5 year Historical Norms (1 Viewer)

[Chris Black]

This is the freelance article that I submitted. Obviously, it didn't make the cut, but I thought I'd post it and see if I could get some feedback. I have the norms for the past 10 years (I only posted the last 5), and I'd be more than happy to e-mail them to anyone who would like them (blackcd@uga.edu). I got the idea for the article after I read a book that suggested using average performance over the past 3-5 years instead of projections for the up-coming season. The author suggested ranking players at a position, and then using norms to "project" their performance. I'm not sure that I completely bye into the theory, but it was fun to complile the numbers.

Projections are the lifeblood of fantasy football. They are the necessary foundation upon which the domination of you league is built. For those of you FFL veterans, you may remember the good old days when if you wanted projections you had toil away for hours in order to produce them yourself. Examine a player’s performance over several previous seasons, make your best guess for the up-coming season, and away you went. These days, obtaining projections is much easier given that just about every FFL magazine and web-site offers them in some for or another (Google 2006 fantasy football projections and prepare to be overwhelmed). If you’re a shark, and if you’re reading this my guess is you are, then I feel relatively certain that you aren’t going to take any projections at face-value. You have your own thoughts and opinions on how things will play out this season, and you’ll likely tweak projections to reflect how you think certain players will perform.

I don’t want to tell you how I think certain players will perform this year. That’s for you to decide. What I do want to offer is a statistical framework to aid you with your projections. While each individual season unique, if we look back over multiple seasons we can derive what I term “historical norms” for the number of points scored at a given position for a given end of season rank. For example, the running back that finished the season ranked 6th, or the 27th wide receiver, or the 8th quarterback. I have compiled lists of the average performance of the top 40 RB’s, top 50 WR’s, top 20 TE’s and top 30 QB’s over the last ten seasons. From these lists I calculated a set of “historical” norms from over the last 5 years. Also included are the most (Max) and least (Min) amount of points scored for a given rank. I also came up with a rough estimate of the total yards and TD’s (1 pt for every 10 rush/rec. yards, 1 pt for every 20 passing yards, 6 pts per rush/rec. TD and 4 pts per passing TD) required to reach the point total for each rank. Additionally, I calculated the value over the “worst” starter (based on a 12 team league with 1QB, 2 RB, 2 WR, and 1TE starting requirements) based on these norms.

I would recommend using the norms as simple guidelines for performance this year. The norms won’t tell you how an individual player will perform this year, but they can give you an idea of how points are likely to be distributed across a position. Consider the talent, opportunity and offensive situation of each player for whom you want to project. Look at the norms and determine if you think a certain player can realistically reach the yardage and TD numbers required to achieve a certain rank. For example, if you think Cadillac Williams will be the 6th best RB this season then based upon 10 year norms you could reasonably expect him to score 243 points. That works out to about 1770 total yards and 11 TD’s. Do you think Caddy can reach those levels of performance? Seems like a reach to me given that he likely won’t be incorporated into the passing enough to reach 1770 total yards and that he will likely lose goal-line carries to Mike Alstott. 1500 total yards and 8 TD’s seem more reasonable and would slot him in at #11 (indeed, this is exactly where Footballguys have him).

How will Caddy perform this year? Realistically, what should your projection set look like? Only you can make those decisions. However, I think a firm grounding in historical performance is good place to start. Enjoy.

Key:

Rank: End of season ranking based upon total fantasy points scored

Mean Pts: Average number of points scored by rank over either the last 5 or 10year.

SD: standard deviation of the average over the last 5 or 10 years.

Max/Min: the most or least amount of points scored by a given rank over the 5 or 10 year period

Total Yds and TD: My rough estimate of the total yards and TD’s required to reach th average point total for a given rank.

VDB #: Points above the “worst” starter (based on 12 teams, starting 1QB, 2RB, 2WR, and 1TE).

