Is Defense Part of the Twins’ Problem?

[Cody Christie]

Minnesota has some of baseball’s best up-the-middle defenders. So, why does the team continue to struggle on the defensive side of the ball?

Defense wins championships is a common mantra in professional sports. In baseball, defensive metrics have taken time and technology to assist in evaluating the actual value provided by players. When the Twins signed Carlos Correa, Minnesota projected to have one of baseball’s best up-the-middle defensive groups. However, the team has struggled defensively this season, and most of the issues are tied to a few key players.

According to Outs Above Average (OAA), the Twins rank 22nd in baseball, with only two American League clubs lower in the rankings. Four players have contributed -19 OAA to the team’s overall total, including Luis Arraez (-7), Gio Urshela (-5), Carlos Correa (-4), and Jose Miranda (-3). Correa’s inclusion on this list might be the most surprising as he was arguably baseball’s best defender in 2021 on his way to winning the AL’s Platinum Glove. Outs Above Average isn’t the only metric that paints the Twins negatively. 

Minnesota currently ranks 19th in runs prevented, slightly better than their OAA ranking. However, the Twins’ defenders have posted a -4 runs prevented with only three AL team’s currently ranking below them. One of the areas the Twins struggle with the most is coming in on the ball. Only one team, the Yankees (-9), has a lower ranking than the Twins (-8) when coming in on the ball. 

Behind the plate, Ryan Jeffers and Gary Sanchez have both made improvements. Sanchez came to the Twins as one of baseball’s worst defenders. His framing skills have jumped from the 17th percentile in 2021 to the 51st percentile in 2022. Jeffers has seen his framing move from the 74th percentile to the 78th percentile but is below average as he struggles to control the running game. Catching defense could have been an issue this year, but it’s hardly been the team’s biggest problem. 

FanGraphs utilizes multiple defensive metrics that also show Minnesota’s defensive flaws. According to FanGraphs DEF rankings, the Twins are currently the 22nd best team with a -5/2 DEF total. Out of AL squads, only the White Sox and the Rangers rank worse than the Twins. Minnesota ranks similarly bad in other more traditional defensive metrics like defensive runs saved (20th) and UZR (15th). Overall, not every Twins player is having a terrible defensive season. 

Five Twins players have a positive OAA total, including Max Kepler and Byron Buxton, who lead the team with five OAA each. Gilberto Celestino currently ranks third with three OAA; Royce Lewis and Trevor Larnach are tied with one OAA each. Buxton has been getting scheduled days off and time at DH, so his total likely would be higher with increased playing time. Kepler has been an above-average defender in the past, and the metrics prove that he is still doing well. Celestino was considered a good defender as a prospect, and that tool has transitioned to the big-league level.

Minnesota’s biggest issue is the negative defensive totals compiled by Correa and Urshela. Arraez has been bad, but he’s also been playing significant time at first base, a new position for him. Correa’s bat continues to provide value, but he needs to produce at a high level on both sides of the ball. When the Twins acquired Urshela, he wasn’t considered an above-average defender, but he’s been one of the AL’s worst defenders at third. 

If the Twins want to improve defensively, it’s hard to pinpoint the best solution. Replacing Urshela at third isn’t an easy fix because the replacement options (Arraez and Miranda) aren’t considered strong defenders. Correa can improve his defensive numbers, which would alter the entire team’s defensive profile. Neither of these solutions is guaranteed to work, and none of these names will be voluntarily removed from the line-up. 

Bad defensive plays can lead to bad innings and a snowball effect for all players involved. How do you think the Twins can improve their defense? Have you noticed the team’s defensive struggles this year? Leave a COMMENT and start the discussion.

View full article

[mikelink45]

N ice review of an essential tool.  Defense is too often overlooked.  But like you, I cannot see anyone change on Def that will correct this.  Miranda to third, Polanco back to 2B and Arraez at 1B might be our best lineup (or Kiriloff at 1B abd Arraez DH).

[stringer bell]

With the limited play Buxton has received in the outfield, he hasn't had as much chance to shine in the field. In addition, with Buxton not anchoring the outfield defense, the two corners suffer as well. In the infield, if you believe Correa's rating, a lot centers on him. There is not a lot of speed on this club, particularly if Buxton is limited. That translates to a lack of range. 

