Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Thursday, May 30, 2019

Not bad for a summer job, but are baseball’s ‘advanced defensive metrics’ accurate?

Slog through the fine print of advanced baseball metrics and you often find a passage that says, Yeah, these numbers might not be accurate, but it’s the best information we’ve got, so let’s go with it.

Which is pretty much the same thought process that led to the belief that the Earth was flat, the sun orbited around it and a good way to identify a witch was to tie her up and throw her in the nearest lake to see if she sank.

“Sure, we might be ‘wrong’ and drown a few people, but it’s the best system we’ve got.”

Here’s the problem with that: When we accept and use these numbers without question we give them validity, and players are now being judged by metrics that even the metrics’ advocates admit may not be accurate.

I’d trust the opinion of a coach or scout who has watched thousands and thousands of baseball games and tens of thousands of players far more than the opinion of a “video scout” watching a game on TV, making $8 an hour, doing a summer job and feeding what may or may not be accurate data into a flawed system.

So why do we continue using these numbers?

Because we want to have opinions without going to the trouble of developing expertise.

RoyalsRetro (AG#1F) Posted: May 30, 2019 at 10:23 AM | 58 comment(s) Login to Bookmark
  Tags: advanced metrics

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Itchy Row Posted: May 30, 2019 at 10:44 AM (#5846758)
Because we want to have opinions without going to the trouble of developing expertise.
Like writing a column?
   2. . Posted: May 30, 2019 at 11:46 AM (#5846797)
Like writing a column?


Columns have never purported to be expert opinion, just opinion.

And the quoted part of this column is 100% accurate. "Yeah, our model sucks, but it's the best we have so we have to go with it" isn't expertise, it's ignorance.
   3. BrianBrianson Posted: May 30, 2019 at 11:53 AM (#5846803)
We've known the earth was vaguely spherical for at least 2500 years. Before that, flat was an excellent approximation for basically every purpose.
   4. geonose Posted: May 30, 2019 at 12:12 PM (#5846807)
I clicked through to see who at the KC Star wrote the column. When I discovered it was Lee Judge, my first thought was "Yeah, disregard." He is not a sports columnist, but an editorial cartoonist. He also fancies himself a baseball expert who once had a blog called "Judging the Royals" in which he exposed himself as an avowed anti-sabermetrician.
   5. flournoy Posted: May 30, 2019 at 12:35 PM (#5846811)
"Garbage in, Garbage out" is as true today as it ever was. On that point, the author is completely correct. Whether that applies to modern defensive metrics is an open question, I suppose.
   6. TomH Posted: May 30, 2019 at 01:00 PM (#5846821)
Are the metrics more accurate than (a) the old metrics [as in "who has the fewest errors, he gets the gold glove"] or (b) the scouts/fans who said "that guys has a great swing/style, he is good"?

Probably in many cases.

Are they better than this writer in judging ballplayers? Tee hee.
   7. David Nieporent (now, with children) Posted: May 30, 2019 at 01:16 PM (#5846824)

I’d trust the opinion of a coach or scout who has watched thousands and thousands of baseball games and tens of thousands of players far more than the opinion of a “video scout” watching a game on TV, making $8 an hour, doing a summer job and feeding what may or may not be accurate data into a flawed system.
"Opinion"?
   8. Bring Me the Head of Alfredo Griffin (Vlad) Posted: May 30, 2019 at 01:28 PM (#5846827)
Columns have never purported to be expert opinion, just opinion.


If there weren't an implicit claim that columnists' opinions were better than regular people's, why would readers/viewers be expected to pay for them?
   9. Rally Posted: May 30, 2019 at 01:38 PM (#5846832)
Not going to say anything about scouts, but the collective opinion of coaches gave us Derek Jeter, 5 time gold glove winner.
   10. . Posted: May 30, 2019 at 01:42 PM (#5846836)
If there weren't an implicit claim that columnists' opinions were better than regular people's, why would readers/viewers be expected to pay for them?


