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

Sunday, December 15, 2019

Nationals ‘love’ Kris Bryant but potential holdup could stymie trade talks

With Anthony Rendon officially joining the Angels, the Nationals have a vacancy at third base.

Washington has options to replace Rendon; Josh Donaldson is still available in free agency, and Cubs third baseman Kris Bryant could potentially be had via trade.

The Nationals have reportedly inquired with the Cubs about Bryant, and while they “love” the 27-year-old, their focus is on Donaldson, according to MLB Network’s Jon Heyman. The Cubs would likely seek center fielder Victor Robles in a deal, a holdup on Washington’s end, Heyman said.

 

QLE Posted: December 15, 2019 at 12:33 AM | 48 comment(s) Login to Bookmark
  Tags: kris bryant, nationals, trade talk, victor robles

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. The Duke Posted: December 15, 2019 at 10:14 AM (#5908776)
It seems like Bryant would love a trade and sign deal but he has seemed focused on hitting free agency, and the Nats don’t seem to want to pay market or they would have Rendon

   2. snapper (history's 42nd greatest monster) Posted: December 15, 2019 at 10:53 AM (#5908786)
It seems like Bryant would love a trade and sign deal but he has seemed focused on hitting free agency, and the Nats don’t seem to want to pay market or they would have Rendon

MLB would love an extension that made Bryant's grievance go away.
   3. puck Posted: December 15, 2019 at 12:35 PM (#5908805)
Why are the Cubs shopping Bryant? (I've not really been paying attention.)
   4. Dag Nabbit at ExactlyAsOld.com Posted: December 15, 2019 at 12:56 PM (#5908808)
puck,

They've been in decline the last few years (especially 2019) and think the club needs a shake up. Also, they've traded away most of their best prospects over previous seasons, so the farm system is pretty bare right now. They're not so much set on trading Bryant, but there's been talk all off-season of shaking things up someone. There was a report a few days ago from Anthony Rizzo's camp about how they approached the Cubs for a contract extension and was esesentially blown off, which Rizzo wasn't happy about. There's been talk about how the Cubs might be able to get more in return from Contreras. There seems to be a desire on the part of the Cubs front office to trade one of their key players away, and Bryant is the former MVP.

That's the general gist. Now, if any of that is a good move is a whole other question.....
   5. The Yankee Clapper Posted: December 15, 2019 at 02:09 PM (#5908822)
The Cubs would likely seek center fielder Victor Robles in a deal, a holdup on Washington’s end, Heyman said.
Earlier reports said the Nationals would not trade Robles, but perhaps that changed after Rendon left. They could slide top prospect Carter Kieboom to 3rd, so a trade or free agent signing aren’t their only options. Robles might be very good, I don’t think I’d trade him.
   6. Walt Davis Posted: December 15, 2019 at 02:57 PM (#5908830)
Robles would join a long line of great Cubs CFs like ... ... ... Hack Wilson sorta.
   7. JJ1986 Posted: December 15, 2019 at 03:02 PM (#5908832)
I feel like 'holdup' implies something smaller than can't-agree-on-the-players-involved.
   8. Bourbon Samurai stays in the fight Posted: December 15, 2019 at 03:03 PM (#5908833)
Trading Robles for Bryant would be malpractice. Robles was better than Bryant by WAR last year and has 5 more years of control.
   9. Copronymus Posted: December 15, 2019 at 03:08 PM (#5908835)
Considering what the return on Corey Kluber was, I would start my offer at Michael A. Taylor and Hunter Strickland or something. If these teams want to dump good players so they can really gear up for a good tank, they shouldn't be rewarded with anything of value.
   10. Red Voodooin Posted: December 15, 2019 at 03:09 PM (#5908836)
They've been in decline the last few years (especially 2019) and think the club needs a shake up.


And for some reason they think that dumping their best player is the shake up they need.
   11. The Yankee Clapper Posted: December 15, 2019 at 03:21 PM (#5908840)
I would start my offer at Michael A. Taylor . . .
If you thought you could teach Taylor some pitch recognition & plate discipline, he’d be a good acquisition. Of course, that could be said of many that never had much of a MLB career.
   12. Bourbon Samurai stays in the fight Posted: December 15, 2019 at 03:23 PM (#5908842)
If you thought you could teach Taylor some pitch recognition & plate discipline, he’d be a good acquisition


All they need to do is tell him it's the playoffs all year
   13. PreservedFish Posted: December 15, 2019 at 03:28 PM (#5908843)
And for some reason they think that dumping their best player is the shake up they need.


