Fantasy Football Today - fantasy football rankings, cheatsheets, and information
A Fantasy Football Community!

Create An Account  |  Advertise  |  Contact      

Staff Writer
Email Matt

Matt's Articles

The Weekly Gut Check - Vol. 127
Crank Score Projections

Rookie Scouting Portfolio The “Gut Feeling” is often synonymous with a sense of desperation resulting from a lack of preparation. The Gut Check is a huge proponent of studying the numbers, but there’s a point where one can place too much emphasis on the wrong information. This can result in the undervaluing or overlooking a player’s potential. Therefore, The Weekly Gut Check is devoted to examining the frame of reference behind certain number-driven guidelines that fantasy football owners use to make decisions.

Although The Weekly Gut Check doesn’t claim to be psychic, he does believe that he can dispel certain numbers biases and help you make the best choices for your team. We’ll keep a running tally of The Weekly Gut Check’s insights. This way you can gauge his views as something to seriously consider, or at least seriously consider running the opposite way as fast as you can!

The response to the new method for scoring Crank has been overwhelmingly positive so I am going to move forward with using it to project 2008 stats. I said I was going to write about projecting rookies, but I’ll focus on it next week when I provide my first complete set of 2008 projections for all positions. This week, I’m going to show you how I use an Average Value Theory-based method to making projections.

In a nutshell, Crank Scores measure how often a player missed, met, and exceeded, preset baselines of performance that a fantasy owner needed to obtain a “quality start” from that player. My theory is a fantasy owner will have more success in a head-to-head league with a non-auction draft if he factors consistent performance to desired baselines as a deciding factor among players rather than simply picking which player is projected to have the most points.

"I was asked on the forums to give an example of a team of consistent players out-performing another squad with higher fantasy totals at year's end, but was less consistent week to week." The tables below show the following information:

  • Two starting rosters
  • Each player’s fantasy point average from 2007
  • 2007 Crank Score for each player
  • The head-to-head record of each player
  • The head-to-head record of each team

If one of the players in a head to head scenario was on bye, I used the average fantasy points of the mid-point non-starting ranked player at each position to compete with the active player from the other roster in a 1 QB, 2 RB, 3 WR, 1 TE lineup. That means in a 12-team league, the substitute points were the 18th ranked QB and TE and the 36th ranked RB and WR by fantasy points per game:

  • Sub Par QB points = 14.9
  • Sub Par RB points = 8.3
  • Sub Par WR points = 5.1
  • Sub Par TE points = 4

I have to admit that this was a bit tougher to do than I expected, because I tried to take this example to the extreme to prove a point: Compile a team that had a lower fantasy point per game average than its competitor, but still won more head-to-head match ups. While it didn’t take me more than a couple of hours to do, I took the extra time to find players who were more consistent despite missing time due to injury so it wasn’t simply a matter of the lower average/more consistent team being healthier than the higher average/less consistent squad.

Team A Fpts/Gm Crank Record Team B Fpts/Gm Crank
Brett Favre 17.47 12 F 9-8 Ben Roethlisberger 19.11 9
Willis McGahee 12.8 20 CP 9-8 Clinton Portis 14.44 17
Joseph Addai 15.57 31 JA 9-7-1 Adrian Peterson 17.06 30
Braylon Edwards 14.06 35 BE 12-4-1 Andre Johnson 14.79 23
Wes Welker 10.56 22 WW 10-5-2 Anquan Boldin 11.73 17
Greg Jennings 12.62 29 TH 10-7 Chad Johnson 12.29 20
Kellen Winslow 8.79 16 KW 8-7-2 Antonio Gates 9.53 12
Team Avg 91.87 A 9-8 Team Avg 98.95

Team B had a total point per game average that was over seven points higher than Team A, but Team A actually edged Team B in head to head match ups among its players. I cheated a bit with Greg Jennings, who scored more per game than Chad Johnson, but I used him as a Team A player because the fantasy points per game averages between Jennings and Johnson were virtually the same, but Jennings missed time and still was more consistent. I could have made this a more dramatic difference if I gave Team B players such as Ronnie Brown, Larry Johnson, or Steven Jackson, which would have been more common. Here’s the actual head-to-head breakdown.

  Team A   Team B
Wk Fpts Winner FPts
1 70.4 B 125.4
2 111.6 B 120.4
3 104.3 A 78.1
4 96.3 A 63.9
5 76.9 A 62.3
6 89.7 A 89.4
7 83.1 B 96.5
8 124.8 A 69.7
9 112.6 B 118.5
10 75.9 A 71.7
11 89.4 A 76.2
12 112.3 A 87.2
13 71.7 B 88.3
14 95.4 A 63.4
15 59.7 B 90.9
16 49.8 B 96.7
17 57 B 94.8
Avg A 9-8
While a boom-bust player’s carry over points will help his team compensate for any of his teammates’ poor showings, I was actually being pretty kind by giving a significant number of fantasy points to any players on bye or injured. This helped the boom/bust team more than the more consistent squad.