Not even sure what "better" means here. A "better" opinion? What does that mean?

As to what you appear to be asking regarding reader payment: There are all number of rhetorical purposes other than communicating and transmitting expertise, and many of these rhetorical purposes are entirely valid and worthy.
   11. Jeremy Renner App is Dead and I killed it Posted: May 30, 2019 at 01:53 PM (#5846841)
Never really understood the hostility to trying to understand something and keeping at trying. But then I don't hang with people who don't question anything. Comforting though I suppose.
   12. David Nieporent (now, with children) Posted: May 30, 2019 at 02:16 PM (#5846845)

Not even sure what "better" means here. A "better" opinion? What does that mean?
One that is more well-reasoned and based on more accurate underlying facts.
   13. What did Billy Ripken have against ElRoy Face? Posted: May 30, 2019 at 02:22 PM (#5846849)
based on more accurate underlying facts
Oh, man, nice retro reference. I totally remember back when "facts" were a thing. Those were the days.
   14. Brian C Posted: May 30, 2019 at 02:27 PM (#5846852)
I’d trust the opinion of a coach or scout who has watched thousands and thousands of baseball games and tens of thousands of players far more than the opinion of a “video scout” watching a game on TV, making $8 an hour, doing a summer job and feeding what may or may not be accurate data into a flawed system.

So why do we continue using these numbers?

Because we want to have opinions without going to the trouble of developing expertise.

I don't know how much I trust modern defensive numbers, either - depending on the specific stat, I guess I think of them somewhere between "rough measures" and "generally reliable". So I understand some skepticism and even think it's healthy. But there's something truly insidious about essentially saying, "the math might be wrong so I'm just going to blindly trust conventional wisdom." As noted above, the coaches and scouts have more than their fair share of collective blunders too, which is what drove the search for better defensive metrics in the first place.

I suppose what I'm saying is that I can't stand when people pat themselves on the back for their own ignorance.
   15. . Posted: May 30, 2019 at 02:37 PM (#5846860)
But there's something truly insidious about essentially saying, "the math might be wrong so I'm just going to blindly trust conventional wisdom."


That's actually quite a bit smarter than going with knowingly wrong math. Being blindly anti-conventional wisdom (*) just for the sake of it is actually really stupid.

(*) I.e., the collected iterative wisdom of lived human experience.

I suppose what I'm saying is that I can't stand when people pat themselves on the back for their own ignorance.


Then you should positively detest saber fanatics.
   16. BrianBrianson Posted: May 30, 2019 at 02:53 PM (#5846864)
The conventional wisdom was probably doing a smidge better than coin flips or chicken entrails, which is very high in WAR (Wisdom Above Replacement).
   17. strong silence Posted: May 30, 2019 at 03:05 PM (#5846867)
Saw Jeopardy the other night. Alex Trebek asked the current champion, who needs another 200K +/- to beat the All-Time Earnings Record, a question about the state of saber metrics in MLB. Holzhauer replied that in his opinion the teams were doing a good job of balancing saber/stat approaches with conventional, traditional measures.

Holzhauer also said he wants a job with an MLB team.

Being so focused on data/facts/trivia, he doesn't strike me as one with the analytical tools to provide a competitive advantage for any MLB tea.

   18. Baldrick Posted: May 30, 2019 at 03:09 PM (#5846870)
Slog through the fine print of advanced baseball metrics and you often find a passage that says, Yeah, these numbers might not be accurate, but it’s the best information we’ve got, so let’s go with it.

Which is pretty much the same thought process that led to the belief that the Earth was flat, the sun orbited around it and a good way to identify a witch was to tie her up and throw her in the nearest lake to see if she sank.

What a doofus.
   19. Never Give an Inge (Dave) Posted: May 30, 2019 at 03:28 PM (#5846883)

I’d trust the opinion of a coach or scout who has watched thousands and thousands of baseball games and tens of thousands of players far more than the opinion of a “video scout” watching a game on TV, making $8 an hour, doing a summer job and feeding what may or may not be accurate data into a flawed system.