5 yr Mean Pts SD Max Min Tot. Yds TD VBD #

RB1 352 28 373 307 2260 21 206

RB2 316 28 345 277 2140 17 170

RB3 304 30 344 264 2020 17 158

RB4 289 20 311 262 1990 15 143

RB5 265 11 275 248 1870 13 119

RB6 254 17 270 231 1820 12 108

RB7 245 21 264 220 1790 11 100

RB8 233 16 259 218 1670 11 87

RB9 222 22 256 202 1620 10 76

RB10 210 14 226 194 1560 9 64

RB11 204 14 220 187 1560 8 59

RB12 200 14 219 185 1520 8 54

RB13 190 17 217 173 1420 8 45

RB14 186 17 209 168 1440 7 40

RB15 185 15 205 168 1430 7 39

RB16 181 16 203 164 1390 7 35

RB17 177 16 201 160 1350 7 32

RB18 172 12 188 157 1360 6 26

RB19 168 13 184 152 1320 6 23

RB20 165 14 183 147 1290 6 19

RB21 158 15 179 142 1280 5 13

RB22 155 12 171 141 1250 5 9

RB23 152 14 171 136 1220 5 6

RB24 146 9 157 135 1160 5 0

RB25 140 9 154 132 1040 6 -6

RB26 135 11 152 122 1050 5 -11

RB27 130 11 145 118 1000 5 -16

RB28 126 9 135 115 960 5 -20

RB29 122 10 133 109 980 4 -24

RB30 120 13 132 106 960 4 -26

RB31 117 8 126 105 930 4 -29

RB32 113 8 126 104 890 4 -33

RB33 111 8 124 102 930 3 -35

RB34 106 9 121 96 880 3 -40

RB35 103 6 111 96 850 3 -43

RB36 101 5 110 95 830 3 -45

RB37 98 6 106 92 800 3 -48

RB38 94 3 99 91 760 3 -52

RB39 90 1 91 89 780 2 -56

RB40 84 5 90 79 720 2 -61

5 yr Mean Pts SD Max Min Tot. Yds TD VBD #

WR1 245 13 267 237 1550 15 113

WR2 224 17 242 205 1400 14 92

WR3 207 9 219 196 1350 12 75

WR4 197 8 204 189 1310 11 66

WR5 191 6 201 186 1310 10 59

WR6 187 6 197 182 1270 10 55

WR7 184 8 197 178 1300 9 53

WR8 182 8 193 175 1280 9 50

WR9 177 8 187 168 1230 9 46

WR10 168 4 175 165 1200 8 36

WR11 166 2 168 163 1180 8 34

WR12 163 2 166 161 1150 8 32

WR13 160 4 164 153 1120 8 29

WR14 159 5 162 150 1110 8 27

WR15 156 4 160 149 1140 7 25

WR16 151 9 160 138 1090 7 19

WR17 148 8 159 136 1060 7 16

WR18 146 9 158 133 1100 6 15

WR19 144 9 156 132 1080 6 12

WR20 140 10 153 126 1040 6 9

WR21 139 10 153 126 1030 6 7

WR22 136 10 151 125 1060 5 4

WR23 135 11 151 121 1050 5 3

WR24 132 12 147 116 1020 5 0

WR25 130 9 139 116 940 6 -2

WR26 128 9 136 115 920 6 -4

WR27 125 10 135 111 950 5 -6

WR28 124 10 133 109 940 5 -8

WR29 123 10 133 107 930 5 -9

WR30 120 10 129 105 900 5 -12

WR31 118 10 127 102 940 4 -14

WR32 115 9 124 102 910 4 -17

WR33 112 7 121 102 880 4 -19

WR34 111 7 118 100 870 4 -21

WR35 108 6 117 100 840 4 -23

WR36 105 5 111 99 810 4 -26

WR37 103 4 108 99 790 4 -29

WR38 102 4 108 98 780 4 -29

WR39 101 5 107 96 770 4 -31

WR40 100 5 106 95 700 5 -32

WR41 97 6 105 91 730 4 -35

WR42 95 8 105 87 710 4 -37

WR43 94 8 105 87 700 4 -38

WR44 93 9 105 85 750 3 -39

WR45 91 9 104 84 730 3 -40

WR46 90 9 103 82 720 3 -42

