Defensive Efficiency

[scottsdale_joe]

by Hy Durham
Baseball Prospectus
Special to the Vancouver Sun
May 11, 2000

Given: "Defense is Crucial to Success"
So What Exactly is Defensive Efficiency?

Defense is the often overlooked or frequently ignored factor that is critical to a team’s success. One way to measure a team’s effective on defense is by using the Defensive Efficiency statistic. This is sometimes called Defensive Average.

Def Eff, or Defensive Efficiency, is the rate at which balls put into play are converted into outs by a team's defense. Def Eff can be approximated with (1 - BABIP) if that is all you have available. The accurate way to compute a team's Def Eff is by using the following formula:

1 - ((H + ROE - HR) / (PA - BB - SO - HBP - HR))

Definitions:
H = hits allowed
ROE = batters who reach first base as a direct result of a fielding error
HR = Home runs allowed
PA = Plate Appearances
BB = Bases on balls (or walks)
SO = Strikeouts
HBP = Batters hit by pitch

This is very obviously a team statistic, and it is also obviously as highly affected by the quality of a team’s pitching as it is by a team’s defensive prowess.

Managers in the MBBA can see the Def Eff of their team in the bottom left corner of the manager screen in OOTP and for all teams on the Fielding Report . Currently Vancouver has the best Def Eff rating in the MBBA at .704. That’s the only rating that breaks .700. Second best is California at .699. The worst two teams in Def Eff are Long Beach and Des Moines, both at .639.

Zone Rating is another statistic that perhaps we will discuss in another column. The best team zone rating doesn’t necessarily translate into the best Def Eff. For example, Valencia, Baltimore, Washington, and Calgary all have a higher team ZR than Vancouver's 8.8. In the Frick League, the New Orleans Crawdads have the best Def Eff (.688) and yet the team ZR is -6.3.

[lynchy34]

The way I've built this team is so uncharacteristic of how usually build my clubs. And it's killing me inside.

[scottsdale_joe]

The way I've built this team is so uncharacteristic of how usually build my clubs. And it's killing me inside.

Your trade block post combined with my surprise at having the best Def Eff in the Johnson League caused me to write this.
My players in general are not great fielders and don't have great range, so I was happily surprised at having the best Def Eff.
I'm still trying to understand it all.
Another 20 years and I'll have it all down. (Oops, I'll probably be dead by then.)

[KCCardinalFan]

This is good stuff, thank you. I have been watching my BABIP the last couple of seasons, and it is way up this year although I only think I have one real defensive turd (Joe Sullivan) playing. Those fluctuations are true to real life, just frustrating!

[KCCardinalFan]

Just to follow up, is this info available in Statslab or just in-game?

[scottsdale_joe]

Just to follow up, is this info available in Statslab or just in-game?

Off hand, I don't know.
I use StatsLab sparingly (mainly because my main other league does not have it).

[jumpmancol]

I think the combination of Defensive Efficiency and Zone Rating is the best way to judge a teams overall defensive ability. One without the other to me is meaningless. I want guys that not only cover a lot of ground, but also do a good job of covering that ground, and do it efficiently. Make sense?

[jcrmoon42]

From the manual.

"Zone Rating" - developed by John Dewan, A player gets credit (a "plus" number) if he makes a play that at least one other player at his position missed during the season, and he loses credit (a "minus" number) if he misses a play that at least one player made. The size of the credit is directly related to how often players make the play. Each play is looked at individually, and a score is given for each play. Sum up all the plays for each player at his position and you get his total plus/minus for the season. A total plus/minus score near zero means the player is average. A score above zero is above average and a negative score is below average.

I haven't been able to find anything regarding how efficiency is calculated for an individual player, unless it is somehow connected to Zone Rating. That is a team stat, and I've never seen it used for individuals until OOTP.

[jcrmoon42]

Looking over my team's individual fielding stats, it appears that Zone and Eff have to be connected. 1.000 would indicate an average defender, which must have to do with making the plays that would be expected on balls hit in your zone. Whether a .800 efficiency means you made 80% of the plays in your zone or not is not explained anywhere, but it would make sense. However, that is an awfully low number, and I'm dubious as to whether or not it is an actual percentage.

As to the assertion that one is useless without the other, I'm not sure what that even means. Both include components of range and efficiency. They are just stated differently. One is a raw number. The other seems to be a percentage. They are both based on how many balls you get to and make a play on compared to how many an average player at that position gets to and makes a play on.

[cheekimonk]

From the manual.

