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.
Page 1 of 2 pages
They probably shouldn't even report defensive stats at this point in the season.
Jon Heyman: i am not a hater of WAR stat, but if someone can explain to me how starling marte & bryce harper are both 1.7, please do
It’s a hardly a clown question, bro.
why don't we just take defense out of WAR? Make it a two number thing, with offensive/baserunning WAR as one number and the defensive +/- as another: e.g. Harper is at 1.2 WAR, +1, and Marte is at 0.7 WAR, +5 or whatever the case may be.
"I am not a hater of the BA stat, but if someone can explain to me how Chris Johnson is ahead of Miguel Cabrera, please do"
If all you want are offensive comparisons, why not use wRC+ or wOBA?
They probably shouldn't even report defensive stats at this point in the season
Fun fact: Harper's career bWAR of 6.9 is already the second highest ever...among players born in Nevada.
No offense intended Sean, you do great work. But 25 games worth of defensive stats doesn't tell me anything I didn't know before the season started.
Seriously. Chris Iannetta was rated higher than Albert Pujols by Batting Average this morning.
#19 - Wow, really? Hits are hits, your defensive numbers are opinion and should be discounted as such. I can't believe that needs to be said. Wow.
Fun fact: Harper's career bWAR of 6.9 is already the second highest ever...among players born in Nevada.
Your all-Nevada team:
The major issue with WAR in small samples is that people use it stupidly.
Considering the two calculations near equal despite their considerable difference in error bars
I know that defensive numbers lack the precision of batting average and the hitting numbers,
#26 - Just disappointing. The varying assessments of a player's defense among different systems tells you all you need to know about the precision of defensive numbers. Meanwhile, a double is a double. Considering the two calculations near equal despite their considerable difference in error bars is going to introduce an error, there's just no way around it. My faith in the system has been shaken. Bummer.
A double isn't a double. Sometimes what I think is a double the scorer rules an error.
A double isn't a double. Sometimes what I think is a double the scorer rules an error. If there are two outs and the pitcher walks the next batter and the following batter homers, the pitcher's ERA isn't impacted. At this point in the season, the difference in ERA can be considerable.
So, a double isn't always a double.
That almost never, ever happens.
This happens all the time. Dexter Fowler lofts a long fly to the wall. Justin Upton takes a triple axel with a back flip route to the ball. It ticks off the top of his glove as he pirouettes toward the spot, 10 feet from his starting point, the ball comes down. Bounces off the wall.
The point is it's wrong to draw an equivalence. There are judgment calls with both hitting and fielding stats, sure, but in the fielding stats they are considerably more frequent and important.
But more importantly many people here and elsewhere do not trust that today's most advanced fielding stats are accurate enough to be considered a simple representation of what really happened on the field. Our faith in hitting numbers guarantees that we believe that John Buck is outhitting Albert Pujols in a small sample. But with the fielding equivalent we're all skeptics.
ROFLMAO. Anyone see those two catches Harper made last night? He's an excellent defensive player. That first one he made, most corner outfielders don't even come close to making that play.
attaching a false equivalency is not an honest argument.
Who is making this argument? Who? The only argument I'm making about the fielding stats is that I think people are too skeptical not that they are equivalent to the batting stats.
To me the problem is that people are trying to make value judgments based on the numbers. It's a little thing but "Starling Marte is as good as Bryce Harper" is not the same as "Starling Marte has been as valuable as Bryce Harper so far." Unfortunately people want to use equivalent WAR as proof that the numbers mean equivalent ability.
But more importantly many people here and elsewhere do not trust that today's most advanced fielding stats are accurate enough to be considered a simple representation of what really happened on the field
War is just a counting stat. Guys are ging to have fluky months. I don't see why that is anything strange.
Not you, pointing to Rickey in post 35, Jim in 31 etc.. they are bringing examples of scorekeeper making judgement calls and trying to say that those are basically the same thing that is happening with defensive stats.
No, what I'm saying is the offensive stats aren't particularly reliable either at this sample size. I am saying that Heyman and anyone else ready to denigrate WAR because Marte and Harper are equal right now without having watched every play the two have made is jumping to a conclusion not supported by anything but baseless conjecture.
gloves never slump
It just seems really a stretch to imagine that in roughly similar number of innings, not only does Harper have 10 more chances, but that he missed another roughly 12-16 (assuming on average each miss is half a run) chances that Marte would have gotten...over a course of just 25 games..
