17 yo Trayvon Martin Shot to Death by Neighborhood Watch Captain #13

[dsntslp]

Somebody had a nice map last thread, I'll see if I can find it.

ETA: NM, I see you guys are using it, I think.

Here it is:
[ame="http://www.websleuths.com/forums/showpost.php?p=7743501&postcount=641"]Websleuths Crime Sleuthing Community - View Single Post - FL 17 y/o Trayvon Martin Shot to Death by Neighborhood Watch Captain #12[/ame]

 

[TorisMom003]

I believe most have agreed that Zimmerman would have been coming back from the North entrance, heading South. He apparently saw Martin at the clubhouse, and parked down past the clubhouse and watched him. Martin passed Zimmerman, and then the entire running/following thing apparently happened.

But Zimmerman's truck is not near the mailboxes. At least from what I have seen on the map it wasn't.

 

[Donjeta]

I mean that the expert considers 90% a match, but when the person demonstrating compared two recordings of the same person (Nixon), it only displayed an 86% match. I guess I'm not understanding how he expects to get over 90% when two very similar recordings were compared and only came out as 86%.

MOO and I am not an expert so I could be wrong.

I think there is always room for inconclusive results in this sort of thing because there probably is some overlap between the results of non-matches and matches.

Suppose you have tested thousands and thousands of samples and have seen consistent patterns.

Suppose you take any two random male voices and they usually get about 40-85 % result.

Suppose you take any two matching voices and they usually get about 80-98% result.

Suppose you and your colleagues have never seen a matching sample that goes below 74%.

Further suppose you've never seen a non-matching sample score over 88%

Now, if you look at this pattern and see that your test scored 48 %. You can be pretty darn sure they're not a match.
Look at another test that scored 95. You can be very sure they're a match.

But if you get a score between 74-88 you just can't say anything for certain and would have to conclude that your test is inconclusive because you've seen those scores go either way.

Add a couple of percent to be sure you eliminate false positives and you come up with the 90 % threshold, knowing that some 86 % Nixons will go inconclusive.

I pulled the percentages out of my hat so don't anybody quote me on this, they're just an illustration and nothing to do with the actual test.

 

[TorisMom003]

According to what I've listened to on the 911, He parked his truck and then got out.

Yes he did. However, I am trying to determine if he followed Trayvon while in his truck first, then parked and followed Trayvon on foot.

 

[GrandmaTo4]

Just got around to reading my morning copy of the Orlando Sentinel. Editorial there on the fact that it has been 4 months since the FAMU hazing death and no arrest there either....My but the wheels of justice sure turn slowly.

Those wheels certainly can be slow, can't they? Sometimes they don't turn at all.

Gee, what race was Robert Champion, (the young man who was killed) again?

I'm sure some have forgotten and others will profess coincidence.

imo

 

[m00c0w]

But Zimmerman's truck is not near the mailboxes. At least from what I have seen on the map it wasn't.

Up until the follow/chase/running episode, it has been thought to have been somewhere near that row of houses on the "cut-through." Whether he moved the vehicle when Martin allegedly started running, I don't think anyone knows for certain. His vehicle, by most accounts, would not be up by the clubhouse.

 

[katydid23]

All this talk about riots, it's almost like wishing for it to happen? I live here in Florida and I am not even remotely afraid of rioting? I've watched the rally's and I see thousands of people of all races standing together for one cause. Justice for Trayvon.

I live in Southern California and I have seen first hand what can happen if some people do not like certain jury verdicts.

 

[OneLove]

And don't forget that when GZ first called 911, Trayvon was NOT facing him. He said after the call was in progress that the suspicious guy was NOW facing him, then walking toward him, hand tucked in waistline.

 

[m00c0w]

Those wheels certainly can be slow, can't they? Sometimes they don't turn at all.

Gee, what race was Robert Champion, (the young man who was killed) again?

I'm sure some have forgotten and others will profess coincidence.

imo

http://edition.cnn.com/2012/03/27/justice/florida-hazing-death/

Investigations take time.

 

[m00c0w]

And don't forget that when GZ first called 911, Trayvon was NOT facing him. He said after the call was in progress that the suspicious guy was NOW facing him, then walking toward him, hand tucked in waistline.

He never said anything about "facing". He made mentions of Marting looking at him, checking him out, and coming towards him.

 

[ynotdivein]

WS public service announcement: Did you know that this case has a Media thread? When a new MSM article comes out, make it a habit to link it there first, for future reference.

Here's a link to the thread: [ame="http://www.websleuths.com/forums/showthread.php?t=166507"]Trayvon Martin Shooting Media Thread NO DISCUSSION[/ame]

This thread is located in the Stickies for the Hot Cases forum.

Please note, this is a no-discussion thread. MSM links only. This thread will be invaluable to us all if this goes to a trial. Thanks to all who have already started to populate it!

We now return you to your previously scheduled conversation.