IMHO, the defense hasn't been that bad. They are mostly fundamentally sound--throwing to the proper base, taking the safe out when called for--but not catching balls that can't be reached will be a problem for this team. 

[Stew]

Looked back the last 20 games. 13 errors. That's high but it's not one guy so I don't see a way of correcting it. Some help will come when polanko comes back. Also if Buxton could see the field more. Maybe other outfielders would cut down on mistakes.

[Whitey333]

Well written and thought out article.  Defense is important and in my opinion, with no stats in front of me to support said opinion, I believe Twins are just an average to below average fielding teams.

[Mark G]

"Minnesota ranks similarly bad in other more traditional defensive metrics"..........

Makes me wonder what "traditional" means, as we went a century or more without any of the "metrics" listed here.  Our eyes used to be the metrics most of us relied on, and I still do.  The article does as least give me encouragement in one area:  it makes me glad I don't pay any attention to analytic metrics, and never will.  

Ex:  When Larnach grades out higher than Correa and Urshela, it reminds me of the saying:  "are you gonna believe me, or your lying eyes?"

[yeahyabetcha]

I haven’t watched every inning of every game, but based on my observation, any metric that suggests Miranda is a better fielder than Urshela currently, should be discarded immediately. 

[FlyingFinn]

Where is Correa failing on defense, what area? Urshela makes some great plays and messes up some easy plays. People have wanted to get rid of Kepler, including me, but we don't have anyone equal to him in regards to corner OF defense. Maybe Celestino but he may end up being Kepler without power. Our current pitching staff needs all the defense it can get. Of course, our hitters have 1 run in the last two games so they aren't carrying the load either.

[Game7-91]

Defense is 1 part athleticism and 2 parts commitment/repetition.

Polanco and Arraez initiated this FO trend of creating interchangeable infield parts. Polancos injuries necessitated a permanent shift to 2b, he has improved since. Arraez is treated like a positional wandering nomad, and the team suffers. His offensive contributions are mitigated by his defense. But its not all on him. 

Whats worrying is the number of guys being developed in the same way. Is Kiriloff a 1b or OF? If 1b, play him there. Regularly. Miranda? Ditto. Arraez? Ditto. Lewis...SS, CF, corner? Really? After how many years , still cant figure out where he plays? Martin? What are they doing with some of the other minor league infielders featured on this site? Advancing high level prospects to MLB roster and deploying them like 26th man utility players? IMHO it is not only depressing wins now, but if our team somehow navigates this god-awful division to make a playoff series, will most certainly cost runs and games when the pressure is really on. 

It is clearly an organizational philosophy, but its the wrong one, and is detrimental to individual player development and the actual MLB product. Identify their best positions, give them rep's in MiLB, let them develop. 

Unless MlB wants to do something really radical, and roster for designated defense and offense....but thats for another discussion?

[TheLeviathan]

I know the eye test isn't terribly reliable, but neither are defensive metrics.  There is a lot in that I just don't think is terribly helpful for analysis.

This team seems to me to be mostly an average to above defense.  Though it has trended the wrong way the last month or so.

[terrydactyls]

The metrics say that Correa is a bad defender.  I think that is all I need to hear to eliminate analytic metrics from any discussion of baseball.  As someone who worked with statistics his entire career, I know they can be manipulated to prove anything you want.  I've done it.  It's easy, deceptive, and should be immoral.  But I worked with government and DOD so it was required.

 

[chpettit19]

1 hour ago, yeahyabetcha said:

I haven’t watched every inning of every game, but based on my observation, any metric that suggests Miranda is a better fielder than Urshela currently, should be discarded immediately. 

It's a cumulative stat. Urshela has played more than Miranda so he's had the chance to accumulate more bad plays. Same reason why the article points out that Buxton playing more would lead to him having more OAA.

