Hardball Times Recap

[DBAHO]

Interesting Read, from one of the best Baseball sites out there;

 

So, now can we stop talking about "small ball?" Please? Rick Sutcliffe, I'm talking to you!

 

The White Sox cruised past their more colorful foes yesterday, 14-2. Ten of the ChiSox's 14 runs scored on home runs: two by A.J. Pierzynski, and one each byPaul Konerko, Juan Uribe and Scott Podsednik. Yeah, Scott Podsednik, who hit zero home runs during the year. I followed the game on TV and on my laptop, logging the Win Probability impact of each play.

 

How useful is Win Probability? Well, when watching the Cardinals-Padres game, I knew that even when the Padres had the bases loaded with two out in the ninth, trailing 8-5, their Win Probability was only 8%. It was a great moment, a home run would have been huge, but the odds were pretty clear. So I thought I would use it while watching the White Sox attempt to win their first postseason game in 12 years.

 

Win Probability even made it easy to write this article. Why? Because when Pierzynski hit a home run in the first inning to put the Sox up 5-0, Chicago had an 88% Win Probability. I pretty much knew what I was going to write about. The home run and those ESPN announcers.

 

If you've been reading The Hardball Times regularly, you probably know that the White Sox actually out-homered the Red Sox this year. You also hopefully know that US Cellular Field has the highest home run park factor in the league. So how hard was it to predict that this game would turn on the home run?

 

To their credit, ESPN broadcasters Chris Berman, Rick Sutcliffe and Mike Piazza acknowledged Chicago's home run record. But they couldn't admit the obvious. They had to keep saying that Scott Podsednik was critical to Chicago's success, and that the key to the White Sox's offense was small ball.

 

The game started out with the pitcher's mound in the sun, home plate in the shade, and I thought the batters would have a tough time picking up pitches in the first inning or two. Maybe they did; neither Podsednik nor Jermaine Dye could get out of the way of Matt Clement's wild offerings, and both wound up on base after being hit by pitches.

 

After Podsednik reached, manager Ozzie Guillen had Tadahito Iguchi sacrifice him to second base. This is a waste of an out in my book. Iguchi is a fine hitter, but the sacrifice took the bat out of his hands and gave an out to the Red Sox. In fact, the bunt lowered Chicago's Win Probability from .580 to .564. Incredibly, Sutcliffe would later state that this was the key play of the game!

 

The next batter, Paul Konerko, gave an indication of what was to come by hitting a huge fly ball that landed just foul in the left field seats. Podsednik had stolen third and then scored when Konerko hit into a force out. But the steal proved unnecessary as Carl Everett and then Aaron Rowand singled, making the score 2-0. Chicago's Win Probability was already 72% even though the Sox had wasted an out.

 

Pierzynski added his first home run next, making the score 5-0 and the probability of a White Sox win 88%. Jose Contreras was on the mound for Chicago and in late-season form (he was the AL Pitcher of the Month in September), so the game was pretty much never in doubt again. Contreras faltered a bit in the fourth inning but the Pale Hose's Win Probability never fell below 80%. Otherwise, Contreras was in control and the White Sox bats went crazy in the Cell, leading to the 14-2 win.

 

Here's a graph of the Win Probability for the White Sox throughout the game. You can see how quickly it became one-sided:

 

 

It wasn't until late in the game, when Berman yelled "Pitching, Defense and Three-Run Homers" that I felt eyes were finally opening in the booth at the ballpark. Let's see what they say about the Sox offense tomorrow.

 

Pierzynski was, of course, the WPA leader of the game, as you can see from this leaderboard. You'll notice that the Red Sox leader was "No One," because two of the biggest plays for the Red Sox's offense were miscues (an error by Joe Crede, and then a wild pitch by Contreras) which weren't credited to any specific player on the Boston team.

 

Team        Player      Off Pitch Field  WPA

White Sox Pierzynski 0.183 0.000 0.000 0.183

            Contreras    0.000 0.107 0.000 0.107

            Rowand    0.087 0.000 0.001 0.088

            Podsednik    0.050 0.000 0.001 0.051

            Uribe        0.048 0.000 0.000 0.048

            Konerko      0.039 0.000 0.006 0.044

            Everett      0.013 0.000 0.000 0.013

            Dye          0.010 0.000  -0.001 0.008

            Politte      0.000 0.001 0.000 0.001

            Cotts        0.000 0.000 0.000 0.000

            Harris    0.000 0.000 0.000 0.000

            Blum      0.000 0.000 0.000 0.000

            Perez        0.000 0.000 0.000 0.000

            Iguchi      -0.021 0.000 0.013  -0.007

            Crede    -0.020 0.000  -0.016  -0.036

White Sox Total          0.390 0.108 0.002 0.500

 