It's worked for Theo before.
   14. Red Voodooin Posted: December 15, 2019 at 03:41 PM (#5908845)
It's worked for Theo before.


Nomar? Obviously he wasn't the best Red Sox player at that time, but damn if that wasn't the most excited I have ever been as a Cubs fan about a mid-season acquisition. I thought it was the move that was gonna push that 2004 team over the edge.
   15. PreservedFish Posted: December 15, 2019 at 03:51 PM (#5908847)
Yes, Nomar. He may not have been the very best Red Sock at the time, but he was an in-his-prime, totally beloved, homegrown Hall of Famer. I was absolutely shocked when they traded him. And absolutely shocked that it worked out so well.

Looking back, I'm amazed to see BR claim that his defense with the Sox in 2004 was so bad that he was actually replacement level despite hitting .321 as a shortstop. Wouldn't have thought that was possible. But he must have been truly brutal, as it was part of the justification of the trade, and indeed he was almost totally done at the position after that year.
   16. Fernigal McGunnigle Posted: December 15, 2019 at 05:01 PM (#5908854)
Nomar had a leg injury (Wiki reminds us it was a borked Achilles) and was completely unable to play SS when he came back from it. It was horrible to watch, especially after seeing Pokey Reese be generally awesome at the position for the first half of the year. Sure, Pokey slugged .300, but at least he could get to a ball that wasn't hit directly at him.

Speaking of Pokey Reese:
If you thought you could teach Taylor some pitch recognition & plate discipline, he’d be a good acquisition. Of course, that could be said of many that never had much of a MLB career.
With his glove and his wheels, all Pokey had to do was show he could put up a consistent 85 OPS+ with a non-horrid OBP. Alas...
   17. bbmck Posted: December 15, 2019 at 06:51 PM (#5908867)
Since Hack Wilson was traded to the Cardinals all seasons with positive WAA and 100+ games in CF for the Cubs total 32.8 WAA, a combined Hall of Fame-ish career since 1932.
   18. Walt Davis Posted: December 15, 2019 at 09:45 PM (#5908891)
Don't look at our post-Banks SS whatever you do.

As to Bryant-Robles ... obviously the different years of control is key to the trade. It's true that, by bWAR, Robles was quite valuable last year. However, by fWAR, it was just 2.5 as all of his value was tied up in defense (getting a +24 from DRS but only a +8.5 from UZR). Statcast also loves him (+23 outs vs average OF, led the league by a lot) while Inside Edge (at first glance) seems to like him a little less than UZR. I assume Steamer relies mainly on UZR and currently projects him to be an average CF overall next year. In short, if Robles can be counted on for 15-20 extra runs on defense or to improve substantially as a hitter over the next few years, then he'll be a very nice player to have; otherwise he's a pretty standard CF.
   19. Darren Posted: December 15, 2019 at 11:12 PM (#5908898)

I feel like 'holdup' implies something smaller than can't-agree-on-the-players-involved.


Ha ha, yeah, me too. I would love to buy a mansion except for this one minor holdup that came up.
   20. Swoboda is freedom Posted: December 16, 2019 at 09:08 AM (#5908910)
What's up with Bryant's defense? Why has it gotten so bad in last couple of years?
   21. Andere Richtingen Posted: December 16, 2019 at 09:14 AM (#5908911)
As to Bryant-Robles ... obviously the different years of control is key to the trade. It's true that, by bWAR, Robles was quite valuable last year. However, by fWAR, it was just 2.5 as all of his value was tied up in defense (getting a +24 from DRS but only a +8.5 from UZR). Statcast also loves him (+23 outs vs average OF, led the league by a lot) while Inside Edge (at first glance) seems to like him a little less than UZR. I assume Steamer relies mainly on UZR and currently projects him to be an average CF overall next year. In short, if Robles can be counted on for 15-20 extra runs on defense or to improve substantially as a hitter over the next few years, then he'll be a very nice player to have; otherwise he's a pretty standard CF.

He's also 22.
   22. Nasty Nate Posted: December 16, 2019 at 09:23 AM (#5908912)
What's up with Bryant's defense? Why has it gotten so bad in last couple of years?
He'd love to be the best fielding 3B but there was a holdup.
   23. PreservedFish Posted: December 16, 2019 at 09:26 AM (#5908913)
I guess those extra two weeks didn't quite take. Shame.
   24. Walt Davis Posted: December 16, 2019 at 04:53 PM (#5909088)
He's also 22.

