defensive metrics?
by travdog6 on Jan 26, 2009 1:44 AM PST
74 comments
this has been discussed to some extent here and somewhat here, but after reading an article that uses plays per game (wtf?) as a defensive metric, ive decided i still don't know defensive metrics/decided to make a fanpost on it.
let me sum up what i know about defensive metrics: i don't know shit. i've read that metrics are unreliable and inconclusive and there's no way to tell with any precision how good a certain player is defensively. which makes me wonder: how do we know that mark ellis is a premium player because of his defense?
so AN, i ask you: which defensive metric (if any) do i trust? i want all the information you have on the mysterious topic that is defense.
0 recs |
74 comments
|
Email
Print
Comments
The best answer is that there are some better than others
None perfect, and it doesn’t hurt to take a look at all of the ones that have some amount of respect in the community and do a bit of an aggregate.
It’s possible that they’re all wrong at Ellis is actually terrible on defense, I guess, but it’s just not very probable.
right
i guess what im wondering is:
ellis isnt very good offensively, but his defense (presumably) more than makes up for this. most around AN refer to him as a premium player. my question is how do we decide his defense is that good?
"True fact: In a global thermonuclear war, the only human who would survive would be David Eckstein" -PT
step 1 – observe his play
step 2 – look at statistics that use raw data like fielding percentage
step 3 – look at metrics that attempt to rate a player like UZR and PMR
Understand the shortcomings of all three. Cogitate away. Talking about those can produce long and contentious threads.
I think hit f/x will produce some good raw data that will give a more accurate picture.
I think Step 2 is pretty close to unnecessary
Just because those stats are so, so meaningless that you literally learn nothing useful from them.
My problem with Step 1 is that most people primarily watch one guy at a position – the guy on their favorite team. So even if you think Ellis is making every routine play and a bunch of tough ones and he never makes an error you have no idea how good he actually is because you haven’t seen the other guys at his position enough. What if every routine play and a bunch of tough ones is what most of the guys do?
Your last line is the best one. Hit F/X will absolutely make the metrics better. UZR is pretty good now. It’ll be much, much better when we’re not relying on PBP data.
Like I said they all have flaws
I don’t think 2 provides no information and 3 has the potential to provide bad information. The best thing about UZR is that it is easy to get. Other than that what makes it pretty good now?
You really think Fielding Percentage provides more useful information than UZR?
Care to explain?
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 26, 2009 12:48 PM PST up reply actions
different information
Here is an article that is not much more than a review of some raw statistics. I don’t think there is a definitive or broad conclusion that can be drawn from the article.
I haven’t kept up with the different flavors of UZR, but I think it has a lot of flaws and they can be fatal to an evaluation of a player.
The link was to an article on OF throwing arms. Was that intentional?
What flaws does UZR have other than potential bias of the observer?
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 27, 2009 11:30 PM PST up reply actions
sure
The link is basically an accumulation of raw data over a season.
Is the question about UZR rhetorical? You seem like a pretty well informed person and the discussion of flaws of defensive metrics is pretty worked over. It also typically leads to discussions that aren’t very fruitful.
UZR was the best metric when it was impossible to get
It’s a system that uses PBP data and determines how many plays a player makes vs. how many he should make, basically. It’s far better than fielding percentage.
It’s theoretically possible to stand in one spot, never move, make a few plays and no errors and look like a good fielder based on fielding percentage. Fielding percentage really doesn’t tell you anything useful. It’s far, far worse than a stat like batting average or ERA.
You have to make a decision on player acquisition. Whether the metrics are
“reliable” or not they’re the best we got, along with the scouting reports. If they agree with each other you feel a bit more comfortable with your viewpoint. It’s all evidence.
I trust in order:
1) Fans Scouting Report
2) Other available scouting reports (BA, Scout.com, etc)
3) bUZR and sUZR (if available)
4) Plus/Minus from Dewan/James
It’s not a question of “trusting” or “not trusting”. It’s a question of making a decision on whatever evidence you have.
It's not the results, it's how you look going about those results -- Tim McCarver
My list differs in order of weight, but I think that's the best approach.
