What's up with ZR/EFF at 3B?

[diamond3017]

Why do 3B apparently do so poorly defensively according to OOTP's metrics? I'm not sure how much the recalculation for past seasons exacerbates this, if at all.

In my experience, it is very difficult to achieve a positive ZR (or EFF over 1.000) with a 3B, even one with a very high rating at the position. I thought this may be a byproduct of 3B getting fewer fielding opportunities, but good corner outfielders routinely post huge positive ZRs.

Here is a sampling of the career 3B fielding numbers from some of the stronger defenders in my current league:

Mike Schmidt, 90 3B: 1071 G, -37.1 ZR, .975 EFF.
Anthony Rendon, 83 3B: 1705 G, -45.7 ZR, .981 EFF.
Manny Machado (2015), 80 3B: 229 G, -13.4 ZR, .953 EFF.
Manny Machado, 79 3B: 1780 G, -78.3 ZR, .968 EFF.
George Brett, 76 3B: 972 G, -76.0 ZR, .942 EFF.
Jose Ramirez (2017), 74 3B: 1606 G, -59.0 ZR, .972 EFF.
Javier Baez, 72 3B: 479 G, -45.5 ZR, .972 EFF

Anybody else notice this, or have any input?

[chazzycat]

I have also noticed this, but have no clue what is going on.

[bailey]

I don't have a very good 3B, so I'm hoping low ZR is because of an overachieving 100+ defensive shortstop. Ditto on the other side of the diamond,

Edit: 1000 posts in 14 years.

[joehart]

Joe Cronin (120 at 3B) is at 14.9 ZR after 750 games.

[waittilnextyear]

Originally Posted by diamond3017

Why do 3B apparently do so poorly defensively according to OOTP's metrics? I'm not sure how much the recalculation for past seasons exacerbates this, if at all.

In my experience, it is very difficult to achieve a positive ZR (or EFF over 1.000) with a 3B, even one with a very high rating at the position. I thought this may be a byproduct of 3B getting fewer fielding opportunities, but good corner outfielders routinely post huge positive ZRs.

Here is a sampling of the career 3B fielding numbers from some of the stronger defenders in my current league:

Mike Schmidt, 90 3B: 1071 G, -37.1 ZR, .975 EFF.
Anthony Rendon, 83 3B: 1705 G, -45.7 ZR, .981 EFF.
Manny Machado (2015), 80 3B: 229 G, -13.4 ZR, .953 EFF.
Manny Machado, 79 3B: 1780 G, -78.3 ZR, .968 EFF.
George Brett, 76 3B: 972 G, -76.0 ZR, .942 EFF.
Jose Ramirez (2017), 74 3B: 1606 G, -59.0 ZR, .972 EFF.
Javier Baez, 72 3B: 479 G, -45.5 ZR, .972 EFF

Anybody else notice this, or have any input?

First of all, I agree with the premise to an extent. I think there's some hidden penalty for playing the hot corner and the resulting defensive metrics tend to be lower than for other positions.

But if I were to play devil's advocate a little bit, I'd point out that pretty much all those guys you listed are decent to solid but none are really elite at 3B. Maybe I'd give you 90-rated Mike Schmidt as an elite defender. I am a bit of a defense snob, though, so I routinely have guys rated 100+ manning the hot corner.

Another thing is to maybe slice these data into 162-game samples. I think the sheer volume of ZR accrued over ~1000 games can make meh results look terrible. If I do that for your list, we have...

90 Schmidt (87 range/59 error): -5.6 ZR per 162
83 Rendon (67 range/84 error): -4.3 ZR per 162
80 Machado (77 range/63 error): -9.5 ZR per 162
79 Machado (65 range/73 error): -7.1 ZR per 162
76 Brett (63 range if this is the gold one/53 error): -12.7 ZR per 162
74 Ramirez (67 range/83 error): -6.0 ZR per 162
72 Baez (79 range/46 error): -15.9 ZR per 162

So Baez/Machado/Brett are pretty bad and the others less so.