WR47 88 8 100 82 700 3 -43

WR48 86 7 99 81 680 3 -45

WR49 85 8 99 81 670 3 -46

WR50 83 7 96 79 650 3 -49

5 yr Mean Pts SD Max Min Tot. Yds TD VBD #

TE1 150 23 174 123 900 10 80

TE2 133 20 168 119 850 8 63

TE3 109 21 134 89 710 7 39

TE4 98 18 119 74 620 6 28

TE5 95 17 118 73 590 6 25

TE6 92 16 112 73 560 6 22

TE7 87 12 103 72 570 5 17

TE8 83 9 95 72 530 5 13

TE9 77 6 86 70 530 4 7

TE10 75 7 85 66 510 4 5

TE11 73 7 82 66 490 4 3

TE12 70 6 77 63 460 4 0

TE13 65 5 73 60 410 4 -4

TE14 63 6 70 56 390 4 -7

TE15 60 7 68 54 420 3 -10

TE16 58 5 66 54 400 3 -12

TE17 56 4 63 53 380 3 -14

TE18 53 3 58 51 350 3 -17

TE19 51 3 55 48 330 3 -18

TE20 48 1 50 47 360 2 -21

5 yr Mean Pts SD Max Min PassYds pTD RushYds rTD VBD #

QB1 362 45 433 318 4300 34 100 1 99

QB2 352 42 418 310 4300 30 100 1 88

QB3 322 22 350 294 4100 28 100 1 58

QB4 314 20 329 281 4000 27 100 1 50

QB5 306 19 321 278 3950 26 100 1 42

QB6 300 17 315 274 3850 26 100 1 36

QB7 296 17 311 271 3850 25 100 1 33

QB8 292 18 310 267 3850 24 100 1 28

QB9 285 15 303 266 3700 24 100 1 21

QB10 280 12 295 262 3650 24 100 1 16

QB11 271 12 283 257 3600 22 100 1 7

QB12 264 5 269 257 3550 22 100 1 0

QB13 258 6 267 251 3600 20 100 1 -6

QB14 251 4 256 246 3500 19 100 1 -13

QB15 238 7 245 231 3450 17 100 1 -26

QB16 232 9 241 219 3400 16 100 1 -32

QB17 226 16 240 204 3300 16 100 1 -38

QB18 220 16 237 201 3300 15 100 1 -43

QB19 213 13 230 199 3200 14 100 1 -51

QB20 206 12 222 192 3200 13 100 1 -58

QB21 202 16 221 180 3200 12 100 1 -62

QB22 194 17 210 173 3100 11 100 1 -70

QB23 188 20 204 165 3000 11 100 1 -76

QB24 182 22 200 153 2900 11 100 1 -81

QB25 173 20 194 145 2800 10 100 1 -90

QB26 166 17 182 144 2700 10 100 1 -98

QB27 151 10 162 138 2600 8 100 1 -112

QB28 143 9 155 134 2500 7 100 1 -121

QB29 134 11 146 119 2400 7 100 1 -130

QB30 129 10 141 116 2300 7 100 1 -135

Notes on QB’s

QB performance has fluctuated widely over the last 10 years. This is likely the result of the fact that QB’s score points using not only their arms, but also their legs. Over the last 10 years “running” QB’s such as Michel Vick, Steve McNair, and Kordell Stewart have turned in top 10 seasons without posting gaudy passing stats. Dante Culpepper and Steve Young posted two of the greatest seasons ever by QB’s by combining significant rushing totals with huge passing totals. A look at the QB landscape this year seems to indicate that outside of Vick and Vince Young (if he plays) the chances of any QB’s posting a strong rushing season are slim. Age and injuries have caught up to the likes of McNair, McNabb, and Culpepper, all of whom have previously posted strong rushing totals.

