F% vs RZR

[ORS]

Cam thinks F% is more important than zone rating stats because errors make him sad. I think making outs is the objective, so stats that quantify that are peachy keen.

 

Discuss.

[example1]

Really ORS? Really? A new thread to remind Cam that he's wrong? I appreciate opening another forum, but I thought you had finished with him earlier...

 

Yes, lots of stats are better than F%, except errors. Errors are worse.

[ORS]

Mod request.

[26 to 6]

who the hell is Cam?

[26 to 6]

Nevermind

[ksushi]

I didn't know if you were finished with him or not, I just poked my head up from studying and saw that Johan talked had turned into fielding metics talk...

 

Apparently you had your way with him. Sweet. I tend to think errors are a pretty irrelevant and extremely subjective stat. RZR I think tells the tale a little more accurately though, with defense, I tend to just grab a scouting report and read because there really is no 100% reliable defensive stat.

[cam780]

I think the most amazing part of this conversation is that the whole thing started with my assertion that Coco Crisp might not have the same season he did this year. But this is the part of our conversation that sticks out to me. I'm not going to quote it but I will instead copy and paste:

 

I don't think you are appreciating what RZR measures. It's been stated already, but I'll repeat it, it measures the rate at which a fielder can turn balls hit into his zone into outs. These balls hit into his zone are either unfielded, fielded and turned into outs, or fielded and not turned into outs. Errors fall into the 3rd category. So, errors are a component of the result.

 

Zone rating goes alot further in identifying bad defensive players than it does identifying good ones and it dissects the field into zones and thus doesn't take into account players making plays out of position. For example, how many times during the course of the year, will Mike Lowell make a play in Lugo's "zone". Doesn't that factor into a players defense in your mind? It does mine.

 

RZR absolutely considers players making plays out of zone. RZR eliminates out of zone plays (OOZ) and creates a new category for them. I'm sure you are thinking, "But Lowell's OOZ plays dropped from 43 to 27. Surely that supports my position." Maybe, but let's look at who was playing next to him. In 2005 and 2006, Alex Gonzalez made 41 and 42 OOZ plays (conveniently they played together in 2005 so we have two years of data to look at), and Mike Lowell made 39 and 43. Looks pretty consistent. Lugo makes 55 last year. Lowell wasn't required to make as many OOZ plays because he was playing next to a SS with superior range.

 

You can't will facts to fit your argument. A stat that considers something is the opposite of a stat the eliminates it. This may be difficult to prove but let me give it shot - Lowell made 227 plays and had a 310 BIZ if you divide the former by the latter you acutally get .732 his RZR or in other words. RZR wouldn't be considering OOZ and OOZ is only a figure of actual out of zone plays not a percentage based on opportunities.

 

Do you know what RZR stands for? It's Revized Zone Rating. One of the flaws in the original ZR was the inclusion of OOZ plays. RZR eliminates them from the ZR, and gives them their own category. I don't know where you are going with this, but it seems this information seems to have eluded you, so there you go.

 

Your first point is basically that I don't know what I'm talking about. My response is I think a players defense isn't limited to his positional "zone" especially in the case of third and and short.

 

You then respond by saying what are you talking about RZR considers out of zone plays because of OOZ.

 

I then respond by saying well zone rating doesn't factor in OOZ because its a seperate category and then you say .. duh, you don't know what you're talking about here's the definition. it's not factored in which was in fact my original point. now the reason why I think there's an issue with RZR is because if you have a good fielder that is consistently making plays outside of his zone then he should receive some credit for it and if your position is that credit is given through OOZ then its not given through RZR.

 

Again, my position is that RZR is a much better way differentiate comparables than to define a season. ie. Inge and Beltre both had 18 E last season and it really is the RZR that serves as the contrast. I realize my defiance to crown RZR the king of defensive stats probably means I'm not getting an inventation to your next tea party but standing alone its not the perfect indicator of defense.

