VOTED #1 FANTASY FOOTBALL SITE
1998, 1999, 2000, 2001 & 2002
PRIORITY NEWS   MESSAGE BOARDS JOIN   
HOME ARTICLES STATISTICS WEEKLY FEATURES TEAM LINKS NFL RESOURCES  
Analyze Your League Using LAG
by David M. Dorey
August 1, 2003
  During the summer, many of us spend countless hours poring over last year's NFL stats, player rankings, schedules and any shred of NFL information. There is so much to learn and discover each season, that people become obsessed with every shred of NFL news, no matter how inconsequential or actually misleading it might be.

Some use esoteric statistical analysis so refined that it relies heavily on using past numbers to guess future numbers, and then makes those guesses pretend to be an accurate enough to base more guesses on yet more guesses. Eventually it becomes like expecting an archaeologist to predict the future. He'd get a few things right, most things wrong and he would never foresee microwave pizza, TIVO or the Bucs success even without any running game. All pretty significant in my book.

The point is this - knowing your players is critical, obviously. But once you have accomplished a good knowledge base of players - reasonable performance expectations, who has upside, who is risky, what are the tiers of equivalent players and the like, stop trying to confuse yourself by extending your analysis deeper and deeper into less and less reliable or accurate conclusions. Turn your attention where it needs to be - your league.

Think about it - does it really matter if you can accurately guess the production of the #55th ranked receiver? Does it really matter if you were completely accurate about the actual order of the top 10 runningbacks for 2003 when the best you could possibly do is to draft one of them? No - what matters most is that in week 16 when your league meets that you get to sing your newly composed song entitled "Hi there, it's me - YOUR NEW DADDY" while dancing that ridiculously annoying Riverdance jig. You should play to win and that means more than knowing players - you must know your league.

If all you do is worry about NFL player projections and news, the best you can do is accurately predict which teams in your league will beat you. In short - you have to know your league. You cannot succeed without knowing some critical elements that will either make your cheatsheet roar or leave you with that ninth round "crash and burn" feeling.

What you need is a League Analysis and Graphing (LAG). It's pretty easy, fun and probably much more related to you winning than projecting kickers or using the standard deviation of anything.

League Analysis and Graphing

There are a few steps that you need to take in order to fully understand your league. While I will try not to get too technical, hang with me and by the end you will be able to do this in about 15 minutes for any league you are in. And you should do it for every league you are in.

Graphing - There is little more important about your draft and season than the scoring used by your league. This is not so much because it reveals who scored well last season, but because it is the blueprint for understanding your league and how point values apply to all those players.

If you have already played in your league before you will undoubtedly have access to last season's numbers. All we are really interested in are the top 20 players in each position and what they each scored for the 2002 season. If you do not have access to your league's scoring the year before or you are in a new league with different scoring rules, you can still determine what the scores would have been the previous season by using almost any league management product out there. Almost all will allow you to download a free copy or setup a league for no charge as they make their money when you buy their stats for the current year.

Setup a mock league using the scoring system you are interested in and then review what they stats would have been the previous season. Excellent products like Fantasy League Manager for the desktop or My Fantasy League online allow you to do this in addition to being robust league management tools.

For ease of discussion, let's use the most commonly used performance scoring system as our example:

Passing TD's are 4 points, -1 point for an interception and passing yards are divided by 20.
Rushing and receiving TD's are 6 points, and all rushing and receiving yards are divided by 10.
Kickers get 3 points per field goal and one point per extra point.
All two point plays award 2 points for the pass, run or catch.
Defense/Special Teams get 1 point per sack, 2 points per turnover, 2 points per safety and 6 points per TD.

After using that scoring system on the 2002 season, I placed the top 20 values for each position into a spreadsheet by copying them and pasting it into Excel (I used Fantasy League Manager, myself). You may have to remove players names or whatever comes out on the particular report you are using, but through whatever means you want to end with a spreadsheet that contains the top 20 scores in each position for last season using your league's scoring system. Once you have that done, you want to end up with something that looks like this (or optionally with the positional values in columns instead of rows):

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
QB 366 364 332 330 318 318 308 306 306 299 283 270 262 257 243 242 237 237 232 225
RB 373 324 307 289 272 270 265 259 256 228 224 219 217 209 207 203 201 190 185 184
WR 241 225 222 189 186 183 179 177 175 168 165 162 161 160 156 151 149 147 147 143
TE 125 119 101 92 87 86 79 76 73 70 69 63 63 62 57 56 53 52 48 47
PK 138 133 130 128 128 128 120 117 117 115 114 111 108 107 107 103 103 100 100 96
DEF 159 153 151 138 133 133 128 126 123 122 120 119 116 107 106 105 105 105 103 101

You can easily use an excel spreadsheet to create a graph by highlighting any table like the one above and selecting the chart wizard button. Once that pops up a small window, just select the line graph. For our purposes here, I only highlighted the top 12 players for our discussion but in your own analysis, string it out to the entire 20. If you want practice, just copy the above table and paste it into a spreadsheet. Then use the chart wizard to produce the table.