Because of this, when projecting the passing yards and passing TD’s necessary to reach the historical norms I have chosen to apply the same number of rushing yards (100) and rushing TD’s (1) to each rank. In theory, this should allow you to see the type of passing numbers required to reach the historic norms. Looking at those numbers, two things jump out at me. First, without the benefit of rushing yards a QB would have to have a truly outstanding season to reach the historical norms of the top 2 QB’s. Outside of Peyton Manning and a healthy Carson Palmer I don’t think any QB can reach these lofty numbers this year. Second, the performance of the QB’s in the 7-15 range has improved over the last 5 years, thus closing the “value” gap between QB’s ranked in this area and those in the top 5. An easy conclusion to draw from this is that it doesn’t appear that any of the top QB’s (since they likely won’t benefit from strong rushing numbers) will outdistance themselves from the pack enough to present strong value this season. Once again it looks like waiting and grabbing a guy in the 8-12 range is the way to go.

 

[Musesboy]

That's an interesting way to look at it.

It seems logical that projections will generally be higher than historical norms due to the fact that injury will affect the production of some players. It would be interesting to compare seasonal productions to actual results.

Did you notice any trends at any of the positions? Moss and Holt put up huge numbers three years ago. The TE position seems deeper than in the past?

 

[Bri]

I don't see the use in this, please explain further

 

[MLBrandow]

You needed to give some specific examples on what your stats would indicate, and why they are as accurate/more accurate than case-by-case analysis.

I'm noticing a trend in these submissions that I've read, and it's that all of these articles seem to try to deliver some super profound message. I don't think that's how you win this competition (although I didn't read the rules).

To me, the key is a) great writing, b) finding some hard-to-explain anomaly, c) explaining that anomaly, d) verifying it with empirical analysis, e) making predictions, f) providing a glut of similar stats to look at and allowing the reader to come to his own conclusions.

You have to take into account that you are writing to an audience that is both well-versed in football and also has a pretty good knack for stats and player analysis. It's by default very difficult to please them at all, let alone a consistent basis, and let alone on a consistent enough basis to get some of them to pay you for your information.

Hope this helps. I'd like to consider myself a pretty good writer, and I can tell you that being able to publish your work before open-ended criticism is a large step to becoming a better writer and a more efficient thinker.

 

[Bri]

You needed to give some specific examples on what your stats would indicate, and why they are as accurate/more accurate than case-by-case analysis.

I'm noticing a trend in these submissions that I've read, and it's that all of these articles seem to try to deliver some super profound message. I don't think that's how you win this competition (although I didn't read the rules).

To me, the key is a) great writing, b) finding some hard-to-explain anomaly, c) explaining that anomaly, d) verifying it with empirical analysis, e) making predictions, f) providing a glut of similar stats to look at and allowing the reader to come to his own conclusions.

You have to take into account that you are writing to an audience that is both well-versed in football and also has a pretty good knack for stats and player analysis. It's by default very difficult to please them at all, let alone a consistent basis, and let alone on a consistent enough basis to get some of them to pay you for your information.

Hope this helps. I'd like to consider myself a pretty good writer, and I can tell you that being able to publish your work before open-ended criticism is a large step to becoming a better writer and a more efficient thinker.

kudos for much research and solid attempt though, don't ya think?

 

[MLBrandow]

kudos for much research and solid attempt though, don't ya think?

Kudos for the effort, but I think had he incorporated a bit of "if you use these numbers, you can do this, here let me show you" it would have been much more effective.I think if he releases a follow-up article it could be a solid read.

 

[Musesboy]

Yep. I think it's a good idea to establish the intent of the article in the opening paragraph.

Explain to the reader what they will gain by giving five minutes of their life to read it.

Then make sure that you deliver on your promise by the end of the article.

 

[Mark Wimer]

Yep. I think it's a good idea to establish the intent of the article in the opening paragraph.

Explain to the reader what they will gain by giving five minutes of their life to read it.

Then make sure that you deliver on your promise by the end of the article.

Musesboy,Yes, the organization you suggest is the classic way to organize any essay (which is essentially what a lot of fantasy football articles are, an essay articulating some issue with / idea about a particular topic). As a simple example:

This article is about how to write an article. Below, the three most essential sections of any article are presented and explained.

First, every article needs an introductory paragraph. This is the portion of the article where the writer states his argument/thesis (or the question that will be answered by the article's end) and presents a general outline for the rest of the discussion. By providing a clear "roadmap" for the reader, the writer is likely to gain and hold the reader's attention.