[ORS]

I then respond by saying well zone rating doesn't factor in OOZ because its a seperate category and then you say .. duh, you don't know what you're talking about here's the definition. it's not factored in which was in fact my original point. now the reason why I think there's an issue with RZR is because if you have a good fielder that is consistently making plays outside of his zone then he should receive some credit for it and if your position is that credit is given through OOZ then its not given through RZR.

 

Again, my position is that RZR is a much better way differentiate comparables than to define a season. ie. Inge and Beltre both had 18 E last season and it really is the RZR that serves as the contrast. I realize my defiance to crown RZR the king of defensive stats probably means I'm not getting an inventation to your next tea party but standing alone its not the perfect indicator of defense.

Great, now I need to define words for you in order to bring you up to speed. My use of the word "consider" for what RZR does with OOZ plays is exactly right.

 

consider - 1. to think carefully about, esp. in order to make a decision; contemplate; reflect on: He considered the cost before buying the new car.

2. to regard as or deem to be: I consider the story improbable.

3. to think, believe, or suppose: We consider his reply unsatisfactory.

4. to bear in mind; make allowance for: The arrest was justified if you consider his disorderly behavior.

5. to pay attention to; regard: He considered the man for some time before speaking to him.

6. to regard with respect, thoughtfulness, honor, etc.; esteem.

7. to think about (something that one might do, accept, buy, etc.): to consider a job in Guatemala.

8. Obsolete. to view attentively; scrutinize.

9. Obsolete. to recompense or remunerate.

–verb (used without object) 10. to think deliberately or carefully; reflect.

11. to view carefully or thoughtfully

 

Considers, not includes. RZR "considers" that some plays in the raw data might not be "in zone". ZR does no such "consideration", and anything a player makes a play on is counted "in zone".

 

I never claimed RZR "included" OOZ plays in the result.

 

Ok, now that we are done with that, let's address the last paragraph. First of all, since RZR is an improvement on ZR, and the improvement is that OOZ plays are isolated as a counted stat and eliminated from the ZR calculation, then I think it would naturally follow that RZR/OOZ are paired stats. I apologize for assuming that was common sense. Finally, this discussion isn't about the absolute value of RZR in the defensive metric spectrum. No, this is about the merit of RZR relative to F%, which, in the discussion of Lowell, you identified as a more meaningful stat. If you are willing to come back from that claim in light of statistical data that shows that position to be untenable, then we can end this.

[cam780]

Great, now I need to define words for you in order to bring you up to speed. My use of the word "consider" for what RZR does with OOZ plays is exactly right.

 

consider - 1. to think carefully about, esp. in order to make a decision; contemplate; reflect on: He considered the cost before buying the new car.

2. to regard as or deem to be: I consider the story improbable.

3. to think, believe, or suppose: We consider his reply unsatisfactory.

4. to bear in mind; make allowance for: The arrest was justified if you consider his disorderly behavior.

5. to pay attention to; regard: He considered the man for some time before speaking to him.

6. to regard with respect, thoughtfulness, honor, etc.; esteem.

7. to think about (something that one might do, accept, buy, etc.): to consider a job in Guatemala.

8. Obsolete. to view attentively; scrutinize.

9. Obsolete. to recompense or remunerate.

–verb (used without object) 10. to think deliberately or carefully; reflect.

11. to view carefully or thoughtfully

 

Considers, not includes. RZR "considers" that some plays in the raw data might not be "in zone". ZR does no such "consideration", and anything a player makes a play on is counted "in zone".

 

I never claimed RZR "included" OOZ plays in the result.

 

Ok, now that we are done with that, let's address the last paragraph. First of all, since RZR is an improvement on ZR, and the improvement is that OOZ plays are isolated as a counted stat and eliminated from the ZR calculation, then I think it would naturally follow that RZR/OOZ are paired stats. I apologize for assuming that was common sense. Finally, this discussion isn't about the absolute value of RZR in the defensive metric spectrum. No, this is about the merit of RZR relative to F%, which, in the discussion of Lowell, you identified as a more meaningful stat. If you are willing to come back from that claim in light of statistical data that shows that position to be untenable, then we can end this.

