Week 6 - Retrospective

Here is a retrospective on how my charts fared in week 6.  I hacked this together last night while watching Russell Wilson eat the Cardinals defense, so here are a couple things I didn't have time to perfect in this first iteration:

1. I pulled this data from http://fftoday.com/stats/playerstats.php?Season=2013&GameWeek=6&PosID=10&LeagueID=1, which doesn't seem to use yahoo standard scoring, so the numbers are off a bit (slightly higher than yahoo).  I'll try to adjust this next time, but for this week they'll have do.

2. Some names did not match (i.e. TY Hilton in fantasypros database vs T.Y. Hilton, Stevie vs Steve Johnson) so they have 0 points to them.  

3. I tried both median and average, and they're pretty similar.  I randomly chose to report median this week.

Tier 1 median: 21.4
Tier 2 median: 19.6
Tier 3 median: 20.7
Tier 4 median: 18.4
Tier 5 median: 21
Tier 6 median: 8.4

If not for Dalton and Foles' outstanding performances (which earned them Fedex air players of the week), who both sat in tier 5, the QB tiers would have looked much more accurate.

Romo fail boated, but I'd like to argue that it's unreasonable to expect everyone in tier 1 to do well. You'll see further down that every position has at least one player in tier 1 who tanked.

Thus, the better interpretation of these tiers is that you should expect:
- most players in tier 1 to score high (with occasional low performances),
- most players in tier 5 to score low (with occasional high performances, like Dalton and Foles this week)

Tier 1 median: 14.1
Tier 2 median: 12.1
Tier 4 median: 9.5
Tier 5 median: 5.4
Tier 6 median: 8.8 (4.2 w/out Ridley)
Tier 7 median: 3.6

If not for Ridley's breakout game of 23 points, Tier 1-7 would have been monotonic decreasing.  With just one week of retrospective so far, RB looks like the most predictable position.

Tier 1 median: 7.8
Tier 2 median: 6.9
Tier 3 median: 10.7
Tier 4 median: 4.2
Tier 5 median: 4.9
Tier 6 median: 4.6
Tier 7 median: 8.1
Tier 8 median: 3
Tier 9 median: 1.5
Tier 10 median: 3.5

I'm proud of myself last week for stating "Anyone in the top 3 tiers this week is capable of producing WR1 numbers".  You can see the separation between tiers 1, 2, and 3 against the rest.

Unfortunately both Dez and Megatron flopped this week, which brought down tier 1.  An unusual high number of breakouts in tier 7 brought that tier up. 

Tier 1 median: 4.6
Tier 2 median: 6.8
Tier 3 median: 2.5
Tier 4 median: 3.2
Tier 5 median: 0

Tier 1 was a disaster this week.  Graham, Witten, Gates.  Those were outlier performances (the bad kind) that no one would expect three of in one week. 

Tier 1 median: 6 
Tier 2 median: 8
Tier 3 median: 7
Tier 4 median: 6

Kickers look noisy, but if Prater hadn't only kicked PAT's, which resulted in his lowest performance of the year, kicker tiers could have also been monotonic decreasing.  

Tier 1 median: 5.5
Tier 2 median: 4
Tier 3 median: 2.5
Tier 4 median: 6 (Rams 21, Cowboys 11, Titans 10)
Tier 5 median: 2

The Rams, Cowboys, and Titans (21 pts, 11pts, 10 pts) dragged up the tier 4 median.  The other 6 teams in tier 4 all failed to score more than 6 points.  Other than that, the tiers were monotonic