Sure, which is why he might improve substantially as a hitter over the next few years. Just pointing out that believing he's a 4-WAR player right now requires believing he's a true +20 CF -- which he might be -- or hoping he's about to take a big step forward as a hitter -- which he might be.

On Bryant's defense ... it's a muddle. First, TZ likes him more than DRS with UZR sorta in the middle. A fair chunk of his good defense has been in the OF, never much more than an average-ish 3B. I'm not sure I can really justify this but he seems to have lost a smidgen of speed -- his steals, triples and OF defensive ratings are all down the last couple of years. There was also the shoudler injury which I think was his non-throwing shoulder but might still impact him on willingness to dive/reach at 3B I suppose.

It's certainly a possible reason the Cubs would make him available. If he's declining at 3B then he's a nice-hitting corner OF/1B -- nothing wrong with that but not a guy you offer a 8/$250+ contract.
   25. Sunday silence Posted: December 16, 2019 at 05:13 PM (#5909098)
. Just pointing out that believing he's a 4-WAR player right now requires believing he's a true +20 CF -- which he might be...


You always seem to take an agnostic stance w/ respect to these differning defensive numbers, such as last year with Harper.

BUt I would like to ask you: Do you think one system is better than the other?

Moreover: Dont you find it all curious that TZ almost never (from what Ive seen) has ANYONE at +20 runs? I mean at any position. Do you think that's defensible?

ANd it is reasonable/defensible then doesnt it call into question the ways teams make their lineups in regard to central defensive positions?
   26. Tom Nawrocki Posted: December 16, 2019 at 05:21 PM (#5909106)
What's up with Bryant's defense? Why has it gotten so bad in last couple of years?


I would like to tell you that a guy's never going to maximize his defensive abilities at third base if you keep sending him to the outfield every couple of weeks, rather than just telling him to focus on playing third base. But I'm not a genius like Joe Maddon.
   27. Sunday silence Posted: December 16, 2019 at 05:38 PM (#5909111)
Its a chicken or the egg problem right? Crappy 3b play OF cause they cant field; or they cant field cause they play the OF?
   28. Tom Nawrocki Posted: December 16, 2019 at 05:46 PM (#5909113)
If I had a decent-fielding third baseman, I would tell him to focus on his defensive skills and try to get better at playing third base.

If I had a bad-fielding third baseman, I would tell him to stop playing third base and go play an outfield corner.

   29. Moses Taylor, glorified meat shield Posted: December 16, 2019 at 05:56 PM (#5909115)
How did playing in the OF make him a worse defender at 3b?

There were concerns about Bryant's defense and ability to stick at 3b going all the way back to the draft, mainly because of his height. He actually improved over time, but then wasn't as good the last couple of years. Could it be because he's had injuries, and perhaps the concerns were justified? Isn't that much more logical than the fact he played other positions? Whether or not that was a good idea is a different discussion*. Oddly enough, I believe Bryant's OF defensive numbers are still fine.

*And one that I think Maddon was justified in making more often than not; usually Bryant would be pushed to the OF for a guy who was better at 3b (be it Javy** or Bote) and he would also be playing ahead of a worse OF defender (be it Schwarber, well, mainly him, or RF with Heyward moving to CF because the Cubs can't keep a CF who can hit***). Of course, none of these decisions are made in a vacuum, and there are always different and conflicting priorities. I agree Joe would go overboard, but I also don't think a blanket dismissal of using his players' positional versatility is a bad thing.

**Somehow he didn't get worse defensively at his main position of SS by playing at 2b and 3b.

***This includes Heyward.
   30. Moses Taylor, glorified meat shield Posted: December 16, 2019 at 06:01 PM (#5909116)
Trading Robles for Bryant would be malpractice. Robles was better than Bryant by WAR last year and has 5 more years of control.