 

I think what we're talking about is purely semantics. I think in order to get a correct view of a fielders ability you have to consider both but keep in mind that this conversation is about statistics. You could also say that batting average considers walks because it eliminates them but that number that you get has nothing to do with his walks. Personally, I would say that OBP considers walks because that figure that you see has walks factored in. Mike Lowells RZR has nothing to do with his OOZ. Each is a separate statistic. I understand how difficult it would be to figure it out but in order to make this a more complete stat I think you need OOZ as a percentage - that is the RZR being opportunities and plays made in the players assigned positional zone and the OOZ being the number of opportunities vs plays made in other zones. When you see a number like 27 (I didn't go back to check it) for Lowell's OOZ maybe he made every play coming to him while he was in the SS zone but what if he had 60 opportunities? I agree that including plays made out of zone gave players some additional benefit in the old ZR but consider your argument.

 

You have said that RZR considers OOZ because it seperates it and you would have to look at them both. But this year, Lowell's RZR was .732 and his OOZ is 27. How would you compare that to Lowell's 2004 - a .654 RZR and a 52 OOZ. You have to admit its difficult because what OOZ presents is an absolute number and not a percentage. Now I don't know the data on how many of Lowell's errors came in zone but what if 5 of Lowell's 15 errors occurred in the SS zone. That wouldn't appear in OOZ because it wasn't a play made and it wouldn't affect the RZR because it was out of zone. The difference there is the difference between Lowell having a .732 and a .720 RZR. I only have the information that Hardball times has on its website so I can't compare it career wise but that number would be less than Lowell's 2005 RZR of .728 and therefore 2nd worst over that span In our previous discussion, you used Lugo's range vs Gonzalez's to state that Lowell didn't have to make as many plays but that is far from a substantiated claim. While I admit its probably not the range you're speaking of, the RF for Gonzo in 04 was 4.33 - and 4.21 for Lugo in '07. If you have another stat that supports your stance on Lugo v Gonzalez, I'm more than willing to consider it.

 

Also, this has nothing to do with Lowell but consider the shift and Ortiz. Obviously based on game situations there are some points when the fielding team can't utilize it but if Ortiz gets to bat 3 times at least once you have to consider that 1 of those at bats the shift would come into play if its a team in the division you're talking about at least 20 plays out of position and thats just for Ortiz. Would it be fair to assume that if you extrapolate that 50-100 plays would take place out of a fielders typical zone both strategically and unintentionally. If you consider that Lowell had 227 plays in zone. That is actually a lot.

 

You also have my original point wrong. If you go back to the original thread, I made a point that Crisp could have a down year that you couldn't count on him making every play that he did in 2007 next year. I stated that Mike Lowell had a down year defensively (again not a bad year). I did not mention RZR or any other stat for that matter in my original thread. Red Sox rules then responded that Lowell didn't have a down year. I mentioned that errors , f% and range factor were all career lows. He stated that he believed that zr (not rzr as he stated it was .778) was a better measure of defense. There was a link to the espn calculator and sure enough his zr was one of the worst in the time span allowed to be calculated. I mentioned the same points I have restated here that I think that if you rely on RZR as the stat of choice to measure someones defense than you basically don't give him credit or admonition for plays he does or doesn't make out of that zone at least fielding percentage takes into account number of plays to a certain player as a whole as opposed to within a certain box. Using an analogy to a different sport, its like pulling apart Rodney Harrison's tackles that he has made playing linebacker because he wasn't playing safety at the time. My original position isn't F% is a better indicator of what Lowell has done vs RZR but instead that Lowell had down year. To that end this is what you have.

 

Errors - 15 worst in his career

F% - .961 worst in career

RF - 2.51 worst in career

RZR - .732 in four years of available statistics 2nd best, his third best year is .728 albeit with 60 more plays in his zone

OOZ - 27 in four years of available statistics, the fewest he has made.

