Defensive Metrics

[Johnny Footstool]

At the same time on extra hit ever 5.6 games means what in terms of OBP and SLG? 27 hits per 150 games. What is that worth to a players offensive line? No matter if you slice it either way, its a lot of extra hits being allowed.

A more simplistic view:

Dunn averages 641 Plate Appearances and 269 Total Bases per year on offense.

If we assume -1.25 TB per missed play on defense, that gives us about -25 Total Bases, which means his defense reduces his offensive value by about 9%, right?

[dougdirt]

A more simplistic view:

Dunn averages 641 Plate Appearances and 269 Total Bases per year on offense.

If we assume -1.25 TB per missed play on defense, that gives us about -25 Total Bases, which means his defense reduces his offensive value by about 9%, right?

Not quite. Your math is assuming HR's, walks, HBP's, sac bunts and strikeouts. Defense doesn't (at least in left field because there will be no sac bunts to the left fielder).
What you need to do if figure out the total BIP he averages, which will be nowhere near 641. From 2005-2009 we are looking at an average of 324 BIP.

[RedsManRick]

A more simplistic view:

Dunn averages 641 Plate Appearances and 269 Total Bases per year on offense.

If we assume -1.25 TB per missed play on defense, that gives us about -25 Total Bases, which means his defense reduces his offensive value by about 9%, right?

A zero baseline is inappropriate. It's not Adam Dunn versus an empty hole in the lineup and in LF. It's Adam Dunn versus what you would expect from somebody else in the lineup and in LF.

[westofyou]

Defensive metrics in today's world

[jojo]

Using WAR, Dunn's bat was worth 56 marginal runs above replacement (36 above average and the difference between average and replacement is roughly 20) last season. A replacement level bat was worth 0 marginal runs but hidden in the value was the difference between the marginal run bar and zero which is about 57 additional runs given Dunn's playing time (i.e. those are the runs Dunn produced that one would expect any freely available bat to produce on average).

So his total offense was worth about 110 to 115 runs.

UZR suggests his defense was last year was worth -36 runs (clearly UZR graded him as giving up more than 25 total bases last season-Dewan's agrees with UZR).

So his defense in this scenario would reduce his overall marginal value by 64% and his overall production by 30%.

Assuming his career UZR as a corner outfielder is more reflective of his true skill (-15), given an offensive year like he had last season, his glove would decrease his marginal value by roughly 30% and his overall value by roughly 15%.

[jojo]

Defensive metrics in today's world

There are too many teams using them as part of their decision making process for that to not be hyperbole.

[westofyou]

There are too many teams using them as part of their decision making process for that to not be hyperbole.

Of course it's hyperbole, but i bet every team has unique method too, so are they all right or some wrong?

Once a king had six blind men gathered together to examine an elephant.

"When the blind men had each felt a part of the elephant, the king went to each of them and said to each: 'Well, blind man, have you seen the elephant? Tell me, what sort of thing is an elephant

The six blind men assert the elephant is either like a pot (the blind man who felt the elephants' head), wicket basket (ear), ploughshare (tusk), plough (trunk), granary (body), pillar (foot), mortar (back), pestle (tail) or brush (tip of the tail).

The men cannot agree with one another and come to blows over the question of what an elephant really is like, and this delights the king. The Buddha ends the story of the king and compares the six blind men to preachers and scholars who are blind and ignorant and hold to their own views: "Just so are these preachers and scholars holding various views blind and unseeing.... In their ignorance they are by nature quarrelsome, wrangling, and disputatious, each maintaining reality is thus and thus." The Buddha then speaks the following verse:

O how they cling and wrangle, some who claim
For preacher and monk the honored name!
For, quarreling, each to his view they cling.
Such folk see only one side of a thing.[2]

[jojo]

Of course it's hyperbole, but i bet every team has unique method too, so are they all right or some wrong?

Once a king had six blind men gathered together to examine an elephant.

"When the blind men had each felt a part of the elephant, the king went to each of them and said to each: 'Well, blind man, have you seen the elephant? Tell me, what sort of thing is an elephant

The six blind men assert the elephant is either like a pot (the blind man who felt the elephants' head), wicket basket (ear), ploughshare (tusk), plough (trunk), granary (body), pillar (foot), mortar (back), pestle (tail) or brush (tip of the tail).

The men cannot agree with one another and come to blows over the question of what an elephant really is like, and this delights the king. The Buddha ends the story of the king and compares the six blind men to preachers and scholars who are blind and ignorant and hold to their own views: "Just so are these preachers and scholars holding various views blind and unseeing.... In their ignorance they are by nature quarrelsome, wrangling, and disputatious, each maintaining reality is thus and thus." The Buddha then speaks the following verse:

O how they cling and wrangle, some who claim
For preacher and monk the honored name!
For, quarreling, each to his view they cling.
Such folk see only one side of a thing.[2]

I'd hardly characterize those in FO's who think enough of the approach/data to include it in their decision making process as quarrelsome. It's a pretty conservative business and to have defensive metrics make the dent they have suggests proponents of such metrics aren't dogmatic.