Another variable is many teams have cranked their ballpark factors up for hits, doubles, etc. That'll nerf the efficiency rating because I believe at its core, EFF is a measure of the rate at which balls in play become outs. If these players are playing their games at Coors Field, then their defensive stats are going to suffer as they're dodging line drives all game.

It might also be helpful to try and correlate different components of the rating (arm weighs heavily in overall 3B ability, but what if that's masking a low error rating?). Guys have to be able to get to a spot and then make a play to get credit for it. Some of those guys with decent-looking OVRs have some pretty glaring component ratings.

As for my own guys, I played Andrelton Simmons most often at 3B last season (91 G, +2.6 ZR, 1.021 EFF) and he has a 117 rating there. Johnny Pesky got some time also (58 G, 0.0 ZR, 1.000 EFF) and he just finished training to 109 there.

[diamond3017]

Let's start with Andrelton, because he's the guy who brought this whole thing to my attention.

With somebody like Andrelton Simmons, you might think that his initial training period was the culprit for his numbers at 3B being underwhelming. So I tried to control for that by limiting my examples to good fielders who were maxed out at 3B out of the box. But if you're interested, take a look at the defensive statistics this Andrelton Simmons card.

.986 EFF at 3B over 516 games. This is a player rated 117 at 3B. He is 14 points better at 3B than he is at SS, and yet a .986 EFF would be unfathomable for him at SS over a single season, let alone 500+ games. Yes, he had the initial training period, but he maxed out about 120 games in.

Originally Posted by waittilnextyear

Another thing is to maybe slice these data into 162-game samples. I think the sheer volume of ZR accrued over ~1000 games can make meh results look terrible. If I do that for your list, we have...

90 Schmidt (87 range/59 error): -5.6 ZR per 162
83 Rendon (67 range/84 error): -4.3 ZR per 162
80 Machado (77 range/63 error): -9.5 ZR per 162
79 Machado (65 range/73 error): -7.1 ZR per 162
76 Brett (63 range if this is the gold one/53 error): -12.7 ZR per 162
74 Ramirez (67 range/83 error): -6.0 ZR per 162
72 Baez (79 range/46 error): -15.9 ZR per 162

Those results look terrible, too, don't you think? What other position can you find a 90-rated defender post a -5.6 ZR over a single season, let alone average that over many seasons?

[Lukas Berger]

Interesting points.

Yeah, just at first glance something seems a bit odd here. I let Matt know to check this out, so we'll definitely take a look and see what we can find.

[humboldt1498]

Is this a result of many SS playing at 3B? Seems like that would skew the results since SS have much higher INF ratings.

Sent from my LG-H910 using Tapatalk

[dkgo]

Has anyone recorded the stats after the end of the each season or are you using the historical career view? I think it gets modified like WAR because it seems to me that the recorded ZR is worse when you are looking at past seasons stats than what it is with the current season. I could swear I've had guys that finish each season around -5 or so then I check career stats and they have -15 seasons

[LandiHero]

I just took a look at my defensive team to see if was happening for them too, and it seems to have been a similar story there. I had Rey Ordonez at 3B (120 rating after training) from the start until 2027, and across those seasons he collected 5 GG's, 10494.2 innings (!), 3122 chances, and only +20.5 ZR! More recently, I've had Aparicio playing at third since I phased out Ordonez. He's won all 3 GG's from 2028 to 2030. He's had 5871.2 innings at third and 1802 chances, but still only has a +3.4 ZR! On the other hand, when he's needed to play some at SS briefly, he's done far better. After playing 50 games, starting one, and amassing only 127.2 innings and 76 chances, he's already up to +6.6 ZR. Part of that is due to the fact that he doesn't have to train up at SS, but that's still nuts.

Across the rest of the infield, I've got Rookie Tulo at 1B from '22-'31 (12463.2 innings, 14966 chances, +62.2 ZR, 8 GG's), Mark Belanger at 2B from '19-'30 (15610 innings, 9242 chances, +367.2 ZR, 10 GG's), and Marty Marion at SS from '19-'31 (16162.1 innings, 9467 chances, +421.6 ZR, 8 GG's). I get the lower ZR from Tulo bc it's tough to get much at first anyways, but it shouldn't be the same way at third.