In a real situation, one wouldn’t presume a fantasy owner found a steady-productive player once Andre Johnson or Ronnie Brown got hurt. Not everyone drafts depth effectively, nor do they hit on waiver wire picks even 30% of the time.

Crank isn’t meant for you to try to pick players who score fewer points per game, but to complement your search for high scorers with consistency so you don’t have to rely upon a big week from one or two players at the same time to keep your team competitive. It’s unlikely you’ll face more than three to five teams where the match ups aren’t more lopsided among 50%-75% of the players in your starting lineup than this example I went through here.

To review from last week, the New Crank Score sports these differences from the old method:

  • Missed games due to injury are factored into the New Crank Score.
  • Elite and #1 quality games are no longer given more weight.
  • Fantasy points per game average no longer dominate the formula.
  • The New Crank Score has a maximum number at each position.

We all know that projecting performance is not science, but a craft. There are numerous variables that go into building a winning roster. Examining consistency is only a small part of the equation, but any edge can help you. I’m sure you have notice that your fellow owners have become increasingly more knowledgeable about the game and prepared on draft day and they incorporate several factors into a draft plan:

My method for projecting Crank Scores will include the factors I mentioned earlier, plus incorporating two-year Crank scores much like one would do to calculate two-year fantasy point totals for the Average Value Theory. This way at least the scores/stats/performances are accurate for the ranking, even if the player selected for the ranking isn’t correct.

Here is the list of the top 32 QBs from a two-year measurement of Crank scores. The table shows the old Crank and New Crank scores. The players are sorted according to the new scores, but the number next to the player’s name reflects his rank according to the old score.

 Top 32 QBs: Two-Year Crank
Player w/Old Crank Rank Team Games Old Crank Fpts/G Elite #1 Sub Par New Crank 2-Yr Avg
1. Peyton Manning IND 32 48.65 19.7 15 24 8 31 15.5
2. Tom Brady NE 32 41.35 20.67 16 20 12 24 12
3. Tony Romo DAL 29 35.31 18.62 12 18 11 19 9.5
7. Drew Brees NO 32 30.82 18.27 14 18 14 18 9.0
8. Ben Roethlisberger PIT 30 30.72 17.39 11 18 12 17 8.5
6. Donovan McNabb PHI 24 32.01 18.74 9 14 10 13 6.5
10. Carson Palmer CIN 32 25.11 17.1 11 17 15 13 6.5
11. Brett Favre GB 32 23.67 16.12 11 17 15 13 6.5
5. Derek Anderson CLE 20 33.21 17.48 6 13 7 12 6.0
4. Michael Vick ATL 16 33.40 18.43 5 10 6 9 4.5
9. Kurt Warner ARI 20 25.67 16.56 7 11 9 9 4.5
13. Jon Kitna DET 32 18.81 16.27 9 15 17 7 3.5
14. Matt Hasselbeck SEA 28 17.05 16.47 9 12 16 5 2.5
16. Marc Bulger STL 28 14.77 14.77 8 12 16 4 2.0
12. Chris Redman ATL 6 21.19 14.13 3 3 3 3 1.5
20. Vince Young TEN 30 13.33 14.29 6 13 17 2 1.0
15. A.J. Feeley PHI 5 15.10 15.1 2 2 3 1 0.5
17. Luke McCown TB 5 14.41 14.41 2 2 3 1 0.5
18. David Garrard JAC 22 14.36 15.04 3 10 12 1 0.5
21. Sage Rosenfels HOU 12 12.34 13.46 3 5 7 1 0.5
19. Todd Collins WAS 4 13.91 13.91 0 2 2 0 0
22. Philip Rivers SD 32 11.38 14.01 6 13 19 0 0
23. Jake Delhomme CAR 16 11.30 15.07 4 6 10 0 0
26. Brian Griese CHI 12 8.83 10.6 2 5 7 0 0
27. Seneca Wallace SEA 10 7.94 9.93 2 4 6 0 0
28. Quinn Gray JAC 9 7.88 11.82 3 3 6 0 0
35. Chris Weinke CAR 4 2.67 10.69 1 1 3 -1 -0.5
24. Jay Cutler DEN 21 10.29 15.44 3 8 13 -2 -1
25. Damon Huard KC 21 9.22 12.1 1 9 12 -2 -1
31. Drew Bledsoe DAL 6 4.96 14.89 0 2 4 -2 -1
29. Jason Campbell WAS 20 7.80 14.18 3 7 13 -3 -1.5

As I hear (and agree) all the time, last year’s stats do not mean this year’s performance. Yet, last year’s stats do help us make some educated decisions. For instance, Peyton Manning has been the top quarterback for the past two years, but Tom Brady’s 2007 vaulted him to a spot close behind. When one examines the Pats and the Colts offenses for question marks heading into the 2008 season, one can argue while New England lost Donté Stallworth to free agency and Ben Watson is still recovering from an injury, Marvin Harrison’s return remains an unknown and Peyton Manning will miss most of training camp. Since Stallworth and Watson were far less integral parts of the Pats offense compared to Harrison’s role with the Colts and Manning is a workaholic who will miss the continuity of practice he normally gets, I’m more inclined to believe Tom Brady is the best candidate for the top spot for 2008’s projections.