"Zone Rating" - developed by John Dewan, A player gets credit (a "plus" number) if he makes a play that at least one other player at his position missed during the season, and he loses credit (a "minus" number) if he misses a play that at least one player made. The size of the credit is directly related to how often players make the play. Each play is looked at individually, and a score is given for each play. Sum up all the plays for each player at his position and you get his total plus/minus for the season. A total plus/minus score near zero means the player is average. A score above zero is above average and a negative score is below average.

I haven't been able to find anything regarding how efficiency is calculated for an individual player, unless it is somehow connected to Zone Rating. That is a team stat, and I've never seen it used for individuals until OOTP.

ZR is a relative metric used as an attribute in OOTP (i.e., it follows a player from year to year and "predicts" his performance). Efficiency, at least in MLB and I would assume OOTP, is what percentage of plays the player made of those he "should" have made (and that's not the same as number of errors). It's not relative but specific to him and it's also a stat not a "metric" (akin to batting average and not ZR).

The problem that game designers have (from Strato to OOTP) is reverse engineering an existing ZR (or any other "Saber" - quotes intended - metric, really) for a certain profile player into impacts on specific games and then ensuring it's relatively consistent at the end of the year to what it was intended to be. We know, for instance, the distribution of ZRs in MLB for 2010 (or 2009, 2008, etc.)...easy numbers to calculate. The hurdle is creating a semi-realistic, fictional league with a similar distribution of ZRs and then - as I said - reverse engineering the metric in performance. Depending on how OOTP did that in v.11 would impact the usability of the rating for an individual player (or an infield/outfield) and predicting their efficiency.

So, from what I can tell at least, ZR relates to a player's defensive efficiency as the Contact/Eye/K ratings relate to batting average.

[jcrmoon42]

We also have to be clear about the distinction between teams and individuals.

Teams - Zone rating is NOT just the added up zone ratings of all of the individual players. It appears to be some manner of weighted zone rating that gives more weight to the middle of the defense. However, it is still going to be heavily affected by a terrible player at a lesser position such as left field or first base. I'm not sure we can actually call it a useful stat without knowing exactly how it is calculated. Efficiency, on the other hand, we know exactly what to do with. What percentage of balls put in play does your team turn into outs? Easy peasy. Clean and efficient.

Individuals - Zone rating is useful in that it compares individuals at the same position across the league. It would appear that efficiency for an individual correlates to this and is affected by it. We don't know that for sure, however. In this case, it would appear that zone rating is more useful without knowing what efficiency actually means.

Without the manual being more clear, we are left somewhat in the lerch.

[jcrmoon42]

ZR is a relative metric used as an attribute in OOTP (i.e., it follows a player from year to year and "predicts" his performance). Efficiency, at least in MLB and I would assume OOTP, is what percentage of plays the player made of those he "should" have made (and that's not the same as number of errors). It's not relative but specific to him and it's also a stat not a "metric" (akin to batting average and not ZR).

I've never heard the term "efficiency" used relating to an individual outside of OOTP. It is always used as a team stat. That's why I'm thinking individual efficiency has to be tied to ZR, because ZR tells us the number of plays a player "should" make. There is no way to tell what he should make without comparing him to other players. Considering that, doesn't it have to be relative? If you look at any individual player, efficiency is above 1.000 when the ZR is positive and vice versa.

I may be missing what you are saying regarding ZR following a player and predicting performance.

[scottsdale_joe]

Just for the record, in the OP I was only using Def Eff as it relates to a team, not to an individual.

[jcrmoon42]

Just for the record, in the OP I was only using Def Eff as it relates to a team, not to an individual.

Yah, that's why I went back and distinguished. I realized I had changed the discussion and probably caused even MORE confusion.

[cheekimonk]

ZR is a relative metric used as an attribute in OOTP (i.e., it follows a player from year to year and "predicts" his performance). Efficiency, at least in MLB and I would assume OOTP, is what percentage of plays the player made of those he "should" have made (and that's not the same as number of errors). It's not relative but specific to him and it's also a stat not a "metric" (akin to batting average and not ZR).

I've never heard the term "efficiency" used relating to an individual outside of OOTP. It is always used as a team stat. That's why I'm thinking individual efficiency has to be tied to ZR, because ZR tells us the number of plays a player "should" make. There is no way to tell what he should make without comparing him to other players. Considering that, doesn't it have to be relative? If you look at any individual player, efficiency is above 1.000 when the ZR is positive and vice versa.

I may be missing what you are saying regarding ZR following a player and predicting performance.

If you look at the definition again, ZR is calculated relative to other players. If I miss a play, but everyone in the league missed that play, then it doesn't hurt my ZR. Then again, since it's relative, it's still a "metric" and not a "stat" so it's really only useful in comparing "how good my shortstop is projected to be compared to all shortstops" and not how good he actually is.

I do see what you are saying about defensive efficiency being a team stat...not sure how that is assigned to individuals.