(10-27-2017 01:31 PM)Seahawk Nation 08 Wrote:  

(10-27-2017 01:20 PM)bricksnivy Wrote:  

(10-27-2017 12:06 PM)Seahawk Nation 08 Wrote:  I agree that analytics should only be used to a point. But the Dodgers are a heavy analytics team and are in the World Series, so clearly it works for them the overwhelming majority of the time.

Note that I wasn't ripping you guys, I was talking about the so-called baseball analysts. I'm not sure who all the color guys are for the series, but after Roberts pulled Hill, one of them was just like "I just don't get it. I get that's what the numbers say but I don't agree with it." Well, analytics are part of the furniture of the game. It's your job as an analyst, especially in a World Series game, to try to understand it. Yet you have these guys acting like when you hand your Mom a TV remote that is more complex than the one they were used to growing up, and they shake their heads like "No, I don't want to try to figure it out."

The one area where analytics has been wrong has been for top-flight hitters. Take Vlad Guerrero for instance. Analytics guys don't like Vlad as much as other great hitters because they think he should have taken more walks. But when you have a transcendent hitter, the rules don't apply to him. You let Vlad swing the bat. Vlad could take bad pitches way out of the zone and turn them into doubles and homers. He's a bonafide first ballot Hall of Famer.

Taking a lot of walks can take so-so hitters and turn them into very productive players. But it doesn't take great hitters and make them better. It actually is counter-productive.

But at the same time, while some people bemoan that analytics has taken a lot of the decisions out of the hands of the managers, many times that's a good thing. Some guys, like Dusty Baker, for instance, try to "over-manage" at times and it fails time and again. Those dinosaur types of managers just aren't useful in today's game.

I don't know if I agree with everything you said, but I generally agree with this summary. Maddon references this as outcome bias, and I agree that you go with the approach that got you there, win or lose.

An example of where I want to see an analytics-minded manager incorporate his baseball intuition is "manufacturing runs". I forget the opponent, but here was Cubs game in late July or early August that was 0-0 going into the 8th. Cubs get their lead off guy on second base with no one out. 1 run wins this game, and they didn't sacrifice because "you don't give away outs in today's baseball." The Cubs were in a race where the division lead is less than 5 games, and they lose because a computer model says they have a better chance of scoring multiple runs by swinging. That's when I long for old-fashioned baseball.

But 1 run wouldn't have won the game. As you stated, it was the 8th inning. If it had been the 9th, the bunt would have been the proper call.

yes, but that's where common sense applies. When you have Wade Davis and it's the 8th inning, you score 1 run the likelihood of winning the game becomes significant. This is the very thing that drove me crazy about Girardi, always waiting for the HR. Teams shifted every single player on the yanks because no one will bunt. They have two of the better base stealers in the game in Ellsbury and Gardner, but yet they never steal either!