This is legit! Thanks for the post. I feel that since OPS is a better offensive metric, evaluating that PER position is a way to know where and why the offense keeps coming up short with RISP, down the line up. How do these numbers compare with clutch hitting stats?

Thanks for this. This look does also offer some hope, if we move Polanco back to SS. One could see a lineup with:

Sano -

Garver +

Kiriloff ?

Arraez +

Donaldson +

Polanco +

Larnach ?

Buxton +

Kepler -

That's 5 plus bats, 2 unknown youths, and cross your fingers on Sano and Kepler finding 2019 again. So it could be competitive, or even good, without having to add free agents

Quote

I looked around the Internet and found an article in ScoreSheetWiz where the author had taken the average of the top 30 players in each position over the last 3 years and averaged their OPS.

I appreciate the effort, but that article is very old (2007), its data even older (2004-2006), and beyond that, it is just generally flawed:

https://twinsdaily.com/forums/topic/49850-624-ops-spring/page/3/?tab=comments#comment-1017903

Fortunately, if you want to do your own research in that direction, there are great public tools available. I'd recommend the Fangraphs Splits Leaderboard, as its "position" split is by position played (not position qualified for, like some other Fangraphs pages), and it can also take custom date ranges, even spanning multiple seasons:

https://www.fangraphs.com/leaders/splits-leaderboards

You'd probably need to copy and paste data into a spreadsheet to average the top 30 by PA at a given position (the PA filter only applies to the player list, not to the league-wide splits).

I really enjoyed this.  It is a very constructive way to look at the positions.  I hope you are able to get more up to date data because I think we will look even worse with more current information. 

I went to fangraphs and wRC -

1. Garver is #20 on the overall list - our highest twin 154 score.

It really hurts to see LaMonte Wade at #27!

Nelson Cruz is the only other Twin (ex) in the top 60.

2. Josh Donaldson #62 127

Then there is Akil Baddoo at number 82 - still higher that Twins OF

3.Polanco is #93 with a 118

There is Robbie Grossman and Jonathon Scoop before we get to 4.Arraez at #136 - 109

5. Sano at #169 is 104

And that completes our over league average group (average 100). 

Jeffers 97

Kepler 96

Astudillo 96

Kiriloff 94

Larnach 89

Simmons 61

Thanks to Garver these 11 average 104 for the team. 

The only real criticism I'd have of this approach is that OPS assumes 1 point of OBP = 1 point of SLG, but this isn't true.  That's why Arraez is IMO, better then his OPS numbers.  His numbers are weighted more towards OBP then SLG.  

It's also why players like Buxton and Eddie Rosario HAVE to run high BA's to be successful.  They don't take any BB's, so they are widly BABIP dependent.  As long as they hit the ball hard and have at least average luck, they'll be fine.  But if they don't...

23 hours ago, mikelink45 said:

I really enjoyed this.  It is a very constructive way to look at the positions.  I hope you are able to get more up to date data because I think we will look even worse with more current information. 

...

If we attached these OPS+ to specific position averages, my guess is Sano falls to below average, but Jeffers and Astudillo rise to above average? It certainly seems like today's OPS averages are lower than 2006, which meets the eye test

Yet for all of the consternation about Arraez playing someplace, it is all how you want to look at it. Take a look at those players  with 100 total AB for this year, 3b  Arraez sits as the  17th best. OPS this year of the 61 players . 19th for the 71 players for LF, and 21 for the 68 at 2B. Below average number, above the median for players playing. 128 place out of the 399 players with 100 AB

Arraez is the utility player. Compare him to the other who do what he does

On 8/5/2021 at 9:34 AM, ToddlerHarmon said:

If we attached these OPS+ to specific position averages, my guess is Sano falls to below average, but Jeffers and Astudillo rise to above average? It certainly seems like today's OPS averages are lower than 2006, which meets the eye test

I think that this would make sense as I'd say since then there has been a deflation on the importance of BA and an inflation on the importance on OBP.  However according to OPS a 1B is MUCH more valuable then a BB.  While I agree that a 1B is more valuable then a BB it's not as much as OPS would think.

E.G. A player goes 1-4 with a 1B.  They have a .250/.250/.250 line.  That is a OPS of .500.

A player goes 0-3 with a BB.  They have a .000/.333/.000 line.  That is an OPS of .333.

I don't seem to believe that a BB is worth only 2/3's of a 1B.  I'd think a BB is worth more then that percentage.

My point is that like WAR, OPS is nice for a quick reference, but it's not the end all stat.  It underweights OBP vs. both BA and SLG.

Would you rather have a player with a .250/.350/.400 line (.750 OPS) or a .275/.300/.450 line (.750 OPS)

Player A in 200 AB's has 50 Hits, 22 Walks, and 80 Total Bases.  This is 102 Total Bases with Hits + Walks in 200 AB's, or 222 PA's.  They have made 128 Outs.

Player B in 200 AB's has 55 Hits, 5 Walks, and 90 Total Bases.  This is 95 Total Bases with Hits + Walks in 200 AB's, or 205 PA's.  They have made 140 Outs.

1 hour ago, RedBull34 said:

I think that this would make sense as I'd say since then there has been a deflation on the importance of BA and an inflation on the importance on OBP.  However according to OPS a 1B is MUCH more valuable then a BB.  While I agree that a 1B is more valuable then a BB it's not as much as OPS would think.

 

E.G. A player goes 1-4 with a 1B.  They have a .250/.250/.250 line.  That is a OPS of .500.

A player goes 0-3 with a BB.  They have a .000/.333/.000 line.  That is an OPS of .333.

I don't seem to believe that a BB is worth only 2/3's of a 1B.  I'd think a BB is worth more then that percentage.

My point is that like WAR, OPS is nice for a quick reference, but it's not the end all stat.  It underweights OBP vs. both BA and SLG.

I agree, and RC and wOBA do better. Even more extreme than your example is the overweighting of HR in OPS. One can be a slumping Sano. and have a good 470 SLG with an unplayable 280 OBP. The resulting 750 OPS may look average, but too many outs makes this hitter a weak spot in the lineup