Is Dan Duquette thinking of Trumbo as an outfielder?

[thezeroes]

43 minutes ago, Can_of_corn said:

I'll heartily agree that pitch framing includes bias but outfield defense?  How so?

I also think they are using the technology and things will be greatly improved in a few years.

Neil Weinberg did an article on Fan Graphs in 2014 that states that if the Stringer 9the guy/gal coding a game) places an out as being hit into a shift, that out is not part of a defensive metric.  Hence an out is not recorded to a particular fielders ledger for that purpose and skews the sample size.  The Orioles use a shift defense in the outfield for most plays, particularly at home so therefore one would think that the defensive metrics are skewed for ALL Oriole Outfielders.

Question asked:

"What if a ball is missed because of defensive positioning other than a shift? Say, the infield is playing in at the edge of the grass with a runner on 3B, and a ball gets through the big hole in the middle, or goes just over someone’s head? Or a shallow single against an outfielder playing ‘no dobules’ deep?

That’s not a ‘shift’, but it’s not the player’s choice to be playing too far in or out. Does that play get thrown out by the stringer, or does the fielder just take the hit on his range score?"

Answer:

 

"Neil Weinberg -It’s my understanding the fielder will get docked there. Although I don’t know the technical definition of “Shift” used to determine if the play gets thrown out."

[weams]

3 minutes ago, Can_of_corn said:

Do they?  I thought they had moved beyond that by now.

I'm sure the teams have.

The teams have. We do not have access to their data. 

[weams]

3 minutes ago, thezeroes said:

Neil Weinberg did an article on Fan Graphs in 2014 that states that if the Stringer 9the guy/gal coding a game) places an out as being hit into a shift, that out is not part of a defensive metric.  Hence an out is not recorded to a particular fielders ledger for that purpose and skews the sample size.  The Orioles use a shift defense in the outfield for most plays, particularly at home so therefore one would think that the defensive metrics are skewed for ALL Oriole Outfielders.

Question asked:

"What if a ball is missed because of defensive positioning other than a shift? Say, the infield is playing in at the edge of the grass with a runner on 3B, and a ball gets through the big hole in the middle, or goes just over someone’s head? Or a shallow single against an outfielder playing ‘no dobules’ deep?

That’s not a ‘shift’, but it’s not the player’s choice to be playing too far in or out. Does that play get thrown out by the stringer, or does the fielder just take the hit on his range score?"

Answer:

 

"Neil Weinberg -It’s my understanding the fielder will get docked there. Although I don’t know the technical definition of “Shift” used to determine if the play gets thrown out."

Thank you.  Bad data in...

[Can_of_corn]

1 minute ago, weams said:

The teams have. We do not have access to their data. 

The groups running the metrics have access to Statcast right?  Or does MLB greatly limit what they release?

[thezeroes]

26 minutes ago, Can_of_corn said:

The groups running the metrics have access to Statcast right?  Or does MLB greatly limit what they release?

From what I can find, the outgoing DATA is limited.  The other issue with a reliance on just STATCAST is it does not finish the product because of no league adjustment, park adjustment, batter handedness adjustment , pitcher handiness adjustment or weather adjustment.  At some point all of the elements that are available mat merge into on defensive metric but as of now the data is flawed.

https://baseballwithr.wordpress.com/2016/11/07/statcast-defensive-gam-2015-2016/

[Number5]

56 minutes ago, Can_of_corn said:

I'll heartily agree that pitch framing includes bias but outfield defense?  How so?

I also think they are using the technology and things will be greatly improved in a few years.

Maybe you can explain how they account for any difference in the speed of the ball off the bat from one ball that drops in a certain spot to another that is caught there.  How about differences in spin?  Trajectory?  Wind?  Positioning?  Game situation?  Having one ball caught by an outfielder and another hit to virtually the same spot drop in by no means proves to me that the two chances were identical.  There are any number of forces causing a difference in the degree of difficulty and/or wisdom in making a particular play.  Anyone that has ever picked up a glove knows this.  Yes, attempting to apply a single standard to cover this multitude of complex and different situations contains inherent bias, based upon the opinion of the non-baseball professionals making these decisions on how plays should be applied to the metrics.

How about the fact that runners and coaches have respect for a player's arm and hold up at a base.  No throw, no play, but a plus defensive result nonetheless.  How would the metrics determine whether this was out of respect for the outfielder's arm, or due to a slow start by the runner, or due to a bad coaching decision, or because the offensive team was already leading the game 17-2?  How about hitting the cutoff man perfectly or not?  How about positioning based on coaching, rather than player choice?  Game-planning how a batter will be pitched combined with pitchers hitting/missing their spots?  Again, there is often no proper accounting for these situations in the metrics.  How about having the intelligence to know when a chance should be taken on a ball, and when to play it safe, given the game situations involved?  Players, coaches, managers, scouts, and general managers all know and appreciate these things.  Defensive metrics can be a useful tool, in a general way, but are very far from a true, accurate, and absolute measure of defensive ability, in the opinion of those actually involved in the game.  Many folks give too much weight and credence to the metrics, IMO.  I agree with the players, coaches, managers, scouts, and general managers who see these things and have clearly stated that they know the metrics to be imperfect, and sometimes even misleading.