I think some skepticism of defensive metrics is warranted, but there needs to be some balance. To the extent MLB coaches and scouts have actually seen tens of thousands of players, how many times have they really seen most of them? One reason we rely on a balance of scouting and statistics is that the scouts/coaches often have to form judgments based on a very small sample size.

So metrics that claim to measure a player’s defensive ability actually have little or no information about how far a player had to run to get to a batted ball.

The metrics don't claim to measure defensive *ability*, do they? They claim to retrospectively measure defensive outcomes and value.
   20. Fernigal McGunnigle Posted: May 30, 2019 at 03:42 PM (#5846893)
Man, I read TFA and it is so ####### stupid.
Yeah, these numbers might not be accurate, but it’s the best information we’ve got, so let’s go with it.

Which is pretty much the same thought process that led to the belief that the Earth was flat, the sun orbited around it and a good way to identify a witch was to tie her up and throw her in the nearest lake to see if she sank.
It's actually the thought process that got us pretty much all of modern science and technology! If an author doesn't know jack #### about intellectual history he shouldn't try to talk about it.
   21. . Posted: May 30, 2019 at 03:54 PM (#5846897)
It's actually the thought process that got us pretty much all of modern science and technology! If an author doesn't know jack #### about intellectual history he shouldn't try to talk about it.


Right, but the point is more that at one point in intellectual history "the sun orbits around the Earth" was "the best information we've got." From which it follows that it's not necessarily smart to simply default to "the best information we've got."

This isn't really that hard to understand, is it? The lesson isn't then to just dispose with intellect, but it is to be humble about its current state -- particularly when the reasons for that humility are right there in front of everyone's face.
   22. What did Billy Ripken have against ElRoy Face? Posted: May 30, 2019 at 03:58 PM (#5846899)
a good way to identify a witch was to tie her up and throw her in the nearest lake to see if she sank
Have we found a significantly better way? I'm not really up to date on the latest witch analytics.
   23. . Posted: May 30, 2019 at 03:58 PM (#5846900)
The metrics don't claim to measure defensive *ability*, do they? They claim to retrospectively measure defensive outcomes and value.


Do they even measure value? In what units of measurement?(*)

I'm not sure any baseball data actually measures ability. Baseball data can certainly be used to infer ability.

(*) Those units will not be the same as the units to measure hitting value, as many on the board have previously noted.
   24. JL72 Posted: May 30, 2019 at 04:02 PM (#5846902)
That's actually quite a bit smarter than going with knowingly wrong math.


Except that is not what is being done. Recognizing the potential limitations in a data set is a far cry from knowing that the math is wrong and using it anyway.
   25. . Posted: May 30, 2019 at 04:03 PM (#5846904)
Recognizing the potential limitations in a data set is a far cry from knowing that the math is wrong and using it anyway.


It's not limitations in the data set; it's that the measurements of the data and the modeled outputs of the data are knowingly inaccurate.
   26. BrianBrianson Posted: May 30, 2019 at 04:13 PM (#5846911)
Have we found a significantly better way? I'm not really up to date on the latest witch analytics.


Yeah, you really gotta control for lake temperature and salinity. Otherwise those living next to the Dead Sea or the Garabogazkol Basin end up thinking everyone is a witch.

It's really about indoor pools, where these things can be closely controlled.
   27. TJ Posted: May 30, 2019 at 04:31 PM (#5846918)
Holzhauer also said he wants a job with an MLB team.