Red Sox  No one    0.029 0.000 0.000 0.029

            Millar    0.028 0.000 0.000 0.028

            Nixon        0.026 0.000 0.001 0.027

            Bradford  0.000 0.008 0.000 0.008

            Graffanino 0.005 0.000 0.001 0.006

            Renteria  0.006 0.000  -0.001 0.005

            Arroyo    0.000 0.000 0.000 0.000

            Olerud    0.000 0.000 0.000 0.000

            Varitek      0.001 0.000  -0.004  -0.003

            Gonzalez  0.000  -0.008 0.000  -0.008

            Ortiz    -0.028 0.000 0.000  -0.028

            Mueller  -0.053 0.000  -0.001  -0.054

            Ramirez  -0.060 0.000 0.001  -0.059

            Damon    -0.064 0.000 0.001  -0.063

            Clement      0.000  -0.387 0.000  -0.387

Red Sox Total        -0.110  -0.387  -0.003  -0.500

 

The ESPN announcers gave their Player of the Game award to Jose Contreras, and I have to say that I actually agree with them on that call. This game was a good example of how WPA is useful for tracking a game but not as useful for measuring player contributions.

 

Pierzynski had the biggest WPA impact by homering early in the game, but that home run was only one of five. If he hadn't put the Sox up early, someone else would have. Contreras, on the other hand, was dominant on the mound and deserved the ESPN recognition. Let's hope they recognize a few other things before this series is over.

[Kyyle23]

That started to look like a Juggernaut post toward the end there

[southsider17]

I LUUUUUUUV the Hardball Times!

 

[Lefty Of Love]

That was a good write up, but when he writes that Pods' steal of third was unnecessary or that Iguchi's bunt was a waste, he fails to recognize that each of those events was another contributor to Clement's meltdown. For instance, Iguchi's bunt closed the possibility of a double play erasing Podsednik, which mighta bailed Clement out of that first and maybe even the game then. It wasn't hte key play of the game, but it was a good one I thought.

 

Also, he wrote that if AJ hadn't put the Sox up, someone else would've....at that point in the game, somebody else had - we were up 2-0.

 

[rangercal]

That was a good write up, but when he writes that Pods' steal of third was unnecessary or that Iguchi's bunt was a waste, he fails to recognize that each of those events was another contributor to Clement's meltdown.  For instance, Iguchi's bunt closed the possibility of a double play erasing Podsednik, which mighta bailed Clement out of that first and maybe even the game then.  It wasn't hte key play of the game, but it was a good one I thought.

 

Also, he wrote that if AJ hadn't put the Sox up, someone else would've....at that point in the game, somebody else had - we were up 2-0.

 

       

^^^^^^^^^^^^^^^^^^^

[bighurt2719]

that graph up there doesnt make sense. how can you graph the probablilty of a win? anything can happen in baseball. i'd like to see the graph from the steve bartman game. WHo could havbe ever predicted what happened after that foul ball? that graph is worthless.

[southsider17]

Yeah, they never get into quantifying psychological or emotional states. Not that they could if they wanted to. So they completely ignore it.

[southsider17]

that graph up there doesnt make sense. how can you graph the probablilty of a win? anything can happen in baseball. i'd like to see the graph from the steve bartman game. WHo could havbe ever predicted what happened after that foul ball? that graph is worthless.

 

Don't be so shallow. It's called "probability" not "prediction". Given a game situation you can most definitely figure a "probable" outcome. And they call it probable because it's not certain.

Edited October 5, 2005 by southsider17

[Kyyle23]

I wonder if the graph starts about .51% because of home field advantage?

[santo=dorf]

at this analysis.

 

Lefty of Love nailed it on the head. Baseball is not a sequence of random events.

[Controlled Chaos]

that graph up there doesnt make sense. how can you graph the probablilty of a win? anything can happen in baseball. i'd like to see the graph from the steve bartman game. WHo could havbe ever predicted what happened after that foul ball? that graph is worthless.

 

So do like writers use those so they can start writing articles on a game before they are over??

[Flash Tizzle]

  at this analysis.

 

Lefty of Love nailed it on the head.  Baseball is not a sequence of random events.

 

I commonly read Hardball Times and Baseball Prospectus for insight--and frequently a good laugh.

 

I wonder how their win-probabilty scale would have predicted Boston's 0-3 comeback last year? Or Florida's improbable rally off Mark Prior in 2003 NLCS?