How do you think the metrics would treat this:

Yep, it is a play that other outfielders catch.  Normally, it is a ball that should be caught.  Letting it drop in this game situation, however, was an excellent play.

[Can_of_corn]

5 minutes ago, Number5 said:

Maybe you can explain how they account for any difference in the speed of the ball off the bat from one ball that drops in a certain spot to another that is caught there.  How about differences in spin?  Trajectory?  Wind?  Positioning?  Game situation? 

 

Statcast tracks spin and trajectory (launch angle).  Wind has been a known quantity for some time.  Positioning is tracked as is reaction time, speed and elapsed time.

Statcast provides an enormous amount of data.

I was mistaken in how much of that data is being shared.

[Number5]

1 minute ago, Can_of_corn said:

Statcast tracks spin and trajectory (launch angle).  Wind has been a known quantity for some time.  Positioning is tracked as is reaction time, speed and elapsed time.

Statcast provides an enormous amount of data.

I was mistaken in how much of that data is being shared.

If you believe that speed off the bat, spin, trajectory, and wind speed and variance is measured exactly by anyone, I have a bridge I can sell you.

[Can_of_corn]

4 minutes ago, Number5 said:

If you believe that speed off the bat, spin, trajectory, and wind speed and variance is measured exactly by anyone, I have a bridge I can sell you.

So you don't believe in Statcast?

[weams]

16 minutes ago, Can_of_corn said:

So you don't believe in Statcast?

I believe in statcast. I dont believe any anything humanly charted and arbritation selected. 

[thezeroes]

59 minutes ago, Number5 said:

If you believe that speed off the bat, spin, trajectory, and wind speed and variance is measured exactly by anyone, I have a bridge I can sell you.

Then you also do not believe in the Pitch f/x??? This is a system in place that measures and tracks the velocity, movement, release point, spin, and pitch location for every pitch thrown in baseball

[Can_of_corn]

44 minutes ago, weams said:

I believe in statcast. I dont believe any anything humanly charted and arbritation selected. 

Obviously I thought baseball was sharing more information than they currently are.

[Number5]

1 minute ago, Can_of_corn said:

So you don't believe in Statcast?

I believe they attempt to account for some of these things, but it really just isn't possible.  In the attempt to do so, the door for bias is opened a bit further, IMO.  How much do they take these factors into account on a play by play and player by player basis?  Wind speed, direction, and variance, air pressure and temperature can change second by second.  A cloud moves in front of the sun, or away from it.  The wind gusts.  The climate control in a dome clicks on or off.  Heck, wind shear downs airplanes, for crying out loud.  And a lot of time, money, effort, and technology has gone into trying to eliminate that type of tragedy.  Trust me, if you ever had a real strong man hit a line drive at you at third base so hard that it was knuckling a foot from side to side and up and down all in a split-second, you would know that all line drives hit at you are identical.  No, I do not think Statcast, or any other earthly entity, has full accounting and control over these things. 

I also know that they do not account for the game situations, coaching decisions, and baseball intelligence issues I addressed.  They simply don't.  The very fact that people somehow believe these unaccountable factors are being accounted for proves weams' point that there is inherent bias in the metrics, IMO.  Yes, the metrics are better than they were six years ago, and I agree that they will likely continue to improve, but I don't believe they can ever be totally accurate, and certain players will tend to benefit from these imperfections more than others.  May even be different players from one year to the next.  Heck, all you have to do is actually look at what the metrics show for various players one year to the next.  These players don't really wildly fluctuate in their defensive ability from year to year, unless they suffer injury, yet the metrics often indicate that they do.  Adam Jones' total runs above average per 1200 innings, for example, has gone 18, 3, 5, -15, 11, -13, -2, 16, and 0 in his years with the Orioles.  This is by no means unique to Adam Jones. 

I've always enjoyed statistics, as well, but I try to keep them in the proper perspective and recognize both what they are showing me, and what they really aren't.

[Number5]

7 minutes ago, thezeroes said:

Then you also do not believe in the Pitch f/x??? This is a system in place that measures and tracks the velocity, movement, release point, spin, and pitch location for every pitch thrown in baseball

Wasn't talking about that.  At all.

[thezeroes]

Just now, Number5 said:

Wasn't talking about that.  At all.

Same system is used for both.  A person can not believe in one without at the very least acknowledging the other.