Ryan on OF defense

[USAFChief]

http://blogs.twincities.com/twins/2014/12/04/twinsights-terry-ryan-defensive-metrics-twins-ignore/

 

With all the talk about OF defense recently, I thought this piece in the PP was interesting.

[gunnarthor]

The problem with defensive stats is that they might not be measured correctly and they might not be valued correctly.  If those two numbers are wrong, you can get some pretty funky conclusions.  Most of us probably like Dozier's defense at second.  Yet he's 29th out of all second baseman with 300 innings.  His negative value was equal to Hunter's 2013 defensive value (which was better than Stanton's value).

 

I do think the various systems can be used helpfully to make a pretty good list but the value part, I'm much less reliant on.  And I think RF in Comerica Park is a huge issue compared to most RFs.  And I think positioning a player (like the Twins did with Orlando Hudson) can make a huge impact on the numbers.

[stringer bell]

I want to agree with Ryan, but the numbers are pretty dramatic. I don't think corner OF defense is quite as crucial as up-the-middle "D", but combining slow guys who take bad routes with inconsistency in center hasn't worked.

[Thrylos]

Hunter failed the eyeball test last season too..

[DocBauer]

Honestly, not being a homer here, (pun sort of intended), I find Ryan's comments to be pretty forthright.

 

The man's background is in scouting. While he might have blown a few here or there, and reached for a couple cheap options with hope that didn't work out, the guy is far from incompetent. His wins in trades and signings outweigh his losses. And that's not me, that's been the general consensus of the baseball world for years. Ryan has always been very respected. I think a few bad moves and a rebuilding team can throw perspective out the window at times.

 

Watching, listening and following baseball most of my life, I feel I know at least a couple things. And one thing I know is that statistics are great, but they also lie at times. I think defensive metrics are OK, and can tell you a lot of things. But over the years, I've seen teams and players that ranked high in defensive rankings, but when compared to other teams and players I witnessed, simply weren't as skilled and productive. And I don't mean just flash. Flash only gets you so far. But there is a difference in teams/players with limited range who are sure and smart vs a team with more errors, for instance, but make big plays and save runs that another player/team simply couldn't make.

 

That, of course, is the eye test Ryan speaks of. I'm not saying Hunter DOES pass the eye test. I simply didn't watch him last season beyond a few innings here and there. I just don't know. But I do find his comments and analysis interesting vs comments and metrics I read here and there.

 

Is it possible Ryan and his scouts are actually right?

[Otto von Ballpark]

Is it possible Ryan and his scouts are actually right?

We'll find out, I suppose.

 

I will say, it's nice to finally start debating his baseball moves and statements from a standpoint of efficacy, rather than be constrained by too many limiters and qualifiers like days past (i.e. small market).

[Halsey Hall]

I think Ryan and his scouts aren't all wrong, by any means. 

[bird]

I suspect gunnarthur is onto something when he makes the distinction between a discussion of the validity of the defensive measurements and the value one places on what's being measured.

 

My hunch is that Ryan undervalues OF defense to some extent, and that may be because he really doesn't accept the numbers. I mean, they can be a bit startling, when we're talking about a 100-run swing between the best and worst OF defenses. Ryan is astute about a lot of things, but he seems to be way off on this.

[Mike Sixel]

The issue I have with those quotes isn't if he's right or wrong on Hunter being "average" on defense (if he is I'll be shocked).....but that he just dismisses defensive stats in a condescending way by saying they aren't even defined. It just shows, imo, a lack of willingness to learn and grow, the kind that led to US car manufacturers almost disappearing from the planet in the 80s. It smacks of hubris, and disdain for knowledge that isn't the kind of knowledge he has now. 

[nicksaviking]

Ugh, I'm not sure where this puts me. I'm very critical of Ryan's reluctance to buy into advanced statistics, but that's based in the fact that I want him to utilize them for offense and shifts, actual calculable data. I also think defensive metrics are more sham than glam.

[nicksaviking]

 

The issue I have with those quotes isn't if he's right or wrong on Hunter being "average" on defense (if he is I'll be shocked).....but that he just dismisses defensive stats in a condescending way by saying they aren't even defined. It just shows, imo, a lack of willingness to learn and grow, the kind that led to US car manufacturers almost disappearing from the planet in the 80s. It smacks of hubris, and disdain for knowledge that isn't the kind of knowledge he has now. 