At the same time, counting on Brady to have consecutive, record-shattering seasons is an improbable bet. This is where plugging in your top-rated player into the best score for the two-year period mitigates this problem. Instead of trying to predict Brady will have another season where he achieves a New Crank Score of 25 (which was the equivalent of 12 Elite Games, 15 #1 Games, and only 1 Sub Par Game), inserting Brady into the #1 spot based on the two-year average of 15.5 (the equivalent of 8 Elite Games, 12 #1 Games, and 4 Sub Par Games) is a far more realistic expectation.

Here is preview of my 2008 Crank Projections for the top 5 QBs based on this method.

 2008 Crank Projections: Preview...
Player w/Old Crank Rank Elite #1 Subpar New Crank Fpts/Gm
Tom Brady 8 12 4 16 18.76
Drew Brees 8 10 6 12 18.75
Peyton Manning 7 9 6 10 17.43
Tony Romo 7 9 7 9 18.46
Ben Roethlisberger 6 9 6 9 17.14

What I’m doing with these projections is simply plugging in the player who I believe will fulfill this two-year average Crank Score. This is very similar to how the Average Value Theory is applied. It’s a quicker method than projecting stats and fantasy points, because the Elite, #1, and Subpar columns are each fantasy point-driven. In this case, it’s for a 12-team league with 1 QB, 2 RBs, 3 WRs, 1 TE. For QBs here’s how the points stack up:

  • An Elite game is at least 21.10 fantasy points.
  • A #1 game is at least 16.42 fantasy points.
  • A Subpar game is less than 16.42 fantasy points.

Let’s use Tom Brady as an example to estimate his fantasy points per game using these figures. If Brady has 8 Elite Games, 12 #1 Games, and 4 Subpar Games, then we know that those 8 Elite Games also count among the #1 Games. Therefore there are only 4 #1 Games that didn’t score in the Elite category (12-4 = 8). To get an estimated fantasy point per game total for the 2008 season, I can simply do this calculation:

(8 Elite Games x 21.10 Fantasy Points) + (4 #1 Games x 16.42 Fantasy Points) + (4 Sub Par Games x 16.4) = 300.08

300.08 Fantasy Points/16 Games = 18.75 Fantasy Points/Game

The number looks too low for Tom Brady, doesn’t it? He averaged 25.9 fantasy points per game lat year, scoring 114 addition points than I credited him. In fact, Brady would have only been the seventh-ranked QB last year with a total I calculated.

What you may not realize is that this total would have made Brady the second-best fantasy quarterback overall in 2006 and the top signal caller in 2005. If you believe quarterbacks performances are trending upward, then my projection is too low. I don’t believe this is the case.

Another criticism I anticipate is how close together these five quarterbacks are two each other in terms of fantasy points per game. This simplified calculation of fantasy points doesn’t factor in how high or how low an Elite Game or Subpar Game can be, respectively. I counter this point by cautioning you to use the New Crank Score as the way you tier players, not fantasy points per game. I’m simply looking for an estimation of fantasy points per game; the Crank Score incorporates enough data for me to see separation alone. The fact that Brady is four points above Brees and six points above Manning is the significant number, especially when I begin applying these rankings across all positions.

 Max Crank Scores By Position
  Max Elite Max #1 Max #2 Max #3 Max Crank
1 QB 16 16 N/A N/A 32
2 RB 16 16 16 N/A 48
3 WR 16 16 16 16 64
1 TE 16 16 N/A N/A 32

Based the maximum Crank Score for each position, the most important positions to have are RB and WR. Based on my experience, it’s far easier to win a championship or at least field a playoff-caliber squad with an average starter or play the match ups with a committee of signal callers than it is to lack quality runners or receivers in your starting lineup. Inherently the Crank Score prioritizes which positions are important to draft.

Don’t mistake this table as reason to draft three receivers in the first three rounds! You have to take this information and apply it with common sense. The demand for runners is higher than any other position in most leagues. This chart should support your reasoning for avoiding the quarterback in the first three to five rounds and opting for surer possibilities at receiver.

Why did I plug in these five quarterbacks into these spots? What all five of these quarterbacks have in common is a quality receiving corps with either a strong TE or excellent slot receiver as a secondary option. Although Brady was phenomenal in 2008, there has never been a quarterback in the history of the game to repeat a top 20, all-time performance. Several have come close, but they could not remain healthy for the season. Tempering expectations on Brady will most likely save you from over valuing him. Drew Brees gets a bump, because of his second-half tear in 2007, an improved linebacker corps, no attrition at the skill positions, and the addition of Jeremy Shockey. Manning slips for the reasons I stated when comparing the Colts off-season to the Pats, but it’s obviously not a big drop when you look at the number of quality games I think he’ll have.

Next week, I’ll incorporate the New Crank Score projections across all positions and provide analysis for the preseason fantasy rankings.