Not to go all SBB here, but this sort of statement with such absolute confidence maybe makes me think things have gone too far. For one, why just bring up last year? Two, and as others have noted, over half of Robles WAR last year was defense so I for one am not totally convinced that's an infallible number; if you go by fWAR, Robles was only 2.5WAR last year while Bryant was 4.8; STEAMER projects Bryant at 4.8 again next year and Robles only at 2.0. I think it'd be a pretty safe bet to say Bryant will be worth more than Robles - and perhaps significantly more - the next couple of years. So that gets us to the "years of control" part, which absolutely matters. But at the same time, as a Cubs fan, I would be absolutely pissed if the Cubs dealt Bryant straight up for Robles.
   31. Sunday silence Posted: December 16, 2019 at 06:08 PM (#5909120)
if you go with fWAR then no one in the league can field more than 10 runs above average, correct?
   32. What did Billy Ripken have against ElRoy Face? Posted: December 16, 2019 at 06:27 PM (#5909124)
I guess those extra two weeks didn't quite take. Shame.
I assume he had to put the improvements he made in Iowa in some sort of escrow account pending disposition of his grievance. If he wins, he would at least have to return the benefit he got from being there.
   33. Tom Nawrocki Posted: December 16, 2019 at 06:35 PM (#5909127)
How did playing in the OF make him a worse defender at 3b?


Skills erode, and there's evidence that defensive skills erode faster than any other type. Whatever time Bryant had to put in practicing for a corner outfield spot was time (and mental energy) he couldn't spend working on his defense at third.

If the only defensive position Bryant ever had to worry about playing was third base, he'd be better at it, wouldn't he? There's a reason even veteran players take 100 ground balls a day.
   34. Sunday silence Posted: December 16, 2019 at 06:40 PM (#5909128)

Skills erode, and there's evidence that defensive skills erode faster than any other type.


what evidence would this be?
   35. Walt Davis Posted: December 16, 2019 at 08:15 PM (#5909145)
BUt I would like to ask you: Do you think one system is better than the other?

Not that I can tell. I assume eventually the precision of statcast (or its successor) will win out.

TZ handed out numbers well above 20 to Andruw all the time in his prime. It gave Buxton +29 in 2017; Inciarte +23 in 2016.
   36. Bote Man Posted: December 16, 2019 at 11:02 PM (#5909167)
usually Bryant would be pushed to the OF for a guy who was better at 3b (be it Javy** or Bote)

Hey, fellas, leave me out of this, OK?

Purely by the fallible eye test, Robles' defense improved noticeably as 2019 wore on. It seems that a flat number from DRS or UZR hides that improvement; trends are informative.

He's not the strikeout machine at the plate that M.A.T. is, but he was rarely the guy you'd want up there in a late-and-close situation. It's possible he could improve at the plate with Kevin Long and his pal Juan Soto giving him pointers.
   37. Walt Davis Posted: December 17, 2019 at 01:11 AM (#5909177)
trends are informative.

Well, that's sorta tautological. If a change over time is informative (i.e. predictive of the future or at least consistently explainable by other factors) then it's a "trend"; if it's not informative, then it's random noise or a one-time spike. Within-season changes (in either direction) tend to be the latter. But sometimes, especially for very young players (or if in response to injury, say), short-term changes are indeed trends.

what evidence would this be?

There have been studies -- not than I can find them and I vouch for none of them. I believe however it would be more accurate to say that they start to decline sooner (i.e. at very young ages) rather than decline "faster" (i.e. at a greater rate of decline).

This one at BtBS was the best I found in a quick search. Runs through 2010. Defense is at a relatively constant peak from 22-27/28 then steadily declines. So defensively, an average player is basically at his peak on the day of his arrival. But it's not that dissimilar from the batting runs curve. But we also need to bring in position which starting a very slight decline around 25. It's trivially small but once position is included, the typical player seems to drop to 0 dWAR (really dWAA) around age 29-30 ... but was never a very good defender.

It's also trivially small but the decline in fielding runs for 1981-2010 is slightly lower than during 1950-79 but the decline in positional value is slightly larger in the more recent period. They basically work out the same such that an earlier player at 36 would have declined to about -4 on runs but only -2 on position while the more modern player would be -3 on both -- i.e. teams today do a slightly better job on finding the right position for older players -- that could be entirely a DH effect.
   38. Walt Davis Posted: December 17, 2019 at 01:37 AM (#5909178)
Back to defensive stats and which I prefer ... I was in a hurry so just gave the quick answer. The slow answer is pretty much the same though. I don't know what to trust. Nearly everything except Inside Edge can produce some numbers I think are probably too big ... but IE seems like it's probably too conservative. I would have thought that TZ would always give us DRS numbers attenuated towards zero but that's not the case. Statcast is probably the most reliable but they haven't figured anything out yet for IFs and I'm not thrilled that they compare with average OF while the others go by average position -- just makes it hard to compare.