I mostly agree in the fact he should know how these things work before dismissing them. Still UZR, critics and Ryan himself agree Hunter's issue is his range. That's still measured by the human eye. Ryan's scouts watched a representative sample of 30+ games and did not see the same drastic range issue. That basically means it's still human vs human discrepancy. Even if Ryan doesn't doesn't realize it, he's basically arguing that he trusts his scouts evaluations more then the UZR or crowd sourced scouting. Trusting your scouts over the other guy's scouts certainly is a valid opinion, even if your scouts turn out to be wrong.

[MileHighTwinsFan]

Maybe we are not giving the Twins enough credit. If it is the case that under Molitor The Twins will do more with infield shifts, they may believe they can compensate a bit for Hunter's defensive limitations. Further, right field at Target Field, particularly in the power ally, is not as spacious as Comerica. A rangy center fielder will be very important, but all things considered he may be adequate out there.

[stringer bell]

I do find it ironic that the Twins run of contention that started in 2001 was based on defense. Outside of catcher, they had plus defenders at every position and won games by being fundamentally sound, making great plays on defense and getting just enough offense. IIRC, Ryan was the architect of this formula and now seems to be discounting or disregarding defense. The game has changed since then, but defense is probably regarded as more important now than it was 14 years ago.

[bird]

I don't interpret his statements as an arrogant dismissal of the stats at all. He didn't reject the data out of hand by any means.

 

Accusing him of hubris and of being generally disdainful of knowledge, however, smacks of condescension and a lack of willingness to give the man any credit at all for acquiring new knowledge.

 

See what I did there, mike?

[TheLeviathan]

I'm a proponent of the eye test and not a strong fan of defensive metrics.

 

But you'd have to be blind to pass this defense with the eye test. It is clearly and demonstrably hurting this team.

[Richie the Rally Goat]

I don't interpret his statements as an arrogant dismissal of the stats at all. He didn't reject the data out of hand by any means.

 

Accusing him of hubris and of being generally disdainful of knowledge, however, smacks of condescension and a lack of willingness to give the man any credit at all for acquiring new knowledge.

 

See what I did there, mike?

if you combine this with the many other times TR has been quoted regarding advanced metrics, especially fielding metrics, you'd see a pattern pretty dismissive of them. For the TD offseason handbook it was partiçularly clear.

[gunnarthor]

 

The issue I have with those quotes isn't if he's right or wrong on Hunter being "average" on defense (if he is I'll be shocked).....but that he just dismisses defensive stats in a condescending way by saying they aren't even defined. It just shows, imo, a lack of willingness to learn and grow, the kind that led to US car manufacturers almost disappearing from the planet in the 80s. It smacks of hubris, and disdain for knowledge that isn't the kind of knowledge he has now. 

 

I think you're wrong here Mike.  Obviously he talked with Jack Goins about Hunter so he got some input from the "stat" point of view.  I don't particularly care if Ryan gives any weight to stats on fangraphs or B-R. The Twins (and all baseball teams) have more access to stats than we do.  Individuals in the A's and Red Sox systems have been critical of published WAR results.  So I don't think Ryan should take flax for not caring about a system of stats that has been demonstratively wrong on both valuation and measurement. 

 

I would hope and expect that Ryan (and Goins) considered RF playing area, positional shifts (or lack thereof) along with more than one years worth of defensive data (scouting or stat) o make a determination on Hunter.  The people attacking this move aren't.  I think it's silly that people think the Twins, of all orgs, are dismissive of defensive importance.  But they might trust other measurements over fangraphs.

 

I've pointed this out before but the Twins have had success with "bad defenders" coming here.  Orlando Hudson had -14 and -5 UZR the two years prior to coming here but managed nearly a 11 (and first in the AL) in 2010. 

[diehardtwinsfan]

I'm a proponent of the eye test and not a strong fan of defensive metrics.

 

But you'd have to be blind to pass this defense with the eye test. It is clearly and demonstrably hurting this team.

 

This sums it up well for me.  I agree with Ryan on defensive metrics.  They suck.  I'm very skeptical of a 100 run differential between the best and worst guys.  You're talking about someone who can make 200 more outs a year just to get into that range...

 

That said, Hunter is old and he's slowed down.  He can still hit, which is definitely a plus, but I don't see him solving anything defensively. 