[chpettit19]

15 minutes ago, terrydactyls said:

The metrics say that Correa is a bad defender.  I think that is all I need to hear to eliminate analytic metrics from any discussion of baseball.  As someone who worked with statistics his entire career, I know they can be manipulated to prove anything you want.  I've done it.  It's easy, deceptive, and should be immoral.  But I worked with government and DOD so it was required.

 

What's the motivation for the companies that produce these stats to manipulate them to say Correa is bad? Their company's future relies on teams trusting their stats to be accurate and useful.

[chpettit19]

1 hour ago, FlyingFinn said:

Where is Correa failing on defense, what area? Urshela makes some great plays and messes up some easy plays. People have wanted to get rid of Kepler, including me, but we don't have anyone equal to him in regards to corner OF defense. Maybe Celestino but he may end up being Kepler without power. Our current pitching staff needs all the defense it can get. Of course, our hitters have 1 run in the last two games so they aren't carrying the load either.

OAA says he's struggling with his range.

[terrydactyls]

1 minute ago, chpettit19 said:

What's the motivation for the companies that produce these stats to manipulate them to say Correa is bad? Their company's future relies on teams trusting their stats to be accurate and useful.

Did I say "manipulate them to say Correa is bad"?  No, I did not.  I just said that I don't put my faith in them.

[chpettit19]

1 minute ago, terrydactyls said:

Did I say "manipulate them to say Correa is bad"?  No, I did not.  I just said that I don't put my faith in them.

If the reason you give for not trusting stats is that they can be manipulated then the obvious logical conclusion to you not trusting stats that say Correa is bad is that you feel they're manipulated.

[terrydactyls]

6 minutes ago, chpettit19 said:

If the reason you give for not trusting stats is that they can be manipulated then the obvious logical conclusion to you not trusting stats that say Correa is bad is that you feel they're manipulated.

You still don't get it.  I am saying I don't like metrics because the end results are predicated on how the algorithms are designed.  The fact that Correa is mentioned is because that's who was being mentioned in the article and, in my opinion, it is an inaccurate representation of his performance.

[mikelink45]

1 hour ago, yeahyabetcha said:

I haven’t watched every inning of every game, but based on my observation, any metric that suggests Miranda is a better fielder than Urshela currently, should be discarded immediately. 

Miranda is playing 1B where he has no experience and has not played 3B enough for us to judge him, but that is the place he has the most experience. 

[old nurse]

15 minutes ago, chpettit19 said:

If the reason you give for not trusting stats is that they can be manipulated then the obvious logical conclusion to you not trusting stats that say Correa is bad is that you feel they're manipulated.

Your logic here is why I am continually amazed by comments sections. A person who says that statistics can be manipulated has their statement manipulated to mean what they want

[chpettit19]

Watching the games the Twins feel like an average-ish fielding team. Some days they're a little above average, some days below. But generally they're just ok. The Correa numbers surprise me. Urshela is a little worse than I'd have guessed, but not terribly. Mostly the numbers feel just about right. 

The defensive metrics are still a work in progress. I've worked for a couple of the companies that develop these stats and I'd suggest not taking any of them as gospel on their own, but if/when they all say someone is good, bad, or ok that's probably what that player is. There's a long ways to go in really dialing these metrics in, but if there's a consensus on a guy it probably means that guy is what the metrics say they are. These aren't dumb people putting the numbers together. "Analytics" haters should sit down before they read this part....the eye test is actually part of how these metrics are calculated ? But the people entering their eye test scores aren't just watching the Twins and a couple random games here and there, they're doing nothing but watching baseball all day, everyday (well 5 days a week, but everyday sounds better). So Urshela is getting his boost for "Wow!" plays. So the fact that he's still rating negative should say something. (Part of that something is that we're emotional beings so when a player on our team does something either great or really bad we tend to remember those things more and give them more weight in our opinions so our eye tests for our own favorite team is heavily biased)

[old nurse]

Roy Smalley was never the greatest athletic shortstop but always seemed to be in the right position to make plays. IIRC he attributed that to be a student of the game. Perhaps the computer takes the knowing individuals out of the equation

[joefish]

Versatility may be over-rated. 

