ZiPS and Pecota Projecting 2017 Twins at .500

[Doomtints]

 

An argument for the optimistic scenario runs something like this:
- Their Pythag record last year was 65 wins.
- Their BaseRuns record last year was 71 wins.
So there were some signs that last year's team was really an unlucky 70-win team talent-wise. (Similar to how the 2015 team was a very lucky 75-win team talent-wise) If you use that as the baseline, then getting to 80 wins is possible. It's still a big lift, but not CRAZY.

 

If Baseruns called them a 71-win team last year and this year's team has the same problems as last year's team, wouldn't 71 wins be the reasonable projection?

[d-mac]

 

It doesn't quite pass the sniff test, does it?

 

It most certainly does not. 

[mazeville]

 

You can be optimistic and still not believe they will have a winning record. One might say optimism could manifest by predicting the Twins will have a loss number in the 80s as opposed to 90s or 100s.

 

I consider myself an optimist going into this Twins season and I'd totally take the under on this. 

 

Seriously: 81 wins would be a ridiculous, 22-win improvement. Certainly not impossible. But that would be nuts. 

[markos]

 

If Baseruns called them a 71-win team last year and this year's team has the same problems as last year's team, wouldn't 71 wins be the reasonable projection?

Maybe? I obviously don't know the inner workings of the PECOTA system, but at a certain level all projection systems predict that young players will get better and older players will get worse. And the Twins, for the most part, are a young team. And they did make one major move. Adding up:
1) Going from Suzuki to Castro is a huge (3ish WAR) improvement for PECOTA because they incorporate their framing metrics into the projections. 
2) Add a mild improvement for every starter under 27, subtract a mild downgrade for everyone over 30, and assume everyone in the middle maintains. This is a pretty young team, with 10-12 more young players than old. So even a mild age adjustment (say, 0.25 WAR) would add 3ish WAR.
This is all very hand-wavey, but at a high-level that is the direction this team is moving. I think the 80 win projection is surprising - definitely higher than I would go - but I don't think that 75 or 77 is unreasonable, especially if one buys into the fact that the 2016 was better than its record indicated.

[Doomtints]

 

Maybe? I obviously don't know the inner workings of the PECOTA system, but at a certain level all projection systems predict that young players will get better and older players will get worse. And the Twins, for the most part, are a young team. And they did make one major move. Adding up:
1) Going from Suzuki to Castro is a huge (3ish WAR) improvement for PECOTA because they incorporate their framing metrics into the projections. 
2) Add a mild improvement for every starter under 27, subtract a mild downgrade for everyone over 30, and assume everyone in the middle maintains. This is a pretty young team, with 10-12 more young players than old. So even a mild age adjustment (say, 0.25 WAR) would add 3ish WAR.
This is all very hand-wavey, but at a high-level that is the direction this team is moving. I think the 80 win projection is surprising - definitely higher than I would go - but I don't think that 75 or 77 is unreasonable, especially if one buys into the fact that the 2016 was better than its record indicated.

 

 

I guess we'll find out what's what in about 6 months.

BPs metrics in general are very difficult to follow. I can't tell if they are geniuses or just nuts.

[markos]

 

I guess we'll find out what's what in about 6 months.

BPs metrics in general are very difficult to follow. I can't tell if they are geniuses or just nuts.

Gleeman wrote an article (might be behind the paywall) discussing the Twins PECOTA projection:

http://www.baseballprospectus.com/article.php?articleid=31187

[Brock Beauchamp]

 

If Baseruns called them a 71-win team last year and this year's team has the same problems as last year's team, wouldn't 71 wins be the reasonable projection?

Only if you expect no progression from Sano, Buxton, Rosario, Kepler, and Berrios.

 

All of those players were a mix of terrible, mediocre, injured, and part-timers in 2016.

 

I don't buy into an 80 win projection but I see how they got there. Doubly so when you expect division-wide regression.

[MN_ExPat]

I think most folks have hit the nail on the head with the 71-77 win range.  While I would absolutely be ecstatic 80+ wins, I think at that point we would have to hope for the plus side of the "lucky bounces" that we saw in 2015 that allowed them to post such a great season (compared to recent years). 

 

The talent is there, just not fully convinced that the cosmic forces of baseball have fully aligned yet for the twins... soon, but most likely not this year.