 

So if Ksushi is visiting this post while he's studying. Say he typically has a 3.5 GPA in most of his subjects. Let's assume his major is biology. Say this semester he gets a 2.2 in English, a 2.5 in Spanish, a 2.5 History, a 3.2 in Biology and a 2.0 in water color painting. Is he going to say he had an off semester? Or is he going to feel real good about himself because he was just slightly down from his usual GPA in his major?

 

I also don't think its untenable that I prefer a stat that, while flawed, doesn't separate the plays a player makes based on where he makes them. Neither stat is perfect mind you, and like I said, RZR is perfect to differentiate between two players with comparable peripheral stats, but it almost seems too specialized to measure the player as a whole.

 

One question and its because I truly don't know the answer. How does RZR address players that play different positions. For example, Eric Hinske. Say he makes 80 starts at first in a given year and he has one error and fails to convert 6 other plays for outs with a 100 BIZ, he plays left for 40 games with 5 errors and fails to convert 7 outs with a 70 BIZ, he then plays 40 games in right with no errors and 3 outs that weren't converted with a 70 BIZ, and he plays 2 games at third, no errors, 100% of outs converted in 4 chances. Would he qualify at first but not in the other three positions and at that point would the plays he made at the other positions be listed as OOZ? Or would he simply not qualify at the other three spots or would you average them and if you averaged them how would you calculate the OOZ?

[ORS]

What is a fielder trying to do? Make outs. The stat you prefer, in the case of a 3B, ignores about 30% of the balls the have an opportunity to turn into outs and only tells you about the fielder's failure on about 5% of them. The other stat, mind you I said RZR and OOZ are a paired stat so continuing to state that RZR is more meaningful only when you look at OOZ makes it look like you have a severe reading comprehension deficiency, RZR/OOZ ignores none of these. I can't state it any clearer than that. You'd rather look at 70% data over 100%. Indefensible.

[cam780]

Usually, when you're having a debate - someone makes a point then the other person makes a counterpoint and then the first person addresses the counterpoint and so on and so forth.This one is going like this I make a point. You make a counter. I counter your point. You repeated your initial point. I once again countered your initial. You once again repeated your initial argument and then claimed I can't read. Apparently, I'm supposed to be changing my mind based solely on your opinion. I don't know if you've managed to bully your competition into concession with arrogance in the past but this ain't happening here.

 

In 2007, Lowell's BIZ accounted for 80% of his chances, in 2006 Lowell's BIZ only accounted for 72% of his total chances, in 2005 it was 71% while in 2004 Lowell's BIZ accounted for 67% of his chances. Yes, RZR accounts for everything in the zone and OOZ accounts for everything made out of the zone but what about the other 20-33% of plays that he had a chance to make?

 

If you add to it that the range for third basemen is from zones C-F and zone G is a "hole" that isn't attributed to any position and that ground balls are the only plays taken into account for infielders and you get a less than perfect number much like fielding percentage.

 

Again, I'm not arguing that F% is far and away better than RZR, I'm just arguing that you can't contend that when all of the supporting stats (including OOZ, its sister stat) are down that a player is having a career-average season based on one metric. Of course, my opinion differs from yours which I'm sure you'll remind me is due to my inability to read as opposed to your incorrigible stubbornness.

[26 to 6]

ouch

[ORS]

Usually, when you're having a debate - someone makes a point then the other person makes a counterpoint and then the first person addresses the counterpoint and so on and so forth.This one is going like this I make a point. You make a counter. I counter your point. You repeated your initial point. I once again countered your initial. You once again repeated your initial argument and then claimed I can't read. Apparently, I'm supposed to be changing my mind based solely on your opinion. I don't know if you've managed to bully your competition into concession with arrogance in the past but this ain't happening here.

 

In 2007, Lowell's BIZ accounted for 80% of his chances, in 2006 Lowell's BIZ only accounted for 72% of his total chances, in 2005 it was 71% while in 2004 Lowell's BIZ accounted for 67% of his chances. Yes, RZR accounts for everything in the zone and OOZ accounts for everything made out of the zone but what about the other 20-33% of plays that he had a chance to make?