[Matt Arnold]

It's not an issue in the current season, since we use actual fielding stats to calculate values, so it's guaranteed to even out at the position. But looking through, it seems that the overall rates aren't quite lining up with expectations, so past season UZR values are a little lower than expected for 3B.

[stl jason]

Originally Posted by Matt Arnold

It's not an issue in the current season, since we use actual fielding stats to calculate values, so it's guaranteed to even out at the position. But looking through, it seems that the overall rates aren't quite lining up with expectations, so past season UZR values are a little lower than expected for 3B.

definitely lower than the original in past seasons at 3B.... I logged a couple players ZR in 2020 on my dynasty thread.... Freddie Lindstrom at 3B originally had a ZR of +9.8 that year (on the Waffletown team).... now his historical stats show the ZR in 2020 as +1.3

[fredbeene]

Originally Posted by waittilnextyear

First of all, I agree with the premise to an extent. I think there's some hidden penalty for playing the hot corner and the resulting defensive metrics tend to be lower than for other positions.

But if I were to play devil's advocate a little bit, I'd point out that pretty much all those guys you listed are decent to solid but none are really elite at 3B. Maybe I'd give you 90-rated Mike Schmidt as an elite defender. I am a bit of a defense snob, though, so I routinely have guys rated 100+ manning the hot corner.

Another thing is to maybe slice these data into 162-game samples. I think the sheer volume of ZR accrued over ~1000 games can make meh results look terrible. If I do that for your list, we have...

90 Schmidt (87 range/59 error): -5.6 ZR per 162
83 Rendon (67 range/84 error): -4.3 ZR per 162
80 Machado (77 range/63 error): -9.5 ZR per 162
79 Machado (65 range/73 error): -7.1 ZR per 162
76 Brett (63 range if this is the gold one/53 error): -12.7 ZR per 162
74 Ramirez (67 range/83 error): -6.0 ZR per 162
72 Baez (79 range/46 error): -15.9 ZR per 162

So Baez/Machado/Brett are pretty bad and the others less so.

Another variable is many teams have cranked their ballpark factors up for hits, doubles, etc. That'll nerf the efficiency rating because I believe at its core, EFF is a measure of the rate at which balls in play become outs. If these players are playing their games at Coors Field, then their defensive stats are going to suffer as they're dodging line drives all game.

It might also be helpful to try and correlate different components of the rating (arm weighs heavily in overall 3B ability, but what if that's masking a low error rating?). Guys have to be able to get to a spot and then make a play to get credit for it. Some of those guys with decent-looking OVRs have some pretty glaring component ratings.

As for my own guys, I played Andrelton Simmons most often at 3B last season (91 G, +2.6 ZR, 1.021 EFF) and he has a 117 rating there. Johnny Pesky got some time also (58 G, 0.0 ZR, 1.000 EFF) and he just finished training to 109 there.

I think it also matters who else in your infield
When I had an all 90 or better infield my zone ratings were POOR for each player but i had high efficiencies....go figure

[diamond3017]

Originally Posted by stl jason

definitely lower than the original in past seasons at 3B.... I logged a couple players ZR in 2020 on my dynasty thread.... Freddie Lindstrom at 3B originally had a ZR of +9.8 that year (on the Waffletown team).... now his historical stats show the ZR in 2020 as +1.3

I've been keeping an eye on this too. Don Hoak and Mike Moustakas typically have positive ZR for me from wire to wire at 3B during a given season, and then when the recalculation happens they end up with negative ZR for that season.

The extreme opposite happens at other positions, too. Last season, Lou Boudreau was +45 ZR at SS and Devon White was +39 ZR at CF. They were Great Glovers but those numbers are absurd. I think Lou was around +25 and Devon was around +15 when I checked their numbers near the end of the season. They both gained 2+ WAR.