 

Honestly, I have no problem with any sabermetric system used to evaluate baseball players/teams. This article in particular I give Hardball Times credit for their critique of "smallball." It's when statistics are used for predicting success that I cringe. Such as saying, "Well Buehlre's statistics have declined the last three years--it's OBVIOUSLY going to continue." Or, "His OBP in the minors was .405--it's OBVIOUSLY going to translate well, although not exactly, into the majors."

 

I'd like to know what tangibles/intangibles are added to Win-Probability to assemble their percentages? Does it take into account the park-factor, pitcher on the mound, weather conditions, offensive capability of opposing team, errors, William Ligue? It's completely unneccesary to calculate win-probability. What purpose does it serve other than statisticians and Beane-lovin' mongers patting themselves on the back for developing a system with a 87.933243% probability of being correct.

 

A 10 year old kid could have told you the second best pitching staff in the American league had a high likelihood of winning with a 5-0 lead in the first.

Edited October 5, 2005 by Flash Tizzle

[santo=dorf]

QUOTE(Flash Tizzle @ Oct 5, 2005 -> 11:17 AM)

I commonly read Hardball Times and Baseball Prospectus for insight--and frequently a good laugh.

 

I wonder how their win-probabilty scale would have predicted Boston's 0-3 comeback last year? Or Florida's improbable rally off Mark Prior in 2003 NLCS?

 

<{POST_SNAPBACK}>

I recall seeing a Red Sox poster on another site post the BP "calculated" odds of the Red Sox winning the ALCS being down 3-0 at 8% and winning the WS 1.95%.

 

I'll try and find it.

[CanOfCorn]

I believe someone, I think it was Rob Neyer saying that Pods, even though we know he is a huge key to the White Sox season, isn't good because his OPS is around .700.

 

Well, KC Rob, it looks as though he was 2 for 3 with his first home run (and not a cheapy, either). Oh yeah, and an HBP AND a stolen base AND getting into Clement's head.

 

Stats mean nothing in the post-season. It's a whole new 8-team league...first one to 11 wins...wins.

 

And I don't understand how you can say that IF Iguchi didn't bunt and "waste" an out, that Everett and Rowand would have even gotten up to bat to drive in the runs that this guy said they were going to drive in anyway. What a crock.

 

But that's what I love about baseball...lots of cooks and no spoiling of the pot.

[Steve9347]

Wow, that article is pointless. That guy with his win probability can stick it up his ass. Anyone can tell you that the team that is up 5 runs in the first has a better chance to win, and that with two outs in the bottom of the ninth, the team up 3 has a better chance. I like baseball statisticians as much as the next guy, but give me a break.

[bighurt2719]

Wow, that article is pointless.  That guy with his win probability can stick it up his ass.  Anyone can tell you that the team that is up 5 runs in the first has a better chance to win, and that with two outs in the bottom of the ninth, the team up 3 has a better chance.  I like baseball statisticians as much as the next guy, but give me a break.

 

that's what i'm saying, and i got called "shallow."

[southsider17]

Stats mean nothing in the post-season.  It's a whole new 8-team league...first one to 11 wins...wins.

 

 

That's why Billy Beane said "My s*** don't work in the playoffs."

 

For stats to be useful (and as predictors they are still, at best, an educated guess) they have to be applied over a long period of time. But over a 5 or 7 game series, any level of performance is possible. Stats will just tell you what's likely.

 

The win-probability graph is pretty useless, it's just a way to graphically look at momentum or critical points in a game. When they first came up with that that's all they claimed.

 

As for the "shallow" post, I believe you can graph the probability of a win throughout a game, contrary to what you said. They simply use historically similar situations and player/team performance stats and make some kind of calculated 'guess'.

 

Win probability is just fluff. The graph adds some splashy graphics to a summary article. I can't believe all the defensive posts on here.

 

That's the funny part.

[SHAFTR]

Regarding "Bartman" and the Red Sox comeback. The reason those moments are memorable is because they weren't predictable. The Propability of Marlins and Red Sox winning those games was very low, not impossible. I don't see how those games are examples of stats failing, but examples of teams winning against the odds (as the stats say).

 

I'm not against looking at potential, talent, intangibles etc when looking at baseball. On the other hand, people will claim that sabermetrics is useless. All of you use BA, ERA, splits, etc to help predict future possiblities. Even WHIP is a sabermetrically created # that is used. Stats simply point to the likelihood of something, not what IS going to happen. It is nothing to be scared of.