Research the metrics, find the ones you like the most (I think UZR and UZR/150 are easily the best, but that’s me) and weigh them against scouting reports and other metrics. Defense is hard to quantify, but it’s easier than it used to be and I think that UZR gives a pretty good idea of a player’s worth. As far as value going forward, that’s where the Fan’s Scouting Report and those types of things come into play.
by Aaron Campeau on Jan 26, 2009 3:59 AM PST up reply actions
Yes for historical performance UZR is what it is....although the differences btw sUZR and bUZR
are a bit of a concern.
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 26, 2009 6:46 AM PST up reply actions
I'm sceptical about Fan Scouting Reports
Big believer in pro reports, however.
Travdog6, the rule I’d preach is to question any defensive metric that doesn’t jive with Mark 1 Eyeball reports. Question, not dismiss. Defensive metrics, at their best, quantify how good (or bad) the best (and worst) defenders are.
Check out this site for a quick and dirty way of weighing defensive ability. Sean gives you 5 options to choose from on defense (Excellent, Good, Average, Poor, Putrid) just take whichever rating jives with what you’ve seen/heard about a defender and there you go… you’ve factored defense into a player’s overall contribution to the team,
The monster at the end of this blog.
I don't care for the fans scouting report much
I like the idea of it. I just don’t trust fans (yet) to honestly and accurately fill it out without a bias.
intentional or unintentional
As your post above (I think, accurately) pointed out, most fans only watch “their guy” with any regularity — so even if they’re being fair (and this is all leaving aside issues of “ballot-stuffing”), they may not be being accurate.
Similar reason is why I don’t trust the “advanced” defensive stats — they’re still all based on eyewitness raw data.
A B -3X = Swedish girls like chocolate @('.')@
On what do you base you views on defensive performance and ability?
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 26, 2009 12:49 PM PST up reply actions
I pretty much throw my hands up and walk away
If I think I see something (Cust improving marginally over the course of a season, Crosby still being pretty good at moving to his left and getting the runner at first), I’ll express it. Otherwise, I pretty much take the collective/triangulated words of andeux, devo, grover, PT, and salb918. And when they don’t agree, I throw my hands up and walk away again …
A B -3X = Swedish girls like chocolate @('.')@
I think the difference is that while there might be a margin of error in using the PBP data
There isn’t an inherent bias*. What i mean is that maybe a ball wasn’t exactly where the scorer said it was, but the human error is random (and knowing how difficult it is to get one of those jobs, because I tried), and not very frequent. It’s not error based on confirmation bias or something. It’s not somebody thinking a guy is good because he made a lot of diving plays because the guy who records who did what isn’t the one who determines if that guy is good.
The best thing to say about the good metrics is that they’re probably not perfect, but they’re not useless. It’s kinda like using OPS or something. Maybe OPS doesn’t always tell you who the best players are, but if one guy has an OPS of 1.000 and another has an OPS of .600, there’s little doubt who is the better player. If all the metrics say Ellis is good and Jeter/Michael Young are bad, we can be pretty sure of those things.
- - There is some current debate over scorers for certain games/at certain parks and what is considered a fly ball and what is considered a line drive. This bias might be something that affects certain data, but the bias here is a systematic thing, and since it might be across the board with proper study it can be accounted for to a great extent.
I'm not sure how this makes sense
Bias you can correct for — its precision means that its inaccuracy can be adjusted.
Noise is neither precise nor accurate.
A B -3X = Swedish girls like chocolate @('.')@
I think there's a lot of merit to the wisdom of crowds approach to doing things.
by Aaron Campeau on Jan 26, 2009 11:31 AM PST up reply actions
Wisdom of crowds doesn't necessarily work when there's a pervasive bias in one direction
The reason why the crowd is invariably right on Who Wants to Be a Millionaire is that no one has a vested interest in any particular answer being correct.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
According to the wisdom of crowds
All math more difficult than addition is incorrect. (citing all the SAT questions more than half the test-takers miss)
"There's never enough time to do all the nothing you want" -Bill Watterson
I know you're trying to be snarky, but this is wrong
The wisdom of crowds would only fail if a particular wrong answer actually got more responses than the correct one. There might be a few intentionally misleading questions designed to trick the unwary into an “easy” but wrong answer, but I doubt that’s what’s going on with most of those SAT questions.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
I decided to test this theory
And googled “math question with results”
That got me to here, so naturally I tried the 4th grade test.