He keeps winning and he might get one called "owner"...
   28. PreservedFish Posted: May 30, 2019 at 05:06 PM (#5846931)
Man, articles like this used to be the bread and butter of this site.
   29. Fernigal McGunnigle Posted: May 30, 2019 at 05:11 PM (#5846933)
This isn't really that hard to understand, is it? The lesson isn't then to just dispose with intellect, but it is to be humble about its current state -- particularly when the reasons for that humility are right there in front of everyone's face.
The author of TFA gets really upset when he reads all of the caveats that the Fielding Bible and Fangraphs give about their data. I.e., when the sabermetric position shows humility, the author takes it as evidence that they aren't showing humility. This isn't really that hard to understand, is it? The author is either a ####### moron, or isn't trying very hard to be intelligent.
   30. Rally Posted: May 30, 2019 at 05:27 PM (#5846938)
Outside of players who make the big leagues and top executives, James has already made more money in his streak than the typical baseball employee makes in a career.
   31. DFA Posted: May 31, 2019 at 02:02 AM (#5847038)
I'm certainly among the skeptics of some of the freely available defensive stats, and thus using them in wins above replacement also gives me pause. But I would also think that actual GMs have access to information that is likely more reliable and valid than what terrific sites like fangraphs and bb-ref have access to.
   32. manchestermets Posted: May 31, 2019 at 04:49 AM (#5847041)
If there weren't an implicit claim that columnists' opinions were better than regular people's, why would readers/viewers be expected to pay for them?


Because they write well. The best columnists are the ones that you still want to read even when you think what they're saying is idiotic.
   33. Fancy Crazy Town Banana Pants Handle Posted: May 31, 2019 at 07:23 AM (#5847043)
Man, articles like this used to be the bread and butter of this site.

I think you mean articles like this used to be the bread and butter of FJM.
   34. ERROR---Jolly Old St. Nick Posted: May 31, 2019 at 08:28 AM (#5847050)
Question: How do the advanced defensive metrics factor in shifts and positioning cards? Do they give a built-in advantage to current players over players who didn't have the advantage of that information?
   35. ERROR---Jolly Old St. Nick Posted: May 31, 2019 at 08:31 AM (#5847051)
The best columnists are the ones that you still want to read even when you think what they're saying is idiotic.

Even better are the ones who can sometimes make you reconsider some of your pre-existing opinions, or bring up information that you didn't know, or perspectives you hadn't previously considered.
   36. Rally Posted: May 31, 2019 at 08:43 AM (#5847052)
Question: How do the advanced defensive metrics factor in shifts and positioning cards? Do they give a built-in advantage to current players over players who didn't have the advantage of that information?


1. Not sure exactly what they do about shifts. I know Baseball Info Solutions tracks shifts and takes them into consideration, but exactly what they do at this point in time, I don't know.

2. No advantage whatsoever. The sum of defensive runs saved for all 2019 shortstops should be exactly the same for all 1959 shortstops: zero. The only advantage you might find is that players on a team that uses the shift well in 2012 might have an advantage over teams that don't shift that year. In 2019 everybody shifts, but maybe some teams are better than others at deciding when to do it and where to put fielders, so their fielders might look better than fielders on teams that are not great at this stuff.
   37. Steve Parris, Je t'aime Posted: May 31, 2019 at 08:59 AM (#5847055)
What this guy is missing is the lack of bias with the video scouts coding balls in play. The judgments will be off sometimes but they will not favor certain fielders, given large enough of a sample. Whereas the confirmation bias whenever Jeter would make a good play reinforced the myth that he was a great shortstop.
   38. Bug Selig Posted: May 31, 2019 at 09:17 AM (#5847058)
Do the numbers say Raffy was a great 1B in 1999? Oops, there goes this guy's appeal to authority argument.
   39. . Posted: May 31, 2019 at 09:30 AM (#5847059)
The sum of defensive runs saved for all 2019 shortstops should be exactly the same for all 1959 shortstops: zero.


That's one way to do it, but it's merely an assumption -- not something baked into the cosmos. We don't do that for hitting or pitching for example.(*) Which means at that point, defense is being measured in different units than hitting or pitching and therefore can't be summed with them in any way that really makes sense.