 

If you add to it that the range for third basemen is from zones C-F and zone G is a "hole" that isn't attributed to any position and that ground balls are the only plays taken into account for infielders and you get a less than perfect number much like fielding percentage.

 

Again, I'm not arguing that F% is far and away better than RZR, I'm just arguing that you can't contend that when all of the supporting stats (including OOZ, its sister stat) are down that a player is having a career-average season based on one metric. Of course, my opinion differs from yours which I'm sure you'll remind me is due to my inability to read as opposed to your incorrigible stubbornness.

Good point about debate format, now if you'll counter my first point, we can get to the give and take. Let's look at the timeline.

 

I make the point that the outcome of failure in either stat is about the same (backed up by the lin wts), and contend allowing more marginal hits than errors is worse.

 

The "counter" point is some blather about what you think ZR stats measure and are good for, a hypothetical that makes you unconfortable for zone rating stats, and ends with how F% is just your preference and we should leave it at that. Note, none of this addresses my original point.

 

We go back and forth clarifying our positions.

 

I contend there is a data gap. Somehow, this is a reiteration of my value point.

 

You call me on my debate tactics and call me stubborn.

 

I'll start with your hypotheticals. You are right, an error made OOZ doesn't show up in the RZR/OOZ data. I don't think that is of significant consequence for a couple of reasons. One, if a guy is playing straight (positioning), it's unlikely he gets credited with an error because OOZ means he's attempting at play that the typical fielder turns into an out less than 50% of the time. Two, we still have the error count to look at when comparing players, so while the first rate stat I'd look at would be RZR, I'd supplement that with F% when RZR performance is close.

 

If the discussion were about the flaw of ZR based stats, and I was defending their perfection, you'd be doing swimmingly. Unfortunately, I, nor anyone else here, has lauded their perfection and have acknowledged they are flawed. What you've failed to do is demonstrate, other than saying it's your opinion, why the flaws of ZR based stats outweigh the flaws of fielding percentage. Give me something. If it's going to be another hypothetical, at least give me some rudimentary estimate of frequency.

 

As for being stubborn, that had me laughing. Nobody on this site or in the field of baseball statistical study values that archaic stat over ZR based stats. None. And in light of compelling analytical (ie not "it's just my opinion") data, you refuse to acknowledge you may be wrong. Gee, I wonder what's causing this?

[Jayhawk Bill]

I thought that I'd weigh in here because I was involved in this genesis of this discussion.

 

I think making outs is the objective, so stats that quantify that are peachy keen.

 

Concur.

 

First of all, since RZR is an improvement on ZR, and the improvement is that OOZ plays are isolated as a counted stat and eliminated from the ZR calculation, then I think it would naturally follow that RZR/OOZ are paired stats. I apologize for assuming that was common sense. Finally, this discussion isn't about the absolute value of RZR in the defensive metric spectrum. No, this is about the merit of RZR relative to F%, which, in the discussion of Lowell, you identified as a more meaningful stat.

 

Concur. RZR has more merit than fielding percentage because it counts outs, not just errors. RZR and OOZ are paired stats, and for a best-available picture one would consider both.

 

I guess some of the issue is the apples-and-oranges nature of RZR and OOZ. RZR is a rate stat, while OOZ is a counting stat. We can try to equalize that by assuming that each player had a chance to make OOZ plays at the same rate as in-zone plays, creating an Out-of-Zone Rate (OOZR = OOZ/BIZ) and an RZR+OOZR stat, a comprehensive rate metric. Here's how qualifying AL third basemen would look with such a stat:

 

[table]Name | RZR + OOZR

Glaus | 0.941

Inge | 0.883

Beltre | 0.869

Mora | 0.855

Gordon | 0.848

Rodriguez | 0.821

Lowell | 0.819

Blake | 0.793

Iwamura | 0.789[/table]

 

It's no surprise that Mike Lowell drops down. His fielding range is primarily forward, not lateral. He's the best in MLB, IMO, at fielding bunts and weak ground balls that often convert into infield hits. When one considers lateral range, Lowell should probably drop further ordinally than any other third baseman.