For question 3 (“The distance from one town to another is 180 km. If Betty has to drive the distance in 3 hours, what must her average speed be in kilometers per hour?” Answer: “180 ÷ 3”) was answered correctly by 37.1% of Americans. Of course, I can’t prove that the wisdom of the crowds picked a different answer, but I was feeling good.
That brought me to question 8 (“0.4 is the same as” Answer: “four tenths”). In America we “nailed” it, with 40% right. In England, however, only 20% of the people got it. In other words, because there are 4 choices I can guarantee to you that the wisdom of the UK crowds thinks 0.4 is something other than four tenths.
So snarky? Yes, guilty as charged.
Wrong? No, not so much. It took me 5 minutes on google to find at least one easy example.
"There's never enough time to do all the nothing you want" -Bill Watterson
Well, OK, if the "crowd" really is that stupid, it won't help you
Seems like an unusual occurrence, and I do wonder whether the sample size was really large enough or there were some factors that we’re not aware of affecting the British score (like decimals not being taught at all until late in the 4th grade).
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
This is why betting markets are superior to the traditional wisdom of the crowds approach
when it comes to aggregating information. It essentially offers a payout to experts who see a current bias in prices/responses etc. to come in and correct that bias. This enables those with more information to “put their money where their mouth is,” while those who lack information sit on the sidelines, or risk losing money.
by CapgrasDelusion on Jan 27, 2009 12:21 PM PST up reply actions
Hm, I'm not sure I agree with that
In many instances, it’s efficient for well-heeled investors to take short-term losses to promote disinformation for the purposes of long term gain. Granted, those situations mostly occur in the context of politics and not math, but I’d guess they occur significantly more often than the “the public is so stupid it actually thinks a wrong answer is the right answer” scenario.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
Other traders are cognizant of this possibility
Any investor attempting to manipulate the price, or matter of fact being decided, would send a signal to others that either a.) he possesses relevant information, or b.) he is acting for some other non-information based purpose. The awareness of possibility b limits the amount of influence such an individual is able to exert on prices. Additionally, price changes brought about by non-informational considerations produces an incentive for those who do posses relevant information to enter the market and mitigate the manipulator’s influence.
I suppose you could argue that a group of traders are in cahoots with one another; but I would argue that, for the most part, information is dispersed too widely for such an occurrence. By no means would I argue that betting markets are perfect, but I think they are superior to any existing methods of aggregating information.
by CapgrasDelusion on Jan 27, 2009 6:39 PM PST up reply actions
The kinds of people likely to respond to a survey initiated by Tom Tango are likely to be fairly well informed.
by Aaron Campeau on Jan 26, 2009 10:44 PM PST up reply actions
This is true, though he is putting it out to non-saber-friendly places, too.
But being well-informed doesn’t mean there isn’t some bias in there.
Being a saberist and being able to observe fielding are two different skill sets.
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 27, 2009 11:32 PM PST up reply actions
My favorites
1). What are you lookin’ at?
2). Wasn’t me…
3). Like I’d ever do that…
4). That’s just stupid!
All come in handy when needing to convince one of your supposed innocence.
As for defense in baseball? Not a clue.
"Their batters are patient to the point that it's annoying." -Ryan Franklin
You forgot the go to one...
5). I have an evil twin that has escaped.
