The Comeback
Tommy Hanson has been tremendous this year. He was not tremendous today, giving up eight runs in the second inning, as the Reds chased him from the game. After Laynce Nix doubled in Jay Bruce and Ramon Hernandez, the Braves odds of winning stood at just 2.3 percent.
The Braves would mount a little bit of offense, eventually cutting the lead to 9-3 and seeing their win expectancy peak at 7.3 percent in the 5th inning. But, after a few more dead end rallies and a six run deficit heading into the 9th inning, the Braves odds of winning stood at 0.2 percent. In other words, not good.
Then, this happened.
Troy Glaus singles.
Eric Hinske singles.
Yunel Escobar singles.
Nate McLouth singles.
David Ross walks.
Martin Prado reaches on fielders choice/error
That brought Jason Heyward, hero of the Braves season, to the plate representing the winning run. The major league leader in WPA was given yet another chance to send Atlanta into delirium.
He struck out.
With their odds of winning back down to 18.5 percent, Brooks Conrad, he of the career .289 wOBA, launched a launched a walk off grand slam off Francisco Cordero.
Baseball – it’s amazing.
41
You forgot the name of the illustrious Brooks Conrad in the last paragraph :)
Dave,
As a Braves fan, this was a great win. Do you know when was the last time a team came back from 0.2% WE to win a game?
I’d guess roughly 500 games ago? Joking of course, but this is a 1/500 chance.
I seem to remember an AL team coming back from down 7 with 2 outs and no one on in the ninth five or six years ago.
http://www.fangraphs.com/wins.aspx?date=2009-04-08&team=Phillies&dh=0&season=2009
this may have been it. i can’t find what the lowest point of WE for the phils was. but on BBref they had it at 1%. t hey don’t go into decimals i don’t think.
I hate charts, but thats the most amazing one I have ever seen. Maybe every game should be charted like this?
They are.
I don’t understand this post.
welcome to fangraphs, james! have a look around!
The non-sarcastic reply:
http://www.fangraphs.com/scoreboard.aspx?date=2010-05-19
Charts are great, don’t hate.
it’s exactly games like this that make me want to start a fantasy league based entirely on WPA. not all home runs are created equal, and this one was worth .815 WPA.
case in point… Joey Votto hit a grand slam earlier in the same game, and it was only worth .263 WPA. not all home runs are created equal, yet all of our current fantasy leagues treat them equally.
Except, they are created equal. The 4 runs driven in by the GS in the bottom of the 9th had no more input in to the final score of the game than the 4 runs Votto drove in in the 2nd.
They might have been many times over more exciting to watch, but the only reason they were worth so much to begin with is because Votto had put the Reds so far ahead to begin with.
but they’re not created equal… the grand slam at the end of the game, while also only driving in 4 runs, did it with much higher leverage. that’s why it was much more exciting, and “excitement” is apparently something we can measure :)
this is an intriguing and interesting idea. i’d probably bite, but i don’t think it would be as popular as the existing fantasy systems because it’s impossible to predict WPA and it would be almost completely based on luck.
maybe weekly, but if you did season long roto style, don’t the best players tend to end up with the best wpas?
actually, i was thinking weekly. sure, WPA is impossible to predict on a weekly basis, but so is the exact amount of singles, doubles, and triples a guy will get in a week. the point is that the best players are likely to accumulate the most WPA over the course of the season, just like they would with normal H2H fantasy points. it’s not a new idea, just a different metric to use on their actual contribution to their team’s winning.
There would be no way to predict that league, but I’d like to play in it.
Crazy that a rookie would be the most valuable player in this league by a pretty good margin so far this year.
Heyward 2.86
M Cabrera 2.61
Votto 2.35
Youkilis 2.32
Ethier 2.17
Soriano 2.05
McGehee 1.75
this may be not so surprising. his pLI is 1.13, so his production (which has been good) means more overall than it would in a regular fantasy league.
it’s also interesting to note that the leader in pLI is Francoeur at 1.27, but he totals a -0.37 WPA.
Wow, I watched part of this game while at a restaurant back when it was 9-3, I thought the Reds had it for sure. Gotta love baseball miracles!
It’s actually not that amazing stastically. 2400 games in a year, so that means when a team is down to a .2% chance of winning, they will come back 5 times.
Next time John Grabow will probably be pitching.
Umm, yes there are over 2400 games in a year, but how many games where one team is down to a .2% chance of winning? More likely this kind of event only happens once a year.
2400 games * .2% = 4.8 games. Simple probability.
You’re missing some math there, Everett.
To expound for Everett, think of a 1-0 game. Neither team will ever be down to a 0.2% chance of winning at any point in the game – even with 2 outs in the 9th and bases empty, the chance is probably in the 5% range.
Lots of close games will work out that way.
It’s not 0.2% of ALL GAMES that would be won in this manner. It’s 0.2% of games with that particular situation that would be won.
It only makes sense that a living legend that is Brooks Conrad would hit the walk off grand slam.
Looks like the Mariners got inspired by the Braves and staged their own mini-comeback!