[Platoon]

So Terry Ryan:

 

1. Ignores metrics?

2. Takes them into consideration, but abides by his eyes?

3. Got Hunter soley for fan appeasement?

4. Wants to show everyone he is the smartest guy in the room?

5. Does not believe catching balls in the air impacts ERA's?

6. Realizes what we do not, that their is no one in the Twins minor league system who can play LF?

 

Some of the above is facetious, sadly I am not sure which?

[Riverbrian]

 

The issue I have with those quotes isn't if he's right or wrong on Hunter being "average" on defense (if he is I'll be shocked).....but that he just dismisses defensive stats in a condescending way by saying they aren't even defined. It just shows, imo, a lack of willingness to learn and grow, the kind that led to US car manufacturers almost disappearing from the planet in the 80s. It smacks of hubris, and disdain for knowledge that isn't the kind of knowledge he has now. 

Just a thought...

 

What if Jack Goin is telling Terry that the defensive stats are half cooked.

 

I have no idea what Jack is telling Terry... But I'm willing to bet that he's in the room.

 

Following half cooked data also led to "New Coke" and the "McLean".

[Kirby_waved_at_me]

Ranking 51st out of 51 is still bad though, as flawed as the metrics may be. Say he should have been 45th or even 40th - That doesn't really make the move much better in my mind. You don't improve your defense by adding the statistically worst defender in the league.

 

I accept that I am not an expert; I'm only casually knowledgeable about scouting and defensive stats - but the implication that Torii will be fine because he looked good in the games against the Twins does not inspire much confidence in me.

 

I trust Terry Ryan that he's doing what is (in his opinion) the best for the short-term and long-term future of the Twins. I just don't see how the dots connect on defense yet.

[Mike Sixel]

It is not a one time event, if it was, I wouldn't have typed that. It is a pattern. And, to pretend it isn't a pattern is to, imo, just blindly ignore what TR and his assistant GM have consistently, publicly, stated for years now. Also, actions.......the actions back up that they largely ignore stats. We can pretend all we want that isn't true, but you can only judge people by what you see and hear from them.

 

Are people really arguing that the Twins appear at all to have embraced what many, many other teams seem to have embraced the last 5-10 years, in terms of use of statistics and math?

[Mike Sixel]

I don't interpret his statements as an arrogant dismissal of the stats at all. He didn't reject the data out of hand by any means.

 

Accusing him of hubris and of being generally disdainful of knowledge, however, smacks of condescension and a lack of willingness to give the man any credit at all for acquiring new knowledge.

 

See what I did there, mike?

 

Except it is a clear, clear pattern. Not a one time statement, but a pattern. Are you saying you actually think, based on their actions and statements the last 5 - 10 years, that they have actively moved forward from how they've run the organization since the 80's? Based on what you see and hear......

[nicksaviking]

Ranking 51st out of 51 is still bad though, as flawed as the metrics may be. Say he should have been 45th or even 40th - That doesn't really make the move much better in my mind. You don't improve your defense by adding the statistically worst defender in the league.

 

Well the same rankings say that Andrew McCutchen is the 3rd worst defensive CF and Mike Trout is the 4th worst.

[markos]

This sums it up well for me.  I agree with Ryan on defensive metrics.  They suck.  I'm very skeptical of a 100 run differential between the best and worst guys.  You're talking about someone who can make 200 more outs a year just to get into that range...

I don't think 100 runs is unreasonable at all. First, that comment was comparing the entirety of the Twins outfield against the Royals - it wasn't comparing individual players. Second, the value adds up very quickly in the outfield as unmade plays are often doubles and triples. Using the linear weights for various outcomes, difference between an out and a single is roughly 0.75 runs, a double is 1 run, and a triple is 1.25 runs. So a rough, back-of-the-envelop calculation would suggest the difference between the Royals and Twins was basically 1 play a game.

 

That may still seem a little crazy, but breaking it down further, I don't think it is unreasonable at all to think that once every three games Alex Gordon would make a play that Willingham/etc would miss, Cain/Dyson would make a play that Hicks/Santana would miss, and Aoki/Cain would make a play that Arcia/etc would miss. That is all it would take to add up to a 100 run difference. The advanced defensive metrics assign value in a much more sophisticated manner than my rough calculations here, but I think the numbers they have come up with are not impossible or even unreasonable. The Royals were really, really good, and the Twins were really really bad. Even if that difference is just a couple of plays per week at each outfield position, that will add up to a lot of runs very quickly.

