I believe there are a few places you can find this, or you can do the research yourself at retrosheet.org, but I stumbled across a site that calculates the Win Expectancy Formula using a sample size of every game from 1977 to 2006. I found the page at a site called Walk Off Balk and the page with the formula can be found here.
The premise is simple. You input certain situations from a particular game, and, using every game from 1977 to 2006, it gives you a percentage that a team can use to determine their chances of winning that game based on the thousands of identical games in those three decades. So, you enter whether home or visitor is batting, what inning it is, how many outs there are, how many men are on base, and what the score differential is (up to seven either way**). Run the program and it will give you the number of times the home team won games in the exact situation and a percentage for home team win expectancy. From this, you can also subtract that percentage from 1, and find out the visitor’s win expectancy.
So I thought I would run some numbers from some game we are all familiar with to see what the win expectancy was for the teams in play (the Astros in most cases).
Opening Night 2007, April 3, 2007
The Astros had a one-run lead in the top of the ninth with two outs:
Team batting: Visitor
Inning: 9
Outs: 2
Bases: empty
Score Differential: -1
In 3728 games, the home team has won this exact situation 3623 times since 1977. The home team’s win expectancy is 97.2% of the time. So there was only a 2.8% percent chance of the Astros losing. And they lost. Brad Lidge.
October 17, 2005 - Pujols/Lidge playoff game
The Astros had a two-run lead in the top of the ninth with two outs:
Team Batting: Visitor
Inning: 9
Outs: 2
Bases: Empty
Score Differential: -2
Astros’ Win Expectancy: 99.3% (3302 of 3325)
October 9, 2005 - 18 inning Burke walk off HR game
The Astros were down 5 runs in the bottom of the 8th with no outs:
Team Batting: Home
Inning: 8
Outs: 0
Bases: Empty
Score Differential: -5
Astros’ Win Expectancy: 1.6% (44 of 2719)
October 15, 1986 - NLCS game 6 - 16 inning classic
The Astros were up 3 runs in the top of the ninth inning with no outs:
Team Batting: Visitor
Inning: 9
Outs: 0
Bases: Empty
Score Differential: -3
Astros’ Win Expectancy: 98% (4929 of 5028)
And for you Cubs fans:
October 14, 2003 - the Bartman game
The Cubs were up 3 runs in the top of the 8th with one out:
Team Batting: Visitor
Inning: 8
Outs: 1
Bases: Empty
Score Differential: -3
Cubs’ Win Expectancy: 97% (3440 of 3545)
Any other games you want me to look at? Or you can follow the link and do it yourself.
**I believe the number was cut off at seven because either the home or visitor coming back from that deficit or more is so rare over the course of 30 years, the percentage would be irrelevant to report.
well, it just goes to show that there is no 100%
and that in baseball, theres just one word - youneverknow
so what was the win expectancy of game 7 NLCS mets vs cards, mets down by 1, 2 outs, bases loaed bottom of the 9th?
Finally some love for my Cubbies. (Although,this isn’t my favorite game to remember). But, at least you’re being fair and including “Let ‘em win Lidge” in there.
I would be curious to see the formula for the 1984 NLCS Game 4 between Cubs and Padres. Padres batting in bottom of 9th, one out, time game. (Tony Gwynn hits a single and Steve Garvey homered. Cubs lose 7-5 and go on to lose game 5 and series.)
that should say “tie” game, not “time” game.
The answer to the two comments left here:
Lisa,
The Cardinals Win Expectancy for that game in that situation (B9, 3 on, 2 outs, 2 runs down) was actually quite surprising. Out of 214 times this exact situation happened, the home team (or the Mets) won 46 of them or 21.5%. So even though the Cardinals were up 2 with 2 outs, they still only had a 78.5% chance that they would win. Beltran just didn’t get the bat off of his shoulder.
Tim,
That exact situation (home team up to bat in bottom 9, 1 out, no one on, score tied) has happened 4,347 times in 30 years and the home team (or the Padres) won 2641 or almost 61% of the time. If they would only have gotten Gwynn out, the Cubs’ Win Expectancy would have jumped from 39% to 45%.