You can't possibly think Soriano's defense is good. UZR may say so, but that is a formula that has flaws just like an eye. UZR is unreliable when used alone.
Not true. If you use a one year sample size of UZR it will more than likely be inaccurate. However, if you use 3 years worth, it becomes reliable.
You obviously don't remember, so allow me to refresh your mind...
rARM, which is number of outfield arms runs saved: 5 years: 34(next is Willingham with 16) 4 years: 24(Willingham again in 2nd with 12) 3 years: 7(2nd to only Crawford and Ibanez who have 8). Pretty freakin' great.
rPM, which is plus or minus the number of runs saved: 3 years: - 10. Which may seem bad, but it's 4th best. 4 years: -11(again 4th). 5 years: 0(3rd best, behind Holliday and Carl.) While these numbers aren't great, they aren't bad. Which is the point I'm trying to get across to you.
DRS, or defensive runs saved. Over 5 years, he is 2nd to only Crawford. 4 years, 3rd to Holliday and Crawford. Same for 3 years.
RZR, revised zone rating, he is in the middle of the pack for each of the 3 year increments I am using.
TZL, total zone with location, is basically another form of UZR. 5 years: Soriano was great with a 23.6 TZL, 2nd to only Mr. Crawford. 4 years: 3rd to Carl and Holliday, with a 2.3, which is still considered average. 3 years: -2.6, which is still average, but it was good enough for 2nd.
TZ, total zone without location has Soriano in the top 3 year each time, too.
This, coupled with high UZR totals is what I have based my opinion on. I don't see how the human eye can be right and all these defensive stats can be wrong.