 

[chpettit19]

1 minute ago, old nurse said:

Roy Smalley was never the greatest athletic shortstop but always seemed to be in the right position to make plays. IIRC he attributed that to be a student of the game. Perhaps the computer takes the knowing individuals out of the equation

A player like you're describing (always in the right spot so doesn't have to range far to make a play, but makes the plays on balls they can get to) would grade as an average fielder. They're not getting their team extra outs by making plays the average player can't because of great range or arm, but they're not making their team get extra outs by booting it on balls hit right at them. So they'd be right around 0 on all these metrics.

[tony&rodney]

Assists and putouts are traditional measures used to make a semblance of judgment about a player's defensive abilities. These are only reasonable across a larger sample of innings and games played. Naturally, there are mitigating circumstances of these numbers related to the type of pitchers, playing conditions, and fields themselves. However, across a large enough sample the numbers tend to level out. For example,  one could compare the assists and putouts from various infielders and outfielders to others playing the same positions.

I'm not finding the current systems in use being any more useful than the former ways. If one can separate their loyalty to one team and the obvious bias that often occurs when assessing your favorite players, the eye test can actually be fair. This is especially true if one has experience within the game as a player, coach, or scout. 

A few examples using Twins players: Kepler is a much superior defender than Larnach, Donaldson is a better fielder than Urshela, and Correa is much better than Simmons was last year but not in Simmon's league when Andrelton was in his prime. Our catchers are mediocre, Polanco is average to slightly above average and better than Arraez at second base. Kirilloff is much better than Sano, Arraez, or Miranda at first base. Buxton is still really good but there seems to be some decline. Celestino and Gordon are average to slightly below average. The Twins have their outfielders play pretty deep and shift a ton in the infield which also affects the defense, sometimes for the better and other times not so good. A concern is that the team itself doesn't seem to have a firm philosophy on their direction. Cleveland and Tampa Bay, for example, decided defense was important and their outfields are quite good. 

[Mander]

it's just a lack of every day players at all positions. too many different people at too many different positions all year.

[stringer bell]

1 hour ago, chpettit19 said:

OAA says he's struggling with his range.

The one tool where Correa is not above average is speed. Positioning is crucial and he turns balls he gets to into outs. But if the range is lacking, balls that other shortstops get to aren’t touched. I don’t know that I trust he’s below average, but I think he’s merely good because he doesn’t have the wheels to have plus range. 

[stringer bell]

1 hour ago, mikelink45 said:

Miranda is playing 1B where he has no experience and has not played 3B enough for us to judge him, but that is the place he has the most experience. 

Miranda isn’t brand new to first base like Gordon was new to the outfield last year. 

[RpR]

39 minutes ago, stringer bell said:

The one tool where Correa is not above average is speed. Positioning is crucial and he turns balls he gets to into outs. But if the range is lacking, balls that other shortstops get to aren’t touched. I don’t know that I trust he’s below average, but I think he’s merely good because he doesn’t have the wheels to have plus range. 

It is amazing, last year Simmons, whose defense was still top five was dissed for his bat (which so far this year is worse) so this year Correa , who is supposed to be better over-all is dissed for his defense, not his bat.

A lot of nit picking .  The rookies, except for Celestino (Larnach has a weak arm but is better than the others over-all) are the defense failing.

Arraez's fielding is a nail-biter most of the time, but I do not see him as a rookie any more.

[roger]

Would the defense be better if they played where they did historically?  Seems every game has a crucial at bat by the other team that would have been an out had the Twins middle infielder been where he belongs...IMO.  We don't see how many outs they get because of the shifts, or at least they aren't as obvious.  Just drives me nuts seeing the team not get a DP last night cause there wasn't anyone able to get to second.

[old nurse]

1 hour ago, chpettit19 said:

A player like you're describing (always in the right spot so doesn't have to range far to make a play, but makes the plays on balls they can get to) would grade as an average fielder. They're not getting their team extra outs by making plays the average player can't because of great range or arm, but they're not making their team get extra outs by booting it on balls hit right at them. So they'd be right around 0 on all these metrics.

Smalley made the plays because he knew positioning for the situation. That is how he got the extra outs. There is what the statistics do not measure. That is the limitation of these statistics.