torvik contribution projection

[Palatine]

No comment.

 

[Chin Diesel]

I'll take the over on Johnson's numbers.

 

[dannyHAM]

No JStew is laughable

 

[awy]

bro probably should have tweaked his model a bit before releasing this.

 

[nomar]

Huh?

Among other weird projections and omissions, McNeeley @ 9 PPG?

 

[pj]

Huh?

Among other weird projections and omissions, McNeeley @ 9 PPG?

What's weird about that? It wouldn't surprise me if McNeeley averages 9 ppg. I would guess he may be around 15-20 points per 40 minutes, and he might get 25 minutes given the depth of the team, so I would project 10-13 ppg. 9 isn't far off.

The oddest thing about Torvik's projections, apart from not including Jaylin Stewart, is the overall low point totals.

Add in Jaylin Stewart for 9 ppg and a few points for Ross and Nowell, you're at 66 ppg. Last year the team averaged 81.5.

 

[KemBuckets21]

These projections would make sense if we ran UVA's offense.

 

[nomar]

What's weird about that? It wouldn't surprise me if McNeeley averages 9 ppg. I would guess he may be around 15-20 points per 40 minutes, and he might get 25 minutes given the depth of the team, so I would project 10-13 ppg. 9 isn't far off.

The oddest thing about Torvik's projections, apart from not including Jaylin Stewart, is the overall low point totals.

Add in Jaylin Stewart for 9 ppg and a few points for Ross and Nowell, you're at 66 ppg. Last year the team averaged 81.5.

I'd be unpleasantly surprised if he averages 9 ppg given how good he supposedly is. Castle averaged 11 and he wasn't nearly as polished a scorer McNeeley is. Plus, we kinda need him to step up and score more than we needed Castle to. Is that projection crazy? No, but I think it's low.

The whole thing was weird. I thought the previous comments covered some of the other obvious issues.

 

[JustbrewitMan]

Doesn't really make you feel too confident about the validity of Torvik's data & models?

 

[JustbrewitMan]

Also *Sophomore not Sophmore

 

[Chin Diesel]

These projections would make sense if we ran UVA's offense.

65ppg from the top 7.

This board will be bananas if UConn is only scoring about 70 points total or less on average.

 

[HooperScooper]

Seems a little strange. Does anyone have the same Torvik projections from this time last year?

 

[Hunt for 7]

We are so talented we broke his Algorithm

 

[Edward Sargent]

All models are wrong some more than others. By the end of a seasons worth of data he will get the numbers right

 

[REB1094]

If Solo/Samson average 6/4, and Stewart averages 0, we’re in big trouble!

 

[CardiacAndre7]

I'd be unpleasantly surprised if he averages 9 ppg given how good he supposedly is. Castle averaged 11 and he wasn't nearly as polished a scorer McNeeley is. Plus, we kinda need him to step up and score more than we needed Castle to. Is that projection crazy? No, but I think it's low.

The whole thing was weird. I thought the previous comments covered some of the other obvious issues.

Steph wasn’t a shooter but his post game and strength added points to his average. Definitely ahead of McNeeley in transition as well. McNeeley is obviously a great shooter but we don’t know how the rest of his offense will translate yet.

 

[Waquoit]

We are so talented we broke his Algorithm

UConn broke KenPom in 2014.

 

[auror]

Doesn't really make you feel too confident about the validity of Torvik's data & models?

Preseason models have a lot more juice to them than just projected contributors, so yeah safely disregard this list. He doesn't go through and curate every team. There's a reason that recent program performance is a huge weight in preseason and early season models. Development systems and coaching matter and the full algorithms know that.

Though Torvik is still generally low on us due to the lack of experienced high major production.

 

[auror]

Seems a little strange. Does anyone have the same Torvik projections from this time last year?

Torvik had us 3, KenPom 4.

 

[Iluvsuebird]

I must wonder if Dan paid them to come up with this??? If this is not bulletin board material, I dont know what is???? If I am Liam, or Solo, or Stew, or Samson, or Ross......I am mad, very mad.

 

[HooperScooper]

Torvik had us 3, KenPom 4.

Thanks. I was looking for Torvik's "Contribution Projection" from last year. Although last year was much easier to project than this coming season.

 

[awy]

in fairness it is very difficult to model this UConn team because a lot of the big impact effects are far from normal. the team/staff level fixed effect is too large, the usage patterns too different from that of a normal team. removing his team and name a player with the same production observables as jaylin would seem to be a very bad bench player, so a production based projection wouldn't like him. in order for a model to like him it'll have to be a lot more complex and granular

 

[Samoo]

I'll take the over on Johnson's numbers.

The chances of SJ dropping 1.4 ppg and .8 rpg with DC gone seem rather low. Overall, he has the center position going from 18.4 ppg and 10.2 rpg to 12 ppg and 8 rpg. If that is the case we have a problem - luckily, it is improbable.

 

[Chin Diesel]

No JStew is laughable

People need to remember this when mid-season some dork is quoting Torvik like it's gospel. Good probability there are errors in all the calculations which go in to the in-season products he produces.

 

[Bboy Lace]

So, Sampson averages less than he did last year? Got it!