As a Jays fan, the only positive from all of this is that the Mariners will be stuck with Griffey for a while longer because of his late inning “heroics”
The sad thing is that exact same swing would have produced a heroic homer in years past. Now a bloop single to RF is the most we can expect.
(I’m not sure I see how the M’s getting/staying worse is a positive for a Jays fan. With the M’s yet to play the Yankees or Red Sox, not to mention the Tigers or Twins, I would think you’d want the Mariners to get better to give you a better shot in the AL East and, especially, the wild card.)
mike leake got a no decision b/c of this dam u cordero
Left him blowin’ in the wind, eh?
The third “single” was basically a tailor made double play ball that Cabrera misplayed. I would joke about him calling the official scorer to get it changed to a hit, but in this case he doesn’t have to.
I don’t absolve the Reds players of their horrific meltdown, but that 9th inning was a classic case of Dusty being Dusty. If you ever want a snapshot of managerial malpractice, that was it. His only job is to get the best out of his players, and he put a number of players in a position to fail today. As we’ve seen time after time, as soon as things move the slightest bit off script from “the book” Dusty has no clue what to do.
- Used Micah Owings to PH for Leake but didn’t have him come in to pitch when he knew he was going to a long reliever anyways.
- Had Mike Lincoln start a 3rd inning of work despite a history of quickly losing effectiveness after 30 pitches or so.
- Did not have somebody up at the start the 9th inning (as far as I can tell) in case Lincoln started to lose it.
- Brought in Massett to try and stop the bleeding, even though it was his 5th appearance in 7 days.
- Had Rhodes face Heyward and then pulled him for Cordero, despite the PH being a lefty and Rhodes being dominant against lefties. Dusty didn’t have to make that call before the PH was named.
- Left both Cairo and Cabrera in when Janish is the best (healthy) defender we have at both positions.
- Started Cairo over Janish to being with.
It’s hard to win when your defense makes 5 errors, but we still should have. Players make mistakes, that happens. But this was Dusty’s version of HanRam’s jog to the LF corner as far as I’m concerned.
- Had Rhodes face Heyward and then pulled him for Cordero, despite the PH being a lefty and Rhodes being dominant against lefties. Dusty didn’t have to make that call before the PH was named.
For the record, Brooks is a switch hitter.
and relatively much better as a lefty (which is how he hit his HR)
Are you basing that on his 16 career ML PA against LHP? As a minor leaguer he actually didn’t have too much of a platoon split, posting a line of .238/.324/.470 against LHP and line of .258/.336/.477 against RHP.
I’m curious, since you have a few years of data now, how often has a team with a 1 in 500 chance of winning come back to win the game?
I just happen to remember it — not sure if there’s a way to search for it — but this game (http://www.baseball-reference.com/boxes/CLE/CLE200905250.shtml) comes to mind, where the Indians scored two runs in the eight and seven in the ninth to win the game against the Rays.
There was that game between the Mariners and Indians a long time back where the Mariners went up 12-0 and ended up losing.
http://www.baseball-reference.com/boxes/CLE/CLE200108050.shtml
The percentages here aren’t rounded to 2 decimal places, but I assume it was close to .2% at some point.
This is actually deja vu for Reds fans. On May 2nd, 2005, the exact same thing happened:
http://www.baseball-reference.com/boxes/CIN/CIN200505020.shtml
Reds went into the ninth up 9-3 on the Cardinals, David Weathers and Danny Graves blew it. John Mabry played the role of Brooks Conrad.
What about last season? Red Sox blew a huge lead against the Orioles?
That wasn’t all in one inning, or in the 9th, the Orioles scored a bunch of runs in the 7th and a bunch more in the 8th.
To expound on the math problems a few of you have made: if a team comes back from a WE of 0.2% to win, that DOESN’T mean that it happens once every 500 games or 4.8 times per year. It means it happens once every 500 times a team is in that specific situation (score, inning, outs, runners, etc.). Of course, multiple situations can have a WE of 0.2% or worse, but they don’t happen every game. Once a year or every few years for this caliber comeback to occur would be a good guess, but it’d interesting to see the exact numbers.
There was a point in this game where the WE stood at .1%, instead of .2% Dave used that since it was, tidily, resting right there at the start of the bottom of the 9th.
How often do teams come back from .1% WE? Maybe about once per year, if that.
http://winexp.walkoffbalk.com/expectancy/search
With the data from 1977-2006 (but not 1999) at the Win Expectancy Finder, the situation at the start of the bottom of the ninth (home team down by 6) has happened 1993 times. The home team won 3 times: 0.002%.
The situation when Conrad stepped into the box – home team down 3, bases loaded, one out – has happened 132 times. Home team rallied 15 times (0.114%).
http://mlb.mlb.com/video/play.jsp?content_id=8181161
MLB video highlight of the play. A must watch if just for the comment made at about 0:55.
Joe Simpson ftw
I know it’s not even Memorial Day, but I think we can declare that the Game of the Year right now. Jesus Jones!