The nuts and bolts of gameplay are apocalyptic failures, but the awfulness doesn’t stop there. Managing games is utterly pointless. [Feb 2009, p.85]
6). I have an evil twin whom I "absorbed" at birth and sometimes controls my actions
A B -3X = Swedish girls like chocolate @('.')@
Range factor
Range factor is the best at showing defense ability to get to the ball not errors per chances. If you ever have played a baseball game called stratomatic you would know that defense is very important to a games outcome. You can go by eyesight but sometimes those great plays are just a player getting lucky. For example Derek Jeter makes the highlights because of playing in NY but his range is average at best. Arod is claimed to be good at third but he is terrible. In outfield startomatic rates Jack Cust as a 5. The lowest rating a player could have. When listening to a broadcaster for another team’s station you will hear them say Jack Cust should have caught that. The A’s broadcaster are bias in not giving the true misplays of Cust. One game I watched Cust cost the A’s 3 of 5 runs scored by not making the routine play in the outfield.
Except range factor is terrible because it doesn't accurately reflect how many chances you have either
In the unlikely, but possible, scenario where a great defensive player doesn’t have a ball hit anywhere near him, his RF is going to be 0. Say this happens for…2 months straight. Yeah, that would never actually happen, but it could. What if that guy is Willie Mays? Is he a bad outfielder because people only struck out and hit ground balls?
Range Factor only tells you how many plays you make against your own team. It says nothing for how many plays you make that you actually can or should make.
Bias alert
I recognize that my view here violates my level-headed reputation on this site, but here goes:
I think defensive metrics are, by and large, a fruitless attempt to quantify the unquantifiable. Telling me, for example, that Jack Cust is a bad defensive outfielder isn’t something I have not learned with my own eyes.
I have a lot of respect for stat geeks, and am happy to cite their research or information about a lot of things. And it’s certainly true, in the case of defense, that some players have reputations that don’t match up with their performance. That is especially true with players who capable of making SportsCenter-type plays but who screw up at other times. And it’s harder to judge a player’s defensive performance if you do not watch them regularly.
But for any player I have seen regularly, I can tell you their defensive strengths and weaknesses. When the Giambi signing was being discussed, somebody on a thread jotted down the obvious things I recall: no range, can’t throw (especially to second base), pretty good at scooping balls in the dirt. He may have changed, especially because he doesn’t play the position much anymore and is a lot older, I doubt he’s changed much.
Mark Ellis’ strength isn’t his range, although it’s good. It’s the fact that he rarely makes an error on a ball he does get.
I also think that, in most cases, defense is overrated – unless the player is really good or really bad. A lot of what we consider good defense involves decision-making, which is difficult to measure.
All of the above may simply qualify me as a fuddy-duddy who is unwilling or unable to learn a new language, but that’s my take.
by bear88 on Jan 26, 2009 9:01 AM PST reply actions 1 recs
First of all, even if you watch players regularly, it’s really hard to tell how good they are relative to the rest of the players at the position who you don’t see regularly.
Second of all, quantifying it is sort of the point. I can tell with my own eyes that Cust has good power and patience, and bad contact skills, and I don’t need stats to be confident in that assertion. But offensive stats tell you how good each player is relative to all the others to a specific enough degree that you couldn’t possibly match it by just watching all the games and making an educated guess. Same goes for defense. The stats are weaker, but then the observation is weaker also. You know Jack Cust is bad, but how bad? You know Mark Ellis is good and can say why, but how good? Various players have strong arms but are slow, which is obvious to anyone who watches, but then what to say about their defensive value overall? However imprecise the defensive stats are,
Also: range tends to be the most important element of fielding. Consider how difficult it is to accurately gauge someone’s range from watching the games. #1, if you’re watching on tv, they don’t even show you where the fielder started from to get to the ball, so it’s nearly impossible to tell on tv how impressive or not impressive some player’s range is on a given play. #2, even if you can watch the whole play unfold, it is incredibly difficult to watch a play and be able to imagine what a range of other fielders would have done with the play in order to discern whether the guy you just watched is better or worse.
With stout hearts, and with enthusiasm for the contest, let us go forward to victory. ----Hero Defector Montgomery
Yes, but ...
… regarding your #2, imagine all the problems with doing just that — and being expected to transcribe the precise location of the fielder at delivery of the pitch plus the trajectory and velocity of the batted ball, and assign several values to each factor. For nine innings. On every play. That’s what defensive stats are built on.
