AquaMan2342 wrote:So you are saying you'd rather have 100 games of Bradley and 62 out of random fan out of section 10 than Adam Dunn? Again...not to mention that there's no way an AL team would ever sign Dunn and not play him at DH the majority of the time. LAA had a poor team OBP last year, and signing a perennial .400+ guy like Dunn significantly helps one of their biggest weaknesses. Taking Dunn at face value, I still see him as a better signing than Bradley/section 10 and that's without the factors of health and consistency.
Random section 10 fan is not replacement level. Replacement level is freely available talent, think a AAAA guy that can be had by any team at any point. If you don't have one of your own to call up, you can deal a D- prospect for one.
Where do you get the consistency argument from? HRs? Did you realize that Dunn was almost twice as productive (in relation to the average player) offensively in 2007 than he was in 2008? In fact, on a per PA basis, Bradley and Dunn were almost the same player offensively. Then Bradley blew Dunn out of the water in 2008.
AquaMan2342 wrote:Anyways, I still think it's funny that you click some buttons, read some articles and charts, act like nobody else is capable of comprehending what they are saying, make the claims that you do and then criticize other people for not agreeing with you. My question to you is whether or not you can prove how tangible UZR, UZR/150 and these other defensive metrics really are. You are naive at best to think that I, or anybody else who doesn't agree with you, completely neglects the value of defense when looking at a certain player. I am indeed guilty of probably not taking into account enough, but if anything you seem to be (at least that's how it comes off) of doing the exact opposite and overestimating it.
I'm pretty sure that I never once have said that no one else is able of comprehending this stuff, rather they either discount most of it or disregard it altogether. There is no way of "proving" UZR or PMR, it's an impossible task. It's a counter-factual analysis. One side of the equation is always going to be "What would have happened if...?" and you can't prove it. The systems all use batted ball data (position and how well hit the ball is) to determine the expected number of outs that an average player would have made given his distribution of batted balls. Then, his actual number of outs is compared to that number. All that's left is a simple out to run conversion and you are there.
The thing that I don't understand is you make this statement like I think I'm some guru because I've read articles and look at the numbers. That couldn't be further from the truth. It's very similar (in fact, the same!) to how when we try to evaluate a player offensively, we do it based on the numbers. I can say with confidence that you don't prefer Dunn over Bradley because you've watched each of their at bats over the past 5 years, you've looked at some numbers (I'll probably argue the wrong ones) and have come to a conclusion. I have to come to the conclusions that I have because I've done a lot of research on the systems and trust the brightest analytical minds in the business. Certainly there are smart guys in front offices that have great minds, but as far as publicly available analysis this is the best of the best. I'm not making up numbers in my basement while scrolling through play by play readouts.
In any case, there are only two aspects to a position player's value - how many runs he creates and how many runs he prevents. They are equal in value and completely additive. You may say the offensive value is easy to figure, but that's not true at all. OPS may be better than BA, but are all .800 OPS hitters created equally? No. Can a .940 OPS guy be more productive than a .965 OPS guy? Sure! OBP is a touch more valuable in run creation than slugging. Run creation has gotten much more quantifiable with wOBA, because it directly correlates to run scoring. An single is worth A runs, a double B runs, a triple C runs, etc, etc, etc. Only using wOBA and then an idea of what a player does defensively (and a positional adjustment and an adjustment for average vs. replacement but we don't need to get into those) can you get an idea of how many runs a player is worth.
There's no "overestimating" about taking defense into account. A run saved is as valuable
as a run scored. (Actually it's a little more valuable [perhaps up to 10%] simply because the differentials are more powerful when the overall runs you're talking about are smaller but we'll leave that part out of it.) You may feel like I'm overestimating the idea of defense simply because I'm one of the few who will actually talk about or quantify it around here, but that doesn't make it any less of a part of what a player gives a team. And I don't think
that people are neglecting defense, I know from their very arguments around how guys are good signings or good values that they are.