 

[TonyGatto]

Anyone have that photo of the area and can bump up. Kinda necessary for this discusson. I promise I'll boomark it this time.

 

[m00c0w]

MOO and I am not an expert so I could be wrong.

I think there is always room for inconclusive results in this sort of thing because there probably is some overlap between the results of non-matches and matches.

Suppose you have tested thousands and thousands of samples and have seen consistent patterns.

Suppose you take any two random male voices and they usually get about 40-85 % result.

Suppose you take any two matching voices and they usually get about 80-98% result.

Suppose you and your colleagues have never seen a matching sample that goes below 74%.

Further suppose you've never seen a non-matching sample score over 88%

Now, if you look at this pattern and see that your test scored 48 %. You can be pretty darn sure they're not a match.
Look at another test that scored 95. You can be very sure they're a match.

But if you get a score between 74-88 you just can't say anything for certain and would have to conclude that your test is inconclusive because you've seen those scores go either way.

Add a couple of percent to be sure you eliminate false positives and you come up with the 90 % threshold, knowing that some 86 % Nixons will go inconclusive.

I pulled the percentages out of my hat so don't anybody quote me on this, they're just an illustration and nothing to do with the actual test.

I get what you're saying. I'd just feel more comfortable with documentation directly related to the test, or results back from Martin's voice. I don't think we'll get that until the FBI releases what they have, though.

 

[Donjeta]

He never said anything about "facing". He made mentions of Marting looking at him, checking him out, and coming towards him.

It kind of implies that TM was facing him imo.

 

[vlpate]

Gee, two acknowledged experts using different methods tell the press that their results are accurate to a degree of scientific certainty. And thus far, we have no experts contradicting them. (That could change in the future, but that's the status quo.)

And in opposition, we have dozens of posters here who, without the benefit of any training or knowledge of voice comparison techniques whatsoever, have decided that a 90% confirmation threshold should be reduced to 40%.

There's plenty of "confirmation bias" (well, actually, that's just the polite term for it) going on here alright.

Anyone can say they are an expert at anything. I have seen nothing that convinces me these two are.

These guys - I would trust these guys over someone who analyzed a Britney Spears tape any day:

http://www.forensictapeanalysis.com/about_us.htm
1. Wal-Mart
2. Motorola Corp.
3. BankOne
4. BlueCross BlueShield
5. Shell Oil Co.
6. United Parcel Service, Inc.
7. Reliant Energy
8. 7-Eleven, Inc.

Doesn't matter what I think - these two "experts" will show up as Bombshell Tonight! pundits on NG's show Monday and many will take their word as gospel.

JMO

 

[uvamerica]

He never said anything about "facing". He made mentions of Marting looking at him, checking him out, and coming towards him.

Well if GZ is looking at TM, and TM is looking at GZ,wouldn't they be facing each other technically ?

 

[dsntslp]

Map
[ame="http://www.websleuths.com/forums/showpost.php?p=7743501&postcount=641"]Websleuths Crime Sleuthing Community - View Single Post - FL 17 y/o Trayvon Martin Shot to Death by Neighborhood Watch Captain #12[/ame]

Thanks "Concerned Papa" for that post.

 

[legalmania]

http://articles.cnn.com/2012-03-16/...president-omar-al-bashir-ahmad-harun?_s=PM:US

Thanks $100 fine. Hope he donates more than that to help the rebuild.

 

[Deleted member 39678]

MOO and I am not an expert so I could be wrong.

I think there is always room for inconclusive results in this sort of thing because there probably is some overlap between the results of non-matches and matches.

Suppose you have tested thousands and thousands of samples and have seen consistent patterns.

Suppose you take any two random male voices and they usually get about 40-85 % result.

Suppose you take any two matching voices and they usually get about 80-98% result.

Suppose you and your colleagues have never seen a matching sample that goes below 74%.

Further suppose you've never seen a non-matching sample score over 88%

Now, if you look at this pattern and see that your test scored 48 %. You can be pretty darn sure they're not a match.
Look at another test that scored 95. You can be very sure they're a match.

But if you get a score between 74-88 you just can't say anything for certain and would have to conclude that your test is inconclusive because you've seen those scores go either way.

Add a couple of percent to be sure you eliminate false positives and you come up with the 90 % threshold, knowing that some 86 % Nixons will go inconclusive.

I pulled the percentages out of my hat so don't anybody quote me on this, they're just an illustration and nothing to do with the actual test.

Your post leaves me even more confused than before I read it.
You have things like 40-85% being found and then 48% means no match and then a score of 74-88 is inconclusive. Sorry I don't get it.

 

[m00c0w]

Well if GZ is looking at TM, and TM is looking at GZ,wouldn't they be facing each other technically ?

Yes, but it doesn't mean that TM was facing away from him at any other time, either. It's not as if Zimmerman said, "Oh, he's facing me now." He could potentially still see Martin's face/front without Martin overtly looking at Zimmerman.