A B -3X = Swedish girls like chocolate @('.')@
Well, they don't have to compare any players to each other
the ball trajectory/velocity problems will lead to some vagueness and various mistakes, but it shouldn’t be all that difficult to mark down where a ball lands when that’s your only job. And these things are only really a problem to the extent that there’s bias. If the tabulators make errors, it’s not a big deal in a large sample. It’s only a big deal different players are being systematically evaluated differently. I don’t know if or to what extent that goes on (it would be a problem if, as I think might happen, each park has a different person doing and they classify the batted balls differently, but I don’t think the method itself is a huge problem.
With stout hearts, and with enthusiasm for the contest, let us go forward to victory. ----Hero Defector Montgomery
And it may be that the defensive stats are mostly worthless, but I'm quite sure that observation is mostly worthless
beyond placing players in very broad categories.
With stout hearts, and with enthusiasm for the contest, let us go forward to victory. ----Hero Defector Montgomery
But unless you have one tabulator for every defender at every game, they have to note the starting position of every single player on the field at a single moment (delivery of the pitch), right? Unless they somehow know which player the ball is going to be hit to before the pitch is delivered … {begins composing conspiratorial screed}.
A B -3X = Swedish girls like chocolate @('.')@
Why do you need to know the starting point? Isn't positioning part of defensive performance?
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 26, 2009 12:50 PM PST up reply actions
Wait. What?
I’m not seeing your apparent intended disconnect between those two questions.
I’d guess that with defensive positioning, it’ll remain impossible to statistically parse out the difference between individual and team/scouting/coaching contributions.
But … if you don’t measure where the defender starts … then how do you calculate his range?
A B -3X = Swedish girls like chocolate @('.')@
they don't take into account where the fielder starts,
which creates some problems, especially in the outfield, but it’s not a huge deal unless some teams are really bad at positioning.
With stout hearts, and with enthusiasm for the contest, let us go forward to victory. ----Hero Defector Montgomery
I don’t agree with that. It can mean that a player gets a lot of balls that are not in his zone and doesn’t get ones that are. There are also different strategic reasons for positioning, like doubles defense that doesn’t necessarily translate into an accurate reflection in rating systems.
Some systems, ex. RZR, separate plays out of zone from plays in zone
Different defenses make such a minimal difference that whatever error they introduce is swallowed up within the various other forms of statistical noise that’s out there.
Let’s say a team spends 5 plays a game in a major variant defense (no-doubles, infield in, infield shift), and another only spends 4 plays a game that way. Overall you’re looking at 162 batters for the entire season where the teams differ. A given fielder might get 20 of those, at most. Most of those will be plays he couldn’t have made anyway or makes no matter what defensive formation the team is in. Maybe 1 or 2 extra plays get made or not by a given player in a given season. Very, very marginal.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
“Major variant defenses” like doubles defense is just one example of a shift. If you want to conclude that a single major variant defense typically doesn’t have a significant effect over the course of a season I would agree. From that premise I wouldn’t conclude that defensive shifts have a minimal difference and that it is swallowed up by statistical noise.
I really presented doubles defense as an example. Some teams shift defensively on most plays.
The calculation of plays in the zone and OOZ plays has changed and not surprisingly it has affected the ratings of players.
Even when that change results in different ratings for players, the people who read those ratings still speak authoritatively about the results both before and after the change.
Position of the player in relation to rigid zones is just one flaw.
Some teams shift defensively on most plays?
Really. Care to back that up?
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
So, that's a "no," then
Ok.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
Not really,
but you seem to be a person who can’t be satisfied. I’m sure nothing I can say will alter your belief in anything.
Can it
This is a simple factual question. Does a team shift its fielders out of normal position “most” of the time, or not?
If you have actual data backing up your assertion that the Brewers do so, then fine— show it to me and I will “alter my belief”. If not, then no— you won’t convince me by just saying it’s true repeatedly.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
I don’t believe you. If you can point me to a discussion where you admitted to being wrong about something other than a minor detail then fine – show it to me and I will alter my belief.