 

A combined RZR + OOZR is a useful stat.

 

Should he drop this much, from first to seventh?

 

Probably not. We made an assumption that OOZ plays come at the same rate as in-zone plays. But THT publishes a stat for annual infield OOZ plays by team. Let's check our assumption against that list:

 

[table]Team | Inf OOZ

TOR | 255

OAK | 216

TEX | 215

BAL | 206

MIN | 203

LAA | 202

KC | 201

SEA | 200

DET | 197

CLE | 187

NYA | 186

TB | 186

BOS | 171

CHA | 169[/table]

 

Holy cow! No wonder Glaus looks so good: his opportunity to make OOZ plays looks to be roughly 50% better than it would be were he playing for Boston or Chicago.

 

Let's consider team OOZ opportunities to be the OOZ plays by team's infield less the player's opportunities, and let's define adjusted OOZR as equal to OOZR x (league-average OOZ opportunities/team OOZ opportunities) and look at that:

 

 

[table]Name | RZR + Adj OOZR

Inge | 0.910

Beltre | 0.897

Glaus | 0.882

Mora | 0.852

Gordon | 0.844

Rodriguez | 0.839

Lowell | 0.826

Iwamura | 0.799

Blake | 0.790[/table]

 

First, Troy Glaus once again looks mortal. Inge is the best third baseman in the AL. Lowell remains seventh-best.

 

This still isn't ideal, and it's frustrating that THT (well, BIS, their data source) has the actual number of adjacent-zone plays for each third baseman. With that information, we could create an actual RZR + OOZR stat that didn't rely upon any assumptions for its OOZR computation.

 

But look back at that original RZR + OOZR rate. Excepting Glaus, the ordinal rankings of third basemen didn't budge much. I went through all of this to address the obvious challenge that Glaus simply isn't better than Inge. Again, RZR + OOZR is a useful stat.

 

Again, my position is that RZR is a much better way differentiate comparables than to define a season. ie. Inge and Beltre both had 18 E last season and it really is the RZR that serves as the contrast.

 

Cam, I think that this is where we differ.

 

It makes little to no difference how many errors Inge or Beltre had. What matters is how many balls in play they did or didn't turn into outs. Remember that the value of an error pretty closely equates to that of a single? An error is just another ball in play where the runner reached base. Minimizing those situations is the job of the defense.

 

In particular, though, judging third base defense by error count can be deceiving. Here are the qualifying AL third basemen, ranked by fielding percentage:

 

[table]Last | FPct | RZR + OOZR

Iwamura | 0.975 | 0.789

Mora | 0.971 | 0.855

Glaus | 0.967 | 0.941

Rodriguez | 0.965 | 0.821

Blake | 0.962 | 0.793

Gordon | 0.961 | 0.848

Lowell | 0.961 | 0.819

Inge | 0.959 | 0.883

Beltre | 0.958 | 0.869[/table]

 

By fielding percentage, the rate stat for errors, Iwamura becomes the best third baseman in the AL. By evaluating of converting balls in play into outs, he's worst. Inge and Beltre are worst by fielding percentage. They're two of the three best at turning balls in play into outs.

 

Lastly, evaluating fielding by counting errors understates the value of defense. The range from best to worst among qualifying AL third basemen by error count is just 11 plays over the course of the last season. The difference in total plays between Inge and Iwamura was 140 plays. Rate stats describe the differences more accurately and precisely, but the difference in counting stats hits home better. If one player had 140 more hits than another, you'd better believe that we'd notice. Evaluating defense with fielding percentage conceals that difference.

