final 3 game serieses.
I know that worked out last year, and against the same teams, but I’d rather not have it come to that.
by Bilzo on
Sep 24, 2008 7:33 AM EDT
reply
actions
0 recs
Reading this gave me a tiny taste of the nauseated anxiety I expect to be feeling over the next 96 hours or so—hopefully only for the first half of that time frame, but my inner Negadelphian is saying, “no shot, chump.”
by dajafi on
Sep 24, 2008 12:44 PM EDT
reply
actions
0 recs
haha, see my post on WholeCamels’ playoff scenario. I am experiencing the same thing. Perhaps it is that Jewish neurosis.
Reading this stuff automatically makes me think of the worst case scenario. It’s like taking a new medication with a laundry list of potentially serious side effects that only like .01% of the test subjects experienced and still worrying that you have a chance of dying from it.
by FuquaManuel on
Sep 24, 2008 12:50 PM EDT
up
reply
actions
0 recs
I think all you did was verify your starting assumptions. You can do the same thing using simple subtraction.
What you are basically saying is there is a 60% chance that any of these teams will win on any given day. Since there are no more than two possible outcomes—win or lose—the Phillies gain 0.60 every day they play, and another 0.40 every day either the Mets or Brewers play. So, no matter how you crunch the numbers, the Phillies will likely game one “magic number” today, 040 on Friday, since they are off. And the result is they are more than likely to clinch a playoff spot on Friday or Saturday, and more than likely to clinch the division on Saturday or Sunday.
See…
WC DIV
Starting Pt 3 4
Wednesday 2 3
Thursday 1.6 2.6
Friday .6 1.6
Saturday .4 .6 -.4
Sunday -
by Squonk64 on
Sep 24, 2008 2:38 PM EDT
reply
actions
0 recs
i don’t know where to start. firstly, this is about probability, NOT CERTAINTY. it’s about figuring out the chances of the phillies going 0-4, 1-3, 2-2, 3-1, and 4-0 and of each of the phillies and brewers winning anywhere from 0-5 games, then determining the order possibilities. there are over 16,000 different outcomes for how the last 14 Phillies/Mets/Brewers games could play out if you account for order.
i’m also calculating joint probability distributions— we only need to finish below both of them. all you just did was figure out the average expected records of each team. that’s not the point here. the point of calculating playoff odds is to measure standard deviations, not means. yes, by your calculation, we will probably make the playoffs on average. that doesn’t tell us anything we don’t know.
by MattS on
Sep 24, 2008 4:12 PM EDT
up
reply
actions
0 recs