Using those values above, your chart will look like this:

So what are we looking for? Viewing the graph allows us to see the overall impacts of positions in the league scoring. Graphs are a great way to observe two characteristics for each position - how quickly the line declines and how it matches up against other positions. In this example:

QB - Definitely the highest scoring position but after the first three or so they decline slowly. Looks like the worst starting quarterback still scores more than all but the top three runningbacks. But the drop-off from fourth to tenth is not that dramatic.

RB - The second highest scoring position, I can see that the decline in the first five runningbacks is pretty dramatic, the next five will score about the same and then it falls to a lower tier which I already know declines at the same rate.

WR - While not as important as QB or RB, the first three receivers are a definite cut above the rest and then there is remarkably small difference in the next eight players. In all cases, WR's are higher scorers than TE, PK or DEF but are also always lower than QB and RB.

TE - Still seeing the top three phenomena, then it mirrors WR's in decline except about 100 points lower at every point. Tight ends are the lowest scoring position and even the best tight end only equals about the worst starting kicker or defense.

Kickers - Not only scores less than all others except tight ends, but it has the most flattened line of all meaning there is little difference between a kicker and the next one.

Defense - Top three make a difference over the rest, but after that they are little different than a kicker in terms of points. If you do not manage to get a top three defense, it appears that one defense is about the same as another.

So far I see top three players in pretty much any position are significant advantages, though the scoring drops pretty dramatically moving down from quarterback and runningback, to receivers then to defenses/kickers and finally tight ends. Just considering the overall look of what points are scored by position, I can already see that quarterbacks and runningbacks command a premium, though quarterbacks do not lose value from player to player as rapidly as runningbacks and as we will discuss later - we'll be starting two runningbacks each week as opposed to just one quarterback.

I can see that receivers are valuable at first and then very slowly decline but they score well above kickers, defenses and tight ends and I probably will need to start two - maybe three.

Kickers, defenses and tight ends are pretty much sad in this scoring system.

3-10-20 Scoring - We need to examine the first three, the tenth and the twentieth highest scorers per position more in depth. This will give us a great look at how the player scoring actually affects players in our league and how to truly value player positions in your draft. As we have shown many other times, fantasy scoring in the NFL is incredibly consistent from season to season - the only thing that changes are the player names beside the numbers. There is almost always a clear top three in a position and each year roughly the same scoring happens down the line within each position.

Let's take a closer look at the actual numbers:

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
QB 366 364 332 330 318 318 308 306 306 299 283 270 262 257 243 242 237 237 232 225
RB 373 324 307 289 272 270 265 259 256 228 224 219 217 209 207 203 201 190 185 184
WR 241 225 222 189 186 183 179 177 175 168 165 162 161 160 156 151 149 147 147 143
TE 125 119 101 92 87 86 79 76 73 70 69 63 63 62 57 56 53 52 48 47
PK 138 133 130 128 128 128 120 117 117 115 114 111 108 107 107 103 103 100 100 96
DEF 159 153 151 138 133 133 128 126 123 122 120 119 116 107 106 105 105 105 103 101

After the starters are taken for fantasy teams, each draft basically turns into filling out back-up positions and those players are where sleepers are searched for and where depth is created for your team. Our main concern in our analysis of the scoring is to determine how to value positions correctly and that means fairly strict adherence to the realities of getting starters. In the average league, that means one QB, two RB's, two or three WR's, a TE, DEF and PK. We also need to decide if some positions mandate taking backups before starters are taken for other positions.

Before we begin crunching a few numbers, let's remember something very, very important :

"Only about half of the players will meet expectations every year"

Hard as it is to believe, neither my rankings nor yours will be 100% accurate thanks to injuries, team dynamics and the wonderful bit of random luck/misfortune that occurs in the NFL each season (I still think mine will be better, but that's just me).This is pertinent in that it must be considered in the equation of when to take backups (AKA "covering your posterior" in case your starter picks were flops or get injured).