But Statcast is pretty stingy with the +20s too. In 2019, the only one was Robles at +23. But again that's compared with average OF and given nearly everybody with 8+ outs above-average is a CF, I assume that comparing with CF average would bring Robles below 20. Last year Cain and Inciarte barely topped 20 with the same issue. Buxton's +26 in 2017 might keep him above 20 (although my rough guesstimate comes out around 17). So my guess is that a +20 CF (relative to other CF) is the max sustainable (i.e. true talent) level and I wouldn't be surprised if it was 15. Maybe that can be pushed to 20 runs saved. And I'd guess that does usually start to fade in the late 20s.

The main thing that IE and Statcast have taught me is that the largest portion of FBs are dead routine plays that basically only drop in on true random errors like dropping it or losing it in the sun. In 2017, Buxton made 389 catches on about 411 non-impossible balls, and only 25-30% of those had any risk attached. So I think it probably is pretty rare for an OF to make 20 extra plays on about 120 non-routine FBs especially when 75% of those are still reasonably routine. Buxton might well have had only about 25 balls all year that aren't caught at least 1/3 of the time of which he caught 7 or 8 while the typical guy probably caught one and the typical CF 2-3. On the one hand, that could add up to 1 win a year; on the other hand, it's a an extra play every 3-4 weeks.
   39. Bote Man Posted: December 17, 2019 at 08:29 AM (#5909186)
Well, that's sorta tautological. If a change over time is informative (i.e. predictive of the future or at least consistently explainable by other factors) then it's a "trend"; if it's not informative, then it's random noise or a one-time spike. Within-season changes (in either direction) tend to be the latter.

I recall someone on this very site a few years back pointing out that UZR does not provide useful results with less than 3 years of input data. Is this true?

If so, then comparing one season's worth of UZR results would not provide meaningful data. Does that make it noise?
   40. Adam Starblind Posted: December 17, 2019 at 09:42 AM (#5909207)
I feel like 'holdup' implies something smaller than can't-agree-on-the-players-involved.


Yeah, "we're not trading you our excellent 22-year-old starting CF" isn't a "holdup." It's a "go pound sand."
   41. Bourbon Samurai stays in the fight Posted: December 17, 2019 at 09:57 AM (#5909213)
Not to go all SBB here, but this sort of statement with such absolute confidence maybe makes me think things have gone too far.


yeah, I didn't meant to imply that Robles was definitively better than Bryant, but that he is already good enough it doesn't make sense to trade him for Bryant, given the economic issues.

I don't know if Robles will post the same defensive WAR every year, but by the eye test he certainly looks elite, and there's reason to hope his bat will continue to improve.
   42. Tom Nawrocki Posted: December 17, 2019 at 12:25 PM (#5909288)
I believe however it would be more accurate to say that they start to decline sooner (i.e. at very young ages) rather than decline "faster" (i.e. at a greater rate of decline).


That's what I meant. Defense peaks earlier and begins to erode earlier than other other baseball skills.
   43. Sunday silence Posted: December 17, 2019 at 12:55 PM (#5909304)

But Statcast is pretty stingy with the +20s too. In 2019, the only one was Robles at +23.


OK but this is on catches alone? Right? There is also value to be found in "kills" as well as holding the runner. Im guessing that an outstanding assist guy could add 10 runs above average and outstanding holds, 5 more. Not a lot. And not sure any one player can maximize his value in all three parameters, but still theres potential there for an elite OFer to be 30 runs above average. Or no?

On the one hand, that could add up to 1 win a year; on the other hand, it's a an extra play every 3-4 weeks.


Dont you think its more than that? (for an elite defender) I watched two weeks of the playoffs at least some of it. I think Robles made what 4 somewhat difficult catches? If say those are 50-50 balls, that's 1 run/week. (maybe more). Same with Rendon at third, I think making 2 fairy difficult plays a week is do-able.

We talk alot about how its well nigh impossible to rate these players by the eye test; memory being what it is. Thats why we have stats, so we can say Gehrig hit 24 grand slams or Parket had 2700 hits.