(*) Defense is being measured by comparison to what peers would have done with the same batted ball. We don't look at a double with that level of granularity or comparison with what peers would have done with the same pitch -- it's just a double. Double on a shitty BP-caliber 3-1 pitch right down Broadway -- a double. Double on a 90 slider on the black on an 0-2 count after you've taken a couple pitches for the team and fouled off some tough ones -- a double.
   40. Rally Posted: May 31, 2019 at 09:32 AM (#5847060)
Who should have won that gold glove? The top TZ belongs to Tino Martinez. He was a fine defender but never got a GG award. He'll have to settle for having earned True Yankee status while winning 4 rings in his 6 seasons with the team.

Next best TZ candidate would have been Darin Erstad, despite playing only half the season at first. He was a left fielder for the other half of his games, and had a +22 rating out there. Erstad would have been a reasonable choice for two gold gloves at 2 positions in the same season. I wonder if any player has ever had a similar circumstance before or since. Maybe Zobrist 2B/RF?

Erstad didn't actually win his first gold glove until a year later when he hit .355
   41. Rally Posted: May 31, 2019 at 09:35 AM (#5847061)
Yup, Zobrist. Going by DRS, in 2009 he saved the most runs of any RF, and was 2nd at 2B, behind only Kinsler.
   42. Never Give an Inge (Dave) Posted: May 31, 2019 at 12:27 PM (#5847100)

That's one way to do it, but it's merely an assumption -- not something baked into the cosmos. We don't do that for hitting or pitching for example.(*) Which means at that point, defense is being measured in different units than hitting or pitching and therefore can't be summed with them in any way that really makes sense.

Partially true. It's different from what we do for hitting but not really different from what we do for pitchers. However, the units are the same and there's not a problem with adding them together. But by effectively forcing all the SS or 3B or CF in a given year to sum to zero I think we may be introducing some bias into the fielding numbers. This is more of an area for potential refinement, however, and not a reason to throw the whole system out. IMO.
   43. . Posted: May 31, 2019 at 01:29 PM (#5847113)
However, the units are the same and there's not a problem with adding them together.


How so, if one is absolute (runs created) and one is versus a baseline (defensive runs saved)? They're both denominated in runs, sure, but they aren't runs as measured by the same scale. It's kind of like adding Fahrenheit and Celsius temperatures. Both measure temperature, but on different scales.

But by effectively forcing all the SS or 3B or CF in a given year to sum to zero I think we may be introducing some bias into the fielding numbers.


The bigger problem is that by this technique, it's theoretically possible for no one to save a run on defense in any given baseball season which ... can't be right.
   44. Rally Posted: May 31, 2019 at 01:44 PM (#5847119)
It's runs saved vs. average, not absolute runs saved. Even Adam Dunn probably made some good plays here and there that saved runs, but he still was well below average. Better example: Manny Ramirez. Horrible fielder, but I'm sure most of us remember him making a running catch near the wall, high fiving a fan, and doubling a runner off base.

Defensive runs saved summing to zero is no different than linear weight batting runs summing to zero for a league.

We don't look at a double with that level of granularity or comparison with what peers would have done with the same pitch -- it's just a double. Double on a shitty BP-caliber 3-1 pitch right down Broadway -- a double. Double on a 90 slider on the black on an 0-2 count after you've taken a couple pitches for the team and fouled off some tough ones -- a double.


Same with a putout. A catch of a screaming line drive, an easy popup, or catching the ball on a force out are all scored the same way - putouts. Defensive runs and batting runs both involve a comparison to the league average.

   45. Rally Posted: May 31, 2019 at 01:48 PM (#5847120)
The bigger problem is that by this technique, it's theoretically possible for no one to save a run on defense in any given baseball season which ... can't be right.


Possible only if all fielders perform exactly at the same level. That is exactly as likely as all batters having a .750 OPS or every pitcher exactly a 4.35 ERA.