Second, every good article contains the body paragraph(s). In the body of the article, various aspects of the topic/arguments supporting the thesis are presented. Depending on how complex the topic being discussed is, the body section of an article may be quite long, containing many charts and etc. to illuminate the article's thesis or answer the central question. Ideally, within the body paragraphs, each paragraph will have a clear topic sentence, an explanation that illuminates the topic sentence, and then a conclusory sentence which ties the sub-topic discussed in the paragraph back to the main argument (thesis). Building a clear structure for each paragraph keeps the body paragraphs speaking to the central argument/theme of the article.

Third, every good article presents a conclusion. In the conclusion, the writer sums up the supporting arguments presented in the body paragraphs, and then shows how the main thesis/argument has been supported (or how the main question has been answered).

To sum up, when a writer sets out to write a clear article, he/she needs to tell the reader what they are going to be told in the coming article (the introduction); tell the reader what they said they were going to tell them (the body); and then tell the reader what the writer told them (conclusion). The above is the classic outline for a clearly written article/essay.

My .02.

 

[Musesboy]

Yep. I think it's a good idea to establish the intent of the article in the opening paragraph.

Explain to the reader what they will gain by giving five minutes of their life to read it.

Then make sure that you deliver on your promise by the end of the article.

Musesboy,Yes, the organization you suggest is the classic way to organize any essay (which is essentially what a lot of fantasy football articles are, an essay articulating some issue with / idea about a particular topic). As a simple example:

This article is about how to write an article. Below, the three most essential sections of any article are presented and explained.

First, every article needs an introductory paragraph. This is the portion of the article where the writer states his argument/thesis (or the question that will be answered by the article's end) and presents a general outline for the rest of the discussion. By providing a clear "roadmap" for the reader, the writer is likely to gain and hold the reader's attention.

Second, every good article contains the body paragraph(s). In the body of the article, various aspects of the topic/arguments supporting the thesis are presented. Depending on how complex the topic being discussed is, the body section of an article may be quite long, containing many charts and etc. to illuminate the article's thesis or answer the central question. Ideally, within the body paragraphs, each paragraph will have a clear topic sentence, an explanation that illuminates the topic sentence, and then a conclusory sentence which ties the sub-topic discussed in the paragraph back to the main argument (thesis). Building a clear structure for each paragraph keeps the body paragraphs speaking to the central argument/theme of the article.

Third, every good article presents a conclusion. In the conclusion, the writer sums up the supporting arguments presented in the body paragraphs, and then shows how the main thesis/argument has been supported (or how the main question has been answered).

To sum up, when a writer sets out to write a clear article, he/she needs to tell the reader what they are going to be told in the coming article (the introduction); tell the reader what they said they were going to tell them (the body); and then tell the reader what the writer told them (conclusion). The above is the classic outline for a clearly written article/essay.

My .02.

 

[Chris Black]

I apprecitate the comments. They will be helpful in shaping anything I write in the future.

In hindsight, I think perhaps the biggest flaw in what I wrote was that I didn't provide a clear argument about how or why it might be beneficial to use the information I provided. My initial idea was to gather some data that I thought was interesting, and then present it so that others could use it as they saw fit. From the opinions I've gotten so far, it seems that you guys would be more interested in how I think the numbers could be applied to projections this year, or how and why projections from previous years deviated from the norms I presented.

I have some thoughts on these subjects, but I'm going to take alittle time to better organize them before posting anything.

Thanks for all the feedback! This is what makes the board so great!

 

[Dinsy Ejotuz]

Chris, I find this information very useful. Can you tell us what scoring system you used?

 

[Chris Black]

I used standard scoring:

1 pt per 10 rush/receiving yards

1 pt per 20 passing yards

6 pts per rush/receiving TD

4 pts per passing TD

I didn't consider points for receptions or negative points for interceptions and fumbles.

I'd be happy to send you the actual excel spreadsheet that contains the year-by-year numbers for each rank at each position if you'd like (just e-mail me at blackcd@uga.edu)