Jump to content

Providing independent coverage of the Minnesota Twins.
Subscribe to Twins Daily Email

Recent Blogs

Photo

A's Trade 3B Donaldson to Toronto

  • Please log in to reply
170 replies to this topic

#161 spycake

spycake

    Senior Member

  • Members
  • 14,829 posts

Posted 23 July 2015 - 07:04 PM

Not trying to be difficult, really.

But overall bad defense is going to affect your overall run differential, just like overall bad hitting or overall bad pitching.

Obviously pythag is jus an estimate, but you can't explain away a large pythag discrepancy by saying bad overall hitting/pitching/defense. It has to be situational, which usually means bullpen, RISP, and/or plain old luck.

So I am saying they can have a bad defense, even a very bad defense, but bad luck may still be the larger factor. Especially if bad defense gave them a bunch of 1 run losses, but did not apparently move the margins much in their other games (which would in total show up in their overall run differential).

#162 TheLeviathan

TheLeviathan

    Twins News Team

  • Members
  • 16,279 posts

Posted 23 July 2015 - 07:12 PM

I didn't explain it all with defense - it is the bullpen and the defense.  Defense, as we Twins fans should be well versed in lately, has a way of creating plenty of "bad luck".

 

Whatever bizarre way you want to say the defense is hurting them, the point is that their defense is awful and much of the cause of that was directly caused by Beane's trades and the players he targeted in return.

 

But I'm done, they have a bad defense and a bad bullpen.  Those are reasons 1A and 1B, whatever order you want to put them in, for their crappy play this year.


#163 spycake

spycake

    Senior Member

  • Members
  • 14,829 posts

Posted 23 July 2015 - 07:12 PM

Put another way: their run differential suggests 53 pythag wins. Simply with better defense, they might have prevented more runs, to suggest, say, 56 pythag wins.

Still doesn't explain the gap between the 53 estimated and 44 actual wins. Unless most of the bad defense was concentrated mostly in close games... which sounds like situational fielding, which in turn sounds like luck.

#164 jimmer

jimmer

    A former AF SNCO who values integrity.

  • Members
  • 10,212 posts

Posted 23 July 2015 - 07:34 PM

Athletics are tied for 11th in DRS (5th in AL) with 11 DRS.  We are tied for 25th (tied for 12th in AL)  with -16 DRS


#165 spycake

spycake

    Senior Member

  • Members
  • 14,829 posts

Posted 23 July 2015 - 09:03 PM

I haven't looked closely at the defensive stats, but I do recall that Lawrie fared much better by DRS than UZR this year, and last I remember Torii Hunter was the reverse...

#166 jimmer

jimmer

    A former AF SNCO who values integrity.

  • Members
  • 10,212 posts

Posted 23 July 2015 - 09:40 PM

 

I haven't looked closely at the defensive stats, but I do recall that Lawrie fared much better by DRS than UZR this year, and last I remember Torii Hunter was the reverse...

Defensive Runs Saved uses a rolling one-year basis for the Plus/Minus system, while UZR uses several years of data to determine each play’s difficulty level. 

 

So DRS uses the most current level of defensive talent at positions to set the plus/minus level and UZR takes into account the talent level over several years to include players who may have retired (or switched positions or are no longer in the game).  

 

It's why I like DRS better.  Why compare a current players defense to a guy who retired two years ago at age 38 who was bringing down the average talent level?

Edited by jimmer, 23 July 2015 - 09:43 PM.


#167 USAFChief

USAFChief

    Anyone got a smoke?

  • Twins Mods
  • 22,341 posts
  • LocationTucson

Posted 23 July 2015 - 09:51 PM

Defensive Runs Saved uses a rolling one-year basis for the Plus/Minus system, while UZR uses several years of data to determine each play’s difficulty level. 
 
So DRS uses the most current level of defensive talent at positions to set the plus/minus level and UZR takes into account the talent level over several years to include players who may have retired (or switched positions or are no longer in the game).  
 
It's why I like DRS better.  Why compare a current players defense to a guy who retired two years ago at age 38 who was bringing down the average talent level?


Most proponents of defensive metrics will say three years of data are needed to get the better picture of ability. Why would using one year as a baseline be better than using several?

Cutting my carbs...with a pizza slicer.


#168 jimmer

jimmer

    A former AF SNCO who values integrity.

  • Members
  • 10,212 posts

Posted 23 July 2015 - 10:09 PM

 

 

 

Most proponents of defensive metrics will say three years of data are needed to get the better picture of ability. Why would using one year as a baseline be better than using several?

I already explained why.  Because it uses the abilities of more current players not players who may have retired, only played one season, have since left the game, moved to other positions, etc.  And I explained why I think that is better. I want to know how a player is doing against players who are more likely still playing the same position.

 

I'm not going to be involved in derailing another post on the validity or application of metrics.  Most moderators have made it clear not to do that, no matter how many times a trolling post tries to turn the discussion that way.

 

I pointed out Oakland's DRS rating because people were saying how bad Oakland's defense was and I showed one stat that we may use to say maybe not so much.  Then I tried to explain why sometimes there's a difference between UZR and DRS. That's it.  People can take it or leave it, doesn't matter to me.  Said my piece and now their debate can continue. But they have to be basing their belief Oakland's defense has been horrendous on something.

Edited by jimmer, 23 July 2015 - 11:03 PM.


#169 Brock Beauchamp

Brock Beauchamp

    Owner

  • Administrators
  • 20,691 posts

Posted 24 July 2015 - 07:44 AM

 

Most proponents of defensive metrics will say three years of data are needed to get the better picture of ability. Why would using one year as a baseline be better than using several?

It's an interesting question but the several years of data required is to evaluate a single player. Over the aggregate, one year should be adequate to rate the current defensive ability of MLB.

 

I think both metrics are flawed but have their place in today's game. They're a good counterpoint to what we watch on a daily basis.


#170 jimmer

jimmer

    A former AF SNCO who values integrity.

  • Members
  • 10,212 posts

Posted 24 July 2015 - 08:00 AM

Brock since all stats, traditional and metric, give us true talent level with three years of data, I'm waiting for them to decide to only give out awards every three years instead of yearly.  MVP, Gold Gloves, Silver Sluggers, Hank Aaron, Cy Young, Fielding Bible awards, etc. Makes sense, no? Since all we care about it true talent level of a player as opposed to what a player did any given year. Pretty much have to stop doing the rookie of the year award completely, because it says year in it :-)

 

Should we do that with that for the manager award too?  Might be hard:-)


#171 drjim

drjim

    Senior Member

  • Members
  • 8,783 posts
  • LocationSt. Paul

Posted 28 July 2015 - 09:34 PM

I'll give Beane credit, he has done very solid work. Didn't mess around, moved the guys that he needed to move, and got solid returns.

Do think this will lead to a little bit of a longer downturn than usual but they are reloading.
Papers...business papers.