That near-impossible scenario becomes impossible once you consider playing time. A batter who gets one plate appearance either gets on base or doesn't, and so must be either above average or below it. Same with a fielder who only is on the field for one chance to make a play. Either he makes it or he doesn't. No chance of his rating being average.
   46. Never Give an Inge (Dave) Posted: May 31, 2019 at 02:06 PM (#5847129)

How so, if one is absolute (runs created) and one is versus a baseline (defensive runs saved)?

The components of WAR are all measured vs. average (and then some additional runs are added to account for the difference between average and replacement value).

The difference is that for batting the baseline is the total league average (batting runs for the entire league add up to zero), and then a standard positional adjustment is tacked on (the positional adjustment is not constant over time, but I believe the changes are very gradual). So it's possible in a year where shortstops hit really well for shortstops in aggregate to accumulate positive or negative offensive value in any given year when including the positional adjustment.

However, for defensive runs, the baseline is the average for a given position. So if shortstops happen to field really well one year, their defensive runs will still aggregate to zero.

I don't think this is a major problem; as I said earlier maybe it's an area for refinement but it's also a philosophical question about what you're trying to measure with WAR.
   47. . Posted: May 31, 2019 at 03:16 PM (#5847142)
It's runs saved vs. average, not absolute runs saved.


I know -- that's the problem. A batter hits a double, he creates 0.43 (or whatever) runs without reference to any other player or any other event but the double. There's no analogue in fielding. (Or more precisely: There is, but no one pays attention to it.)

Defensive runs saved summing to zero is no different than linear weight batting runs summing to zero for a league.


Right -- both are purely conventions.

Defensive runs and batting runs both involve a comparison to the league average.


Nope. They can involve that. But the concept of batting runs created has easily-understood meaning entirely independent of any baseline or reference point of comparison.

Possible only if all fielders perform exactly at the same level.


Yes, exactly. If all major league shortstops in 2034 field exactly like Ozzie Smith, they're deemed to have saved the exact amount of runs defensively as if they had all fielded exactly like Derek Jeter. The senselessness of that needs no additional explanation.

That is exactly as likely as all batters having a .750 OPS or every pitcher exactly a 4.35 ERA.


Likelihood is not relevant to model and premise quality, but in any event, we wouldn't say the same about OPS. If everyone in the league has a 1.000 OPS, we'd say it was a big offensive year and conversely if everyone had a .500 OPS, we'd say it was a weak offensive year. We wouldn't compare them all to their median peer and say since the deviation sums to zero, they're the same. That would make no sense.
   48. . Posted: May 31, 2019 at 03:21 PM (#5847143)
The components of WAR are all measured vs. average (and then some additional runs are added to account for the difference between average and replacement value).


It's still a different scale for fielding, because the underlying premises are entirely different. That the models wind up calling the thing being measured "runs" is neither here nor there. The "runs" are still not the same thing.

Every batting event creates runs independently of comparison to baseline.(*) Yes, you can then compare the runs created to some baseline, but that's entirely different from fielding where there's no runs saved calculated or calculable other than by the comparison to baseline. (See, e.g., the Smith/Jeter example.) (**)

(*) The double Joey Gallo hits tonight will create x number of runs. Full stop. The ball Rougned Odur ranges far to his right to field will not save x number of runs, full stop.

(**) Park effects show the same feature. Thirty identical bandboxes will be deemed the same as thirty identical caverns, because the reference point is peer group rather than absolute. That's, again, only convention.
   49. Never Give an Inge (Dave) Posted: May 31, 2019 at 03:25 PM (#5847144)

Nope. They can involve that. But the concept of batting runs created has easily-understood meaning entirely independent of any baseline or reference point of comparison.

Not in stats like WAR which add defensive runs and batting runs, which is what I thought you were taking issue with in post #39. Batting runs in WAR are relative to a league average baseline.