BUt you'd think if we watched enough playoffs, we could start to put rough numbers on these things and we (primates in general) wouldnt have such disparate views. Like discretionary fly balls, that's a pretty important factor in how we evaluate Undruh Jones. Cant we all agree on about how many of those we see per week? Using playoffs as a proxy for entire seasons/careers makes sense because we have footage and we have millions of eyes watching this.
   44. Sunday silence Posted: December 17, 2019 at 01:25 PM (#5909312)


But Statcast is pretty stingy with the +20s too. In 2019, the only one was Robles at +23.


You make it sound like a 20 catch above avg. guy is some sort of statistical aberration. Lets look at this again:

In 2016 we had 2 guys w/ 20 OAAs. Plus Marisnick would have made it if he played a whole season (311 AB). Heyward would have 17 if he could make one more catch in 10 games and still take off 12 games.

In 2017 we have 3 guys w/ 20, plus Kiermaier if we extrapolate (he played 1/2 season). Plus there's 2 guys w/ 18 OAA and JBJ would have made 18 too if he played full season.

In 2018 we have 2, plus Bader w/ 19 and he missed 24 games so its reasonable to think he could make one more catch in 15 games say. Engel probably gets 17 or 18 in a full season.
In 2019, theres one, plus Kiermaier prolly gets 19 or 20 w/ 25 more games.

So if we extrapolate to full season, every year there's on average 3 elite OFs who can save us 20 runs/season on catches. And there's at least one guy who's close at 18.

So on average, there's 4 guys in MLB who can save us 20 runs a season on catches. That seems perfectly reasonble in terms of a bell curve of abilities out of 30 CFers. Its also verified by stat cast and it seems to coincide with what me eyes are seeing.

Agreed? Now what is TZ giving us for these guys? Im too lazy to check since I just got done doing the stat cast stuff manually.

Isnt there an obvious discrepancy here? ANd arent GMs and managers playing guys as if they believe the stat cast stuff and not the TZ stuff?
   45. Lance Reddick! Lance him! Posted: December 17, 2019 at 03:11 PM (#5909347)
The main thing that IE and Statcast have taught me is that the largest portion of FBs are dead routine plays that basically only drop in on true random errors like dropping it or losing it in the sun.

I learned this looking at STATS zone data, where it turned out impossibly bad late-career Bernie Williams was converting something like 90% of all his chances, making any standout number like Erstad's 2002 facially absurd.
   46. Walt Davis Posted: December 17, 2019 at 05:21 PM (#5909394)
I recall someone on this very site a few years back pointing out that UZR does not provide useful results with less than 3 years of input data. Is this true?

If so, then comparing one season's worth of UZR results would not provide meaningful data. Does that make it noise?


A good point and thanks for the reminder. It all comes down to what accuracy level you like. But the basiccs are right there in the numbers I cited for Buxton. When we look at a hitter, we talk about needing 3 fullish years to come up with a "good" projection for year 4 -- say 1800 PAs. BIP are similar to PAs which means we need 4 full years to get to about 1600 opportunities for Buxton. Given something like 1100-1200 of those would be dead routine balls that every OF catches, we might need even more than that.

But like I said that's a question of how accurate you want your projection to be. In projections, that's basically handled via how heavily a projection is regressed towards the mean. Somebody like Robles would still get heavily regressed whereas Kiermaier would be regressed only a bit. But I'm not sure the projection systems are doing it in this manner.

You make it sound like a 20 catch above avg. guy is some sort of statistical aberration.

Again, compared with an average OF, not an average CF. And yes, 1-2 guys a year saving +20 is ... what term do you want, uncommon? And while the top 10 is reasonably stable, the +20 guys are usually not the same names from year to year, suggesting that is not their true talent level. So yes, I'm guessing that the max sustainable level for a CF compared with other CFs is around 20 runs and that's probably something we've rarely seen. But maybe it's +22 -- so what?

Yes, statcast is doing just range and there may be some CFs adding assists and that might push them above 20.

As to TZ -- I am not defending it or deriding it. You seem to think that not producing numbers of 20+ is problematic and I showed that it has often produced numbers above 20. In 2017, statcast rated Buxton 26 catches above the average OF; in 2017 TZ rated him 29 runs better than the average CF. If you want to work through the numbers (converting outs to runs, adjusting for throws, adjusting for average CF vs average OF) to see if there's some substantial difference there, be my guest. Looks close enough to me to say they came out the same for that player-season.