I had felt for years that defense was not properly reflected in the advance stats. There was way too much emphasis on offence, whether or not the new defensive stats are accurate or not, I don't know, but I think they are closer than they were 5-7 years ago.
Never thought I'd see a thread where Sean Forman was the irrational guy trying to argue an incontrovertibly wrong point while the unwashed BTF masses were the ones preaching common sense and sanity.... but here we are.
Treating WAR, or any stat based on PBP defensive metrics, as a fact, like the number of HRs a guy has hit is flat out wrong.
All plays with a hit location of 7%+------------------+------------------+----------+-------------------+| pitching_team_id | batted_ball_type | count(*) | sum(outs_on_play) |+------------------+------------------+----------+-------------------+| PIT | F | 47 | 30 || PIT | L | 46 | 17 || WSN | F | 63 | 47 || WSN | L | 61 | 16 |+------------------+------------------+----------+-------------------+
Perhaps he gets crazy good jumps, so he tracks more of these down and Harper gets poor jumps (being a lifelong catcher) and he uses his speed to catch up to fly balls.
HR robbing plays (of which Marte has one).
It's got to be a fluke. Nobody is that good a fielder to make up for such a massive difference in offensive production. Projected out to a full season, Harper's numbers are comfortably in line with Ruth in his prime. Ozzie Smith couldn't make up for all that offensive value.
Projected out to a full season, Harper's numbers are comfortably in line with Ruth in his prime. Ozzie Smith couldn't make up for all that offensive value.
I don't think that's why people are objecting. People are objecting because they don't understand how a left fielder with 114 PAs and hitting .314/.386/.431 could be have provided as much value as another left fielder with 103 PAs and hitting .356/.437/.744.
Treating WAR, or any stat based on PBP defensive metrics, as a fact, like the number of HRs a guy has hit is flat out wrong. No matter what expert says so.
Or it could be, because of the small sample size, that Marte has been lucky so far in that he has gotten an unusually high percentage of line drives hit right at him.
Perhaps it's a small sample size fluke.
Where's the line though? Home runs are facts because of replay now, but some percentage of singles/doubles/triples are going to actually be errors and vice versa.
(71) No. If Matte has gotten a particularly easy distribution of plays, he has not generated real value. Any average LD would have caught those balls and had gaudy stats.
That's my best guess as to what's happened here.
The potential "fluke" results the same way that Justin Upton hit 12 HR in 23 games. Upton isn't a legit 80+ HR guy, nobody is. But he has performed that way over a small amount of games, and that has generated real value. Similarly, Marte could have simply done an exceptional job with the difficult chances he has had so for. Doesn't mean he will continue to do so, but it can happen, and it does add real value.
There are many gradations of "line drive" and "fly ball" into a particular zone. It is entirely possible in a 25 games sample that Marte has gotten "soft" line drives and flyballs relative to average.
Offensive and defensive stats measure fundamentally different things. Offensive stats take actual events that we know happened, and estimate a run value. Offensive stats do not include degree of difficulty in any way.
Offensive stats do not include degree of difficulty in any way.
The advanced offensive stats used to calculate WAR absolutely do. They take into account that it is harder to hit a HR in San Diego than in Philly, or easier to foul-out in Oakland than in Boston.
If it's the latter, I wouldn't say that WAR factors in the degree of difficulty as much as incorporates the particular value of an event in the context that it's happening. It would be related to DoD, but it's not directly taking DoD into account.
Very, very unlikely, but in any case, the runs scored really scored. The runs Marte allegedly saved (vs. an avg. LF) may not have actually been saved.
I mean if Marte has 10 catches that he made because he played no doubles defense 20 times, while Harper has 3 catches because he only played it 6 times, it will make Marte look better even though it was a team decision, not his individual positioning or his greater quality.
This is no different for WAR than guys who accumulate plate appearances because their teams score more runs. If Marte had not been standing where he was in the no doubles defense, then there would have been negative value to the team for him to not be there.
anything beyond raw slash lines is probably a bridge too far.
It is different. Guys who accumulate plate appearances are still being graded against what other people do relative to a standard plate appearance. In the example I gave, I didn't feel that I had to go over all the advantages the situation was going to create, but here we go. Even assuming that there is an equal distribution of balls hit to an area....
You must be Registered and Logged In to post comments.
Login to Join (8 members)
Page rendered in 1.6144 seconds, 58 querie(s) executed