Data scientist, physicist, and fantasy football champion

Week 6 DEF Predictions

It looks like Model F is the one to go with, though Yahoo’s experts are no slouches.

Let’s see some predictions for this week:

ModelTypeScoreOppScoreOppHomeAwayDataYearsPointSystem
ALMXXXX2017IOUH/Standard
CLMXXXX2015-2017IOUH/Standard
FLMXXXX2015-2017TruScor
GGibbs SamplerXXXX2015-2017TruScor
HLMXXXX2017TruScor
IMany LMsXXXX2015-2017IOUH/Standard

 

Model A:

.Model A-1.png


Model A parameters:

The team that’s playing
Who their opponent is
Their predicted score
Their opponent’s predicted score
Whether the team is home or away
Model A Notes:

Uses only data from 2017
Standard points
Did well in 2016, but didn’t stand out


Model C:

.Model C-1.png


Model C parameters:

The team that’s playing
Who their opponent is
Their predicted score
Their opponent’s predicted score
Whether the team is home or away
Model C Notes:

Uses data from 2015-2017
Standard scoring
Did well in 2016, but didn’t stand out


Model F:

.Model F-1.png


Model F parameters:

The team that’s playing
Who their opponent is
Their predicted score
Their opponent’s predicted score
Whether the team is home or away
Model F Notes:

Uses data from 2015-2017
Truscor ignores defense TDs but weights interceptions and fumbles to try to account for it
I really liked this one in 2016. It was marginally better than A and C

Model G:

.Model G-1.png


Model G parameters:

The team that’s playing
Who their opponent is
Their predicted score
Their opponent’s predicted score
Whether the team is home or away
Model G Notes:

Uses data from 2015-2017
Truscor ignores defense TDs but weights interceptions and fumbles to try to account for it
Uses Bayesian statistics (Gibbs sampler) to create a model
Would have done very well in 2016 based on simulation


Model H:

.Model H-1.png


Model H parameters:

The team that’s playing
Who their opponent is
Their predicted score
Their opponent’s predicted score
Whether the team is home or away
Model H Notes:

Uses only data from 2017
Uses TruScor points to reduce the effect of TDs

Model I:

.Model I-1.png


Model I parameters may include:

The team that’s playing
Who their opponent is
Their predicted score
Their opponent’s predicted score
Whether the team is home or away
Model I Notes:

Data from 2015-2017
This model tries to predict all of the components of scoring separately (interceptions, sacks, etc.). Each component comes from a linear model containing only the significant terms (p < 0.05) from the list of possible terms above.

Week 6 model summary:

RankACFGHI
1BALDENDENDENBALDEN
2JACTBHOUHOUJACHOU
3TENHOUTBTBINDTB
4CARLARCARNENEATL
5INDINDNEWASCARKC
6CLEPHIWASCARCLEBAL
7DENKCKCBALHOUNE
8HOUNELARKCWASIND
9WASCARPHIATLDENCAR
10SFBALBALOAKPITLAR
11MIAATLINDLARTENWAS
12NEJACOAKINDSFPHI
13PITWASATLPHIMINMIN
14NOMINGBJACPHIJAC
15TBOAKJACGBNOTEN
16MINTENTENTENNYGOAK
17KCNYGNYGMINLARARI
18ATLARIMINARITBGB
19DETGBARINOCHINYG
20PHICLEPITNYGATLLAC
21LARPITNOPITKCCHI
22NYGLACLACLACGBPIT
23LACCHICLECHIMIANO
24GBNOCHICLEDETCLE
25CHIDETDETDETLACMIA
26ARISFSFSFARIDET
27NYJMIAMIAMIANYJSF
28OAKNYJNYJNYJOAKNYJ


Conclusions

My models break into two rough groups this week: the models tha have been slightly more accurate this year like DEN, HOU, and TB this week while the models that have been a little less accurate like BAL and JAC. I’d be surprised if any of these but TB are available in your league, but go take a look.

Some of these other top options should be available:

  • NE hasn’t been great but they’re up against NYJ and favored to win by a lot. They’ve done a little worse this year than expected, but last week really turned it around. Worth a good look.
  • CAR hasn’t done much this year, but they’ve done alright against some of teams that give up the fewest points to defenses. They did very poorly against NE and NO, but very well against SF, BUF, and DET. PHI gives up about as many points as those teams, so my models like CAR.
  • WAS has been good against some great teams (or at least onces that don’t give up a lot of points). But watch for Josh Norman; his injury may give you pause to put them up against SF. But not too much pause I hope.
  • LAR is up here, but I wouldn’t really trust them against Jacksonville who have given up the 3rd fewest points for the year.

If you’re like me and the dummies in your league stash defenses (WHY???), it’ll be tough to get some of the other top teams. One of those three I listed above should be available, and I like all of them this week. Yahoo’s experts have these in their top-10 as well, so pick one up.  They're better than me. 

Shit.  I really hate that.

Week 6 QB Predictions

Week 6 K Predictions