If everyone in the league has a 1.000 OPS, we'd say it was a big offensive year and conversely if everyone had a .500 OPS, we'd say it was a weak offensive year. We wouldn't compare them all to their median peer and say since the deviation sums to zero, they're the same. That would make no sense.

I believe (but am not sure) that is actually what bWAR would do. Whether that's the right approach depends on what you're trying to measure with the metric.
   50. . Posted: May 31, 2019 at 03:30 PM (#5847148)
Not in stats like WAR which add defensive runs and batting runs,


Not in stats like WAR, but I wasn't referring to any particular stat but instead first principles. I've never suggested that there weren't metrics out there that purported to add things I've suggested can't really be effectively added.

I believe (but am not sure) that is actually what bWAR would do.


Well, then that would make no sense. I wouldn't think we'd have to re-litigate the idea that a league where everyone OPS'd 1.000 would be a big offensive league and one where everyone OPS'd .500 would be a weak one. Once we're re-litigating that one, things have gone a bit astray and it's probably time for a re-examination of premises and first principles.

More batting runs were created in 2000 than in 1968, right? So then how is "adding" those runs scaled that way, with the possibility of absolute measurement and big yearly variance, to runs "saved" on a scale measuring relative things where every year is deemed by convention to save zero runs adding the same thing? It can't be.
   51. Never Give an Inge (Dave) Posted: May 31, 2019 at 03:39 PM (#5847150)
A batter hits a double, he creates 0.43 (or whatever) runs without reference to any other player or any other event but the double.

What stat are you referring to here, out of curiosity? I believe linear weights also gives you value vs. league average, for example.
   52. . Posted: May 31, 2019 at 03:45 PM (#5847153)
What stat are you referring to here, out of curiosity?


I'm not referring to any stat. I'm referring to the concept/first principle that a double creates a certain amount of runs independent of anything else and doesn't need an external reference point to be deemed valuable.

There's no fielding analogue to this concept. And there's no fielding analogue to the simple concept that a year where everyone OPSs 1.000 is a bigger offensive year than one where everyone OPSs .500.

We can, but don't need to, refer to peers to measure offense. We have to refer to peers to measure defense. A double needs no external reference to make sense. A caught ball in centerfield does. That's the fundamental difference between the two.
   53. BrianBrianson Posted: May 31, 2019 at 03:55 PM (#5847158)
. is just trolling you. Trying to explain thing to him is a fool's game, because he has no interest in understanding.
   54. . Posted: May 31, 2019 at 04:09 PM (#5847161)
Nah, the explanation failed. Even if you baseline against average, different things are still being baselined against average. It's been successfully counter-explained. If people want to rally around WAR or whatever in response, there's nothing stopping them and the world isn't going to end. It's just baseball.
   55. strong silence Posted: May 31, 2019 at 05:07 PM (#5847180)
"We don't look at a double with that level of granularity or comparison with what peers would have done with the same pitch -- it's just a double."

"We". Who is that? The creators of WAR? It sounds like you are arguing with a fictitious BTF sabermetrician.

I gotta believe the clubs are analyzing hits to that detail, and further. Of course, those secrets won't reach these pages.
   56. Never Give an Inge (Dave) Posted: May 31, 2019 at 05:20 PM (#5847185)

I'm not referring to any stat. I'm referring to the concept/first principle that a double creates a certain amount of runs independent of anything else and doesn't need an external reference point to be deemed valuable.

I don't know what that means. A double creates a certain number of runs? So if Joey Gallo hits a double in his second at-bat, his team will score X number of runs more than if had done...something else in that at-bat(?)

Why can't you do the converse -- if Roughned Odor fields a ball that would have otherwise gone for a double, he has saved X runs?
   57. Kiko Sakata Posted: May 31, 2019 at 05:46 PM (#5847189)
I'm not referring to any stat. I'm referring to the concept/first principle that a double creates a certain amount of runs independent of anything else and doesn't need an external reference point to be deemed valuable.