As to shifting fielders you are welcome to march forward thinking that there is a “normal position” and that infielders don’t shift or that outfielders don’t play shallow or deep or that the handedness of the batter is irrelevant to positioning or that SSs don’t watch the pitch call and/or catcher positioning and cheat or that the adjacent fielder doesn’t play into positioning. The list goes on beyond that. The Brewers assign positioning responsibility for infielders and outfielders between different coaches and you can see the outfielders check their “cheat sheet” all the time.
Do the A’s all stay in the “normal position” other than extraordinary circumstances ilke double defense? I don’t believe it.
I don't think Range should be defined as how far a fielder can run to catch a ball,
it should be how large an area he can cover. If he always stands where the ball is going, that should count in his favor.
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 26, 2009 1:44 PM PST up reply actions
Why?
It's not the results, it's how you look going about those results -- Tim McCarver
by WaddellCanseco on Jan 27, 2009 11:33 PM PST up reply actions
the line drives to the head?
"True fact: In a global thermonuclear war, the only human who would survive would be David Eckstein" -PT
This does not comport with my understanding of how defensive stats are compiled
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
The observational part of defensive stats has to do with the speed, trajectory, and landing point of the ball
not with where the fielder starts the play. If a guy is out of normal position, he’ll either improve the number of plays he makes or reduce it. Either way, the denominator (the number of balls hit to his zone) is going to be the same, because the fielder doesn’t control what the hitters do.
Many years from now, when his name's recalled
Everyone will say, "He should have passed the ball"
-- Al Stewart, "Football Hero"
Saying "stat geeks" and then claiming you respect the people who put in all this work is kind of a contradiction
Also, Mark Ellis’ strength IS his range.
It's not very easy
the metrics tend to disagree with one another to some extent, there are pretty large differences depending on the two data sources, one year of data is not very reliable even putting aside the quality of the data/metrics, and some of the best metrics are not freely available.
For guys who have been in the league for awhile though, the metrics will be more reliable than anything else. The best/easiest thing to look at is uzr on fangraphs. It’s much better and easier to understand (because it gives a number in runs) than the other free ones.
What I do once in awhile when I have time is look at uzr on fangraphs, zone rating on espn, rzr on hardball times, and pmr on pinto’s site. If you look at all of those for 2-5 years or whatever you can get a pretty good sense of roughly how good a player is. The Fans scouting report on insidethebook.com is very good for players you don’t know anything about who haven’t been in the league very long. “professional scouting reports” would also be good, but it’s not very easy to find that.
Ellis: The defensive metrics are not particularly precise, but they give a range that a player is very likely in. Every metric loves Ellis and has had him in the top 1-2 2Bs the last few years. So people are confident in his value because every metric would have to be very wrong for him to not actually have that value.
With stout hearts, and with enthusiasm for the contest, let us go forward to victory. ----Hero Defector Montgomery
Do you think measures/projections of overall value (which as far as I can tell accord equal weight to offensive and defensive statistics) are … accurate? … misleading? … useful if considered with a grain of salt?
Lies... damn lies...
and false modesty.
"There's never enough time to do all the nothing you want" -Bill Watterson
Well, the general method is right
what you absolutely should not do is say “the offensive stats are more accurate/reliable so I’ll give more weight to offense.” But, individual published overall value measures (e.g. win shares, win value on fangraphs, oense +defense,etc.) don’t do a very good job with defense, usually picking some single metric. they should therefore be taken with a huge grain o salt as to the defensive component. they will be misleading when, as often happens, whatever metric the total value measure uses is an outlier relative to the rest of the def. metrics. chone’s projections are also iffy, but not a terrible way of doing it. they can be improved by looking at all the available defensive data/fans scouting report/what have you, and coming up with a guess at a number for defense based on that.
With stout hearts, and with enthusiasm for the contest, let us go forward to victory. ----Hero Defector Montgomery
You know that is wonderfully common-sensical
My rating system for how often they are treated that way is somewhere between now and then and once in a blue moon.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Certain photos copyright © 2010 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