But I dunno. Maybe TZ gives out fewer +20s even after you adjust for the average OF vs average CF thing. I have no idea, if you've done the grunt work then please share. The quick and dirty estimate for the difference between average CF vs. average OF is to use Rpos from bWAR. Robles got +3, Soto -6 and Harper (152 starts in RF) -6. That puts the average OF around -3 runs and the difference between the average CF and the average OF around 6 runs, probably 5-6 catches. So take statcast and deduct 5-6 catches from the CFs and the best guys are usually under 20 catches compared with the average CF. (FWIW, in the past I've eyeballed the CFs in statcast and +5 seemed about right.)

I don't mind comparing with average OF rather than average position, there's an argument in favor of that, but it's important to make sure different stats are being compared on the same basis. By going out on its own (using outs not runs, using average OF not average position), statcast has made it difficult to compare its results with others. (Note IE is even worse.) Anyway, TZ (and DRS and UZR) is measuring runs relative to average CF (and includes throws); statcast is measuring outs relative to average OF. Their (raw) magnitudes shouldn't match and by comparing to a higher average (for CF), we'd expect TZ to be closer to zero after adjusting runs/outs.

Anyway, did Robles in 2019 make a play a week that an average OF (or CF?) would not have made. Maybe so, it's certainly not impossible and maybe not even rare. Will Robles do that for the next 5 years? That's very, very unlikely. In general but especially in the context of this thread, I don't particularly care how outstanding Robles' defense was in 2019, I'm interested in how outstanding it's going to be for 2020-2024. I would bet the under on +20.

Nobody's even mentioned yet that Robles is surely one of the all-time great disagreements. Statcast at +23 outs, DRS at +22 runs, UZR at about +9 ... TZ at -3. And that's giving him +3 for his arm. Note that DRS gives him +9 for his arm and just +11 on plays made in CF so it doesn't agree nearly as much with statcast as it appears. UZR gives him 5 for range and 2 for arm (and I guess about 2 for position). By RF9, he made about 20 more plays than the average CF (plus another 6 or so in RF) so that seems in rough agreement with statcast.

So ... which to choose? They don't agree on arm, they don't agree on range, they don't agree on total, I don't have a clue how TZ gets to negative range for Robles. FWIW, TZ was also negative on Cain for 2018 and 2019. Meanwhile TZ swooned over Keon Broxton at +19 in just 450 innings (statcast +9); also big on Mallex Smith, usual big love for Marisnick and, among full-time CF, puts Trout at the top with +12. It really disliked JBJ this year (-11).

I have no particular love for TZ. Prior to statcast, DRS critics of the "there's no way he saved 40 runs" variety would often point to TZ since it was easily available. So if other folks want to believe TZ over DRS, who am I to argue, I might as well put it out there. I'm happy to go with an average of all of them as long as somebody else is willing to go to the trouble of putting them all on the same scale, averaging them and making that info easily available. As I said, if any of thme is correct, I assume statcast is -- objective and hopefully precise measurement is a pretty strong argument over the others. It also seems to me to be reasonably conservative -- most everybody seems to be in the +5 to -5 range (esp relative to position) and, over a few seasons, I'm not sure there are more than a couple of people who are consistently +15 or better (though there will be more by rate). So if we want to declare statcast the winner, that's fine with me too.

But what I see is numbers bouncing around across statistics, numbers that even when they are close at the total often don't agree very well in the components. If there's consistency of one always being lower than the other, I haven't noticed it -- again, if somebody has done this work, please share. For TZ and Statcast, they disagree a lot on Robles (TZ much lower), Bradley (TZ much lower), Broxton (TZ much higher) and Trout (TZ much higher). DRS and Statcast sorta agree on Robles but not really, differ not hugely on Broxton and Bradley, agree on Trout. UZR and statcast differ pretty substantially on Robles, differ not hugely on Broxton, differ not hugely on Bradley, agree on Trout.