There's no fielding analogue to this concept. And there's no fielding analogue to the simple concept that a year where everyone OPSs 1.000 is a bigger offensive year than one where everyone OPSs .500.


At its core, baseball statistics are a perfectly offsetting double-ledger accounting system (I think that's the right terminology). If a double is worth +0.43 runs to the offensive team, then it must, by definition, be worth -0.43 runs to the defensive team. My Player won-lost records build up from this concept where positive events are "wins" and negative events are "losses" and if an event is worth 0.43 wins to the offensive team, the same play must be worth 0.43 losses to the defensive team (and vice versa - outs are "wins" for the defense and "losses" for the offense).

The challenge with fielding vis-a-vis batting is simply that it's less obvious which specific players to charge the 0.43 losses - pitcher, fielder(s); if the latter, which specific fielders. But of course one can (in theory) measure offensive runs created and defensive runs saved on the same scale: they're both measuring the same thing - runs.
   58. . Posted: May 31, 2019 at 08:10 PM (#5847213)
The challenge with fielding vis-a-vis batting is simply that it's less obvious which specific players to charge the 0.43 losses - pitcher, fielder(s);


I won't quibble with the "less" part, but it's far less obvious on review than on first blush that the hitter should just get all the credit for the 0.43 runs. If a hitter gets a single -- let's call it .20 runs -- on a crap pitch on 3-1 where the league average production on the same count/pitch is .23 runs ... has the hitter actually even created value?

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
robneyer
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogGiants change bullpen location, move fences in at Oracle Park
(19 - 10:27am, Dec 15)
Last: ERROR---Jolly Old St. Nick

NewsblogRyan Thibs’ Hall of Fame Tracker
(468 - 10:23am, Dec 15)
Last: SoSH U at work

NewsblogSources: Astros witnesses admit sign stealing to MLB, but deny special camera | SNY
(4 - 10:20am, Dec 15)
Last: The Duke

NewsblogNationals 'love' Kris Bryant but potential holdup could stymie trade talks
(1 - 10:14am, Dec 15)
Last: The Duke

NewsblogThe Marlins are talking to Yasiel Puig
(6 - 10:13am, Dec 15)
Last: PreservedFish

NewsblogMinors: MLB's proposal a 'shell game' to yank 42 teams out of their communities - The San Diego Union-Tribune
(8 - 9:47am, Dec 15)
Last: Sunday silence

NewsblogVideo: Josh Donaldson Bought His Mom a Maserati After She Quit Smoking on a Bet
(10 - 9:05am, Dec 15)
Last: Swoboda is freedom

NewsblogFlorida Fire Frogs Mess Comes At Bad Time For MiLB
(10 - 8:52am, Dec 15)
Last: Der-K: at 10% emotional investment

NewsblogOT - NBA Thread, Start of the 2019-2020 Season
(1487 - 11:57pm, Dec 14)
Last: Athletic Supporter is USDA certified lean

NewsblogOT - College Football Bowl Spectacular (December 2019 - January 2020)
(30 - 9:25pm, Dec 14)
Last: Howie Menckel

Hall of Merit2020 Hall of Merit Ballot Discussion
(360 - 9:15pm, Dec 14)
Last: Dr. Chaleeko

NewsblogRangers honor retired jerseys with dimensions at new stadium
(10 - 7:52pm, Dec 14)
Last: Itchy Row

NewsblogReport: Investigation into Astros' alleged cheating likely not completed until 2020
(4 - 6:58pm, Dec 14)
Last: The Yankee Clapper

NewsblogRays sign Japanese outfielder Yoshitomo Tsutsugo to two-year deal
(10 - 5:46pm, Dec 14)
Last: Qufini

NewsblogRed Sox, Pitcher Martin Perez Agree To One-Year Deal
(5 - 4:59pm, Dec 14)
Last: What did Billy Ripken have against ElRoy Face?

Page rendered in 0.4946 seconds
46 querie(s) executed