 

So if Ksushi is visiting this post while he's studying. Say he typically has a 3.5 GPA in most of his subjects. Let's assume his major is biology. Say this semester he gets a 2.2 in English, a 2.5 in Spanish, a 2.5 History, a 3.2 in Biology and a 2.0 in water color painting. Is he going to say he had an off semester? Or is he going to feel real good about himself because he was just slightly down from his usual GPA in his major?

 

And your example here displays your core misunderstanding of what ORS is trying to communicate.

 

Here's the real analogy: let's define an academic error as a complete brain fart, such as being the only student in the class to miss a particular question. Let's say that ksushi makes two such errors in a typical semester, but that he made six this semester. Was it his worst semester?

 

I dunno. :dunno:

 

It could be that ksushi was so far ahead of his class that his perspectives exceeded those of his classmates, and that he was reading too much into those six questions. Maybe he got a 4.00 semester regardless. The GPA is what matters, and while six questions per semester might affect the GPA, the other several hundred questions are overwhelmingly more important than the six "stupid error" questions.

 

Likewise, errors don't make a fielder's GPA. The fielder's ability to turn balls in play into outs is what matters.

 

Hope that this helped.

[cam780]

And your example here displays your core misunderstanding of what ORS is trying to communicate.

 

Stat geeks have a tendency to over think sometimes. My GPA example was just to illustrate that Lowell had an off year in just about every other defensive measure other than RZR it wasn't meant as a comment on either stat.

 

This conversation has gone on way too long at this point. The bottom line is you can't trust either stat as the be and end all and yes maybe baseball analysts prefer RZR maybe it is the defensive measure that all stat guys lust for but personally I don't watch baseball as an analyst. I'm not sitting in Theo's seat so most of what I know is from I what I see and how I interpret it. What I saw was a down year for Lowell and most of the stats support it and like I said previously if you asked Lowell or Tito if he defensively did what he expected to do I don't think he'd respond "hell yeah, my RZR was the second best ever" and thats what it boils down to.

 

As far as stubbornness goes, maybe I'm incorrect but I've never said your opinion was wrong or indefensible or untenable or unjustifiable. I respect that you have a different opinion. Or maybe thats the arrogance instead of stubbornness?

[ORS]

Yeah, you respect my opinion enough to call me a geek for it. How arrogant of me to not notice how magnanimous you've been.

[DUSTINMOHR4LIFE]

What is a fielder trying to do? Make outs. The stat you prefer' date=' in the case of a 3B, ignores about 30% of the balls the have an opportunity to turn into outs and only tells you about the fielder's failure on about 5% of them. The other stat, mind you I said RZR and OOZ are a paired stat so continuing to state that RZR is more meaningful only when you look at OOZ makes it look like you have a severe reading comprehension deficiency, RZR/OOZ ignores none of these. I can't state it any clearer than that. You'd rather look at 70% data over 100%. Indefensible.[/quote']

 

 

That should end the debate. Well said.

[Jayhawk Bill]

Stat geeks have a tendency to over think sometimes. My GPA example was just to illustrate that Lowell had an off year in just about every other defensive measure other than RZR it wasn't meant as a comment on either stat.

 

This conversation has gone on way too long at this point. The bottom line is you can't trust either stat...

 

1) I notice that you've resorted to name-calling.

 

2) Your GPA example was either intentionally misleading or ignorant. It illustrated nothing--that's why I posted that it displayed a core misunderstanding. Your protest suggests that your misunderstanding has not been corrected.

 

3) "Off year?" "Every other stat?" Let's check Lowell's FRAR by year over his entire career:

 

[table]AGE | YEAR | FRAR

24 | 1998 | 0

25 | 1999 | 10

26 | 2000 | 17

27 | 2001 | 33

28 | 2002 | 27

29 | 2003 | 9

30 | 2004 | 26

31 | 2005 | 27

32 | 2006 | 34

33 | 2007 | 27[/table]

 

Looks as if 2007 was a tie for third-best season by Lowell by FRAR...hardly an off year.

 

4) Why do you insult me and then try to say that the "conversation has gone on way too long?"