So 3 of 4 agree Trout was nothing special defensively in 2019 but TZ puts him among the best. All agree that Broxton was above-average but everywhere from a little bit to a good bit to a best-in-league bit to an astronomical bit. Statcast puts Bradley above-average, UZR and DRS about average, TZ well below -- maybe the first differences are the average OF vs average CF thing. They don't really agree at all on Robles who is anything from average to best-in-league (among fullish-time CF) but even if he is best in league that's either outstanding range (statcast) or very good range and outstanding arm (DRS). So ... rankings by TZ, DRS, Statcast, UZR for these 4 (ratings include arm for all but statcast, no adjustment for average OF)

TZ: Broxton (by a mile), Trout, Robles, Bradley (30 run spread top to bottom)
DRS: Robles (by a mile among these 4), Broxton, Bradley, Trout (tied ... 23 run spread)
SC: Robles, Broxton (prob #1 by rate), Bradley, Trout (25 out spread)
UZR: Robles, Broxton, Bradley, Trout (9 run spread)

For this "random" sample of 4, UZR is the one limiting the range/variance. But reasonable agreement on Robles #1 but, again, that's not as agreeable as it appears. SC gives him credit for big range (and unmeasured arm); DRS gives him credit for good range and big arm (+9); UZR gives him above-average range and arm, putting his total 13 runs below DRS, probably at least that below SC. So, by range, we still have Robles below-average (TZ), above-average (UZR), very good (DRS) and best-in-league (SC) ... and an arm ranging from +2 to +9. Neither DRS nor UZR was all that impressed with Broxton. SC was really the only one impressed at all by Bradley but again that might be the average OF vs CF thing.

So my best guess -- you want a guy who can really run and catch fly balls like nobody's business, you probably want Broxton. You want a guy who can run and catch fly balls really well and hit well enough to justify starting, you want Robles. You want a guy who's probably still above-average defensively and hits a bit, you want Bradley. If you're lucky, maybe Trout will hit enough to make up for his defense. :-)
   47. Sunday silence Posted: December 18, 2019 at 02:05 PM (#5909637)
thanks for this in depth analysis, Walt. I will have to look at all this that you said and continue to study this. you've given a lot to go on here. Thanks again.
   48. Bring Me the Head of Alfredo Griffin (Vlad) Posted: December 18, 2019 at 03:40 PM (#5909678)
I recall someone on this very site a few years back pointing out that UZR does not provide useful results with less than 3 years of input data. Is this true?


My recollection is that you can extract useful-but-noisy results at ~1,500 innings (i.e. about one full season's playing time), and that they get progressively less noisy after that.

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
Phil Birnbaum
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogJeter's Hall call near-unanimous; Walker elected
(27 - 5:04am, Jan 22)
Last: Bring Me the Head of Alfredo Griffin (Vlad)

NewsblogWith no sure things, next year's Hall of Fame ballot could be unpredictable
(1 - 1:39am, Jan 22)
Last: Walt Davis

NewsblogCrane: Players will apologize in Spring Training
(4 - 1:28am, Jan 22)
Last: The Yankee Clapper

NewsblogL.A. City Council to vote on asking MLB to award Dodgers championship trophies
(39 - 12:53am, Jan 22)
Last: QLE

NewsblogRed Sox legend Pedro Martinez doesn’t support Astros whistleblower Mike Fiers; ‘You’re just a bad te
(1 - 12:39am, Jan 22)
Last: Howie Menckel

NewsblogRyan Thibs’ Hall of Fame Tracker
(1343 - 12:20am, Jan 22)
Last: Kiko Sakata

NewsblogNolan Arenado feels “disrespected” as Rockies GM Jeff Bridich nixes trade talks
(41 - 12:01am, Jan 22)
Last: Walt Davis

NewsblogOT - Soccer Thread - January, 2020
(224 - 11:37pm, Jan 21)
Last: AuntBea calls himself Sky Panther

NewsblogCubs Fans Boo Chairman Tom Ricketts at Team Convention
(35 - 11:22pm, Jan 21)
Last: Zonk didn't order a hit on an ambassador

NewsblogPrimer Dugout (and link of the day) 1-21-2020
(12 - 11:01pm, Jan 21)
Last: bobm

NewsblogEx-M’s ace Félix Hernández has minor league deal with Braves
(27 - 10:40pm, Jan 21)
Last: baxter

Hall of MeritLarry Walker
(42 - 10:16pm, Jan 21)
Last: Bleed the Freak

NewsblogWATCH: Carpenter says lack of identity at plate is why he struggled; New focus to not pull the ball
(9 - 9:56pm, Jan 21)
Last: Walt Davis

Hall of MeritGavvy Cravath
(305 - 9:37pm, Jan 21)
Last: DL from MN

NewsblogJ.D. Martinez says Red Sox are innocent. Now what?
(27 - 9:28pm, Jan 21)
Last: Darren

Page rendered in 0.5171 seconds
46 querie(s) executed