[TheNext44]

I think that the current, new defensive stats are actually more accurate than the offensive stats that we use. I know it's crazy, but here's my argument.

The defensive stats are based on someone watching each play and determining how well that defensive player played on that play. We don't have that for offense.

We are actually using the opinion of the local official scorer to determine the offensive stats. And those guys, with all due respect, are a joke. A hit is not a hit, until the official scorer rules it a hit. There are a lot of easy ones, but also enough tough ones that they clearly get wrong that it makes a real difference in a batters stats.

On average around 5% of all plays are ruled errors. There are probably half as many that should be ruled errors, but aren't because of the ridiculous rule that no error can happen when no one touches the ball. That's around 7% of a batters BIP that I think are clearly in question. Let's say an official scorer gets 2/3 of them right, which is very generous imo, that would mean that around 8 hits a year for each batter are called incorrectly, assuming around 325 BIP a season. That may not seem like much, but that the difference between a .333 OBP and a .350 OBP.

Now one thing to consider is that maybe the missed scoring calls will even themselves out, just like the bloops that fall in, even out the line drives caught. But that's the crux of my point. We don't know that is true. It probably is for some players but not for others. We are basing our offensive stats on this notion that the missed scoring calls, the line drives caught, the bloops that fall in, will all magically even out over time, equally for each player.

But with the defensive metrics, we don't have to assume that, as we have someone who judges each and every play individually and decides exactly how the fielder should be judged on each and every play. To me that seems to be more accurate than relying on official scorers and the luck of averages.

[mth123]

I agree that the translation of defensive skill into metrics scaled on runs, wins, dollars, etc. is considerably less accurate at this point than offensive skills. The problem is, because the defensive metrics aren't as accurate, there's an impulse to ignore them or punt the question. But total valuations can't be done without considering defense. At some point in the valuation process, defense has to be interjected using the same scale that's used to evaluate offense. Even if we refuse to put a specific number on it, we're going to do it implicitly.

I'd rather it stay implicit. Once we start publishing numbers that are no better than a SWAG, too many people treat them as gospel. I'd prefer to do my own SWAGs.

[Johnny Footstool]

Not quite. Your math is assuming HR's, walks, HBP's, sac bunts and strikeouts. Defense doesn't (at least in left field because there will be no sac bunts to the left fielder).
What you need to do if figure out the total BIP he averages, which will be nowhere near 641. From 2005-2009 we are looking at an average of 324 BIP.

Why would you ignore part of his offensive contributions simply because his defense can't influence those same categories? That makes no sense.

I'm talking about every opportunity Dunn has to make a play, on offense or defense. The guy provides X total bases on offense, and costs X total bases on defense. Shouldn't it be that simple?

[dougdirt]

Why would you ignore part of his offensive contributions simply because his defense can't influence those same categories? That makes no sense.

I'm talking about every opportunity Dunn has to make a play, on offense or defense. The guy provides X total bases on offense, and costs X total bases on defense. Shouldn't it be that simple?

I assumed your math used TB/PA above was his TB/PA on average, which really doesn't work on the defensive side of the ball because of the things I listed. That is all I was getting at in regards to your post. Unfortunately I misread what you said.

It should be simple in the TB category, but the number you got for TB per hit wasn't quite working because of the data I thought you had used.

I am curious as to the TB per hit to left field though....

[Ron Madden]

If you shake a tree, ten gloves will fall out — but no bats.

Rogers Hornsby

The Minor Leagues are full of good defensive players-if they could hit they would be playing in the Major Leagues.

Ron Madden

[Ron Madden]

I believe all baseball fans appreciate good defense.

It's the determination of runs scored against individual fielders that I have no faith in.

I'm not being rude here, I sincerely want to understand how this works. I'd just like to see some/or more factual evidence before I buy into it.

Like I said in the first post it's an honest question.

[blumj]

The Minor Leagues are full of good defensive players-if they could hit they would be playing in the Major Leagues.

Ron Madden

The minors are pretty full of guys who can hit but can't field, too. The system seems to be working fairly efficiently that way, Jack Cust and Adam Everett both still have jobs, but the line on each side of the ball is probably almost right behind them. If you're not about as good at your one side of the ball as those two are at theirs, you're going to have a tough time getting in or sticking. And, for both entertainment value and competition level, I'd say that's a decent place for the line to be.