[gmarais66]

I think some people are so fixated on these defensive stats that they can't see beyond them, to what's really happening on the field. Basically, Hunter is replacing Willingham in the outfield. My eyes told me that Willingham was just a touch better than placing a statue in the outfield. I didn't see a lot of Hunter last year, but I saw enough to know that at age 39, he's still magnitudes better than Willingham was last year. Bottom line: this significantly improves the outfield defense. Also, Ryan made a good point about not blaming the outfield for the team's pitching problems last year. This is something many people seem to be overlooking too, with their fixation on outfield defense. No matter how poor the Twins' outfield played last year, it didn't seem to have much effect on Hughes' numbers. Why was that? Because he pitched extremely well. The bottom line here is that pitching needs to improve significantly. If you have a lot of suspect pitchers in your rotation, you need gold-glove fielding to make up for their deficiencies. If you have a high caliber pitching staff, the defense becomes less of a factor. The outfield doesn't have to do much if the batter strikes out or grounds out harmlessly to the infield... Right? 

[Otto von Ballpark]

I wonder what Statcast ("FieldFX" player tracking) has to say. Target Field was one of the first places to be equipped.

[gunnarthor]

It is not a one time event, if it was, I wouldn't have typed that. It is a pattern. And, to pretend it isn't a pattern is to, imo, just blindly ignore what TR and his assistant GM have consistently, publicly, stated for years now. Also, actions.......the actions back up that they largely ignore stats. We can pretend all we want that isn't true, but you can only judge people by what you see and hear from them.

 

Are people really arguing that the Twins appear at all to have embraced what many, many other teams seem to have embraced the last 5-10 years, in terms of use of statistics and math?

I think Ryan's actions have shown that he clearly knows how to build a baseball team.  The problem I have with your POV is that I don't think the Twins (or any team) should care much about the stats that are available to the public.  Ryan has stated that he talks to his stat people before signing Hunter, Correia, Pelfrey, Willingham, Suzuki, Hughes.  We don't know what Goins tells him but Goins did imply on TD that the Twins thought Correia would pitch about as well as he did here, so they were clearly using some information we weren't privy to.  Despite having a horrible OF defense, Hughes put up the 3rd highest pitching WAR in the AL last year, so something must have worked.  

 

And most national bloggers have long left the Twins out of the list of non-stat teams.  Phillies, Royals, Giants tend to dominate those lists.

[bird]

Except it is a clear, clear pattern. Not a one time statement, but a pattern. Are you saying you actually think, based on their actions and statements the last 5 - 10 years, that they have actively moved forward from how they've run the organization since the 80's? Based on what you see and hear......

With this, I agree to a great extent. Clearly, Ryan, not surprisingly given his scouting background and his fine track record that reinforces his view, has tremendous conviction about the value of scouting and the ability of a trained evaluator to see all the pertinent things. It's not as if he doesn't have a valid reason to value what his scouts see more than he values the things the advanced data somewhat inadequately measures. Ballpark nuances, routes, cutoff men, arm accuracy, all brought up here, require eyeballs, right? 

 

Clearly, Ryan has been stubborn, and slow to accept the value in some of the new data. They have moved forward at a much slower pace than many organizations, no argument there.

 

But is it perhaps a cheap shot to attribute his stubbornness to hubris and an unwillingness to learn? I think it is. That's all I'm saying. We can be harsh in our criticism without falsely attributing negative personality traits to people. 

[jay]

I don't think it is unreasonable at all to think that once every three games Alex Gordon would make a play that Willingham/etc would miss, Cain/Dyson would make a play that Hicks/Santana would miss, and Aoki/Cain would make a play that Arcia/etc would miss. That is all it would take to add up to a 100 run difference. The advanced defensive metrics assign value in a much more sophisticated manner than my rough calculations here, but I think the numbers they have come up with are not impossible or even unreasonable. The Royals were really, really good, and the Twins were really really bad. Even if that difference is just a couple of plays per week at each outfield position, that will add up to a lot of runs very quickly.

 

Great post.  I think the biggest mental block in the perceptions of measuring defense is extrapolating from the few plays we see and know should have been made to all the plays that a better defender would have made and a run value over the course of a long, 162 game season. 

 

We can easily see and count home runs.  We can easily see and count RBIs and runs.  We can't easily see and count a missed play every three games or what that means as a run difference over 162 games. 

 

I think this leads people to dismiss the merits of defense metrics and the impact defense has -- whether that's in the outfield or behind the plate.