So then, let's use 3-10-20 analysis on this example league scoring, knowing what the graph has already shown us. Let's make this easier by doing some averaging that holds true every season. This will be a big help in valuing the positions within themselves and then as they relate to other positions.

  Top 3
Average
4th - 10th Best
Average
Top 3 vs 4-10
Pts. Lost
11 - 20th Best
Average
4-10 vs 11-20
Pts. Lost
QB 354 312 42 249 63
RB 334 262 72 203 59
WR 229 179 50 154 25
TE 115 80 35 57 23
PK 133 122 11 105 17
DEF 154 129 25 109 20

Quarterbacks - We now see that a top quarterback can score more than any other position but that falling back to taking a later one in the draft does not hurt as much in points as missing a top three runningback or receiver. Conversely, the drop-off in points from the 4-10 range to the 11-20 range is the highest in points of any position. This strengthens what we have often preached - do not be in a rush to get a quarterback but do be in a rush to get a backup quarterback. In case our starting QB does not perform to expectations, it will pay off to have a good replacement.

Runningbacks - No position declines as rapidly as runningbacks and only quarterbacks score as much as a runningback. And I will need two starters every week. If I get one wrong, I need a backup to fill in and I have to cover two bye weeks, not just one like for quarterback. I have to give this very early attention in my draft if not the first two picks. And I cannot wait to get a decent backup either unless I am very lucky landing a sleeper which this season is not shaping not so well with an absence of rookie starters beyond possibly Onterrio Smith and fewer unsettled situations.

Receivers - While receivers are less highly scoring than the previous positions, those top three are pretty valuable. I can gain 50 points taking one of them instead of any other position. If I do not have a high draft pick that can reach a top three RB, then I should consider the top three receivers if they are available. I just have to ensure I do not end up with extraordinarily poor runningbacks. If I do not get a top three receiver, the decline in points is very small waiting all the way to the tenth or even twentieth best receiver. I need to consider that I have to start two or more though, and I need to fill in for bye weeks and earlier pick flops.

Tight Ends - For all the hoopla over Gonzalez, Heap and Shockey, in this scoring system I really do not lose much missing them and taking another top ten tight end. Considering that I would have to use a pick for either a starting runningback, quarterback or receiver in order to reach them, the benefit does not seem to be a big payoff, particularly knowing that I need to get my quarterback and runningback backups early as well. There is just minimal value in taking a tight end early in this scoring system and I would be better off shooting for a sleeper tight end later on than squandering an early pick on one and only netting about one or two points per game over later tight ends.

Kickers - Yessh - already almost impossible to predict, even if I get the best kicker over a later one I am rewarding myself with a big single point per game overall. I am better off waiting with the rest on kickers and trying to find two later ones that I can mix and match each week. Ever heard of a stud kicker? Me neither.

Defenses - There is an advantage in this scoring system to owning a top defense and the couple extra points would be nice each week. The reality too though is that they too are hard to accurately pick and what is much more effective - since I am dealing with an entire team here - is to get at least two middle of the road defenses that actually match up well with each other week to week. Playing the matchups with defenses are a good idea, though in this scoring system I need a decent defense to rely on since it needs the top players to get those turnovers and sacks. I am in no hurry to get one, but I do not want to wait too long once they start to be taken.

Again - this shows the value of positions both within themselves and how they relate to each other. On one hand I know that not all picks are golden for a variety of reasons and I have to be good enough to know where I need to start taking backups. And when it is appropriate to take a backup player for one position when I do not yet have a starter for another position.

What this sort of analysis shows me about the league scoring is that there is no waiting on runningbacks. Knowing the high value and rapid decline of them, I want three runningbacks as quickly as I can, with only a quarterback and no more than two receivers taken before my third runningback if they are either top 3 players or at least I feel strongly enough about their chances to outplay most of the others in their position. I want a strong quarterback but there does not seem to be a rush to get one, just as long as I can get one in the first eight or so. Even if I wait on a quarterback, I can be okay, it just means I need to take a backup quarterback earlier in case I am wrong about my first QB or he gets injured.

Receivers score well, but there is little difference between the fourth and tenth one, nor even down to the twentieth one. Since I have to start two or three of them, I would not want to wait too long to begin taking them and I do not want to need to rely on any receiver taken after around the 40th one is taken or so.

In this scoring system, I can see there is an advantage to having a top three tight end, but nothing nearly as dramatic as it might seem. I would only take one if I felt he fell in the draft to maybe the sixth round or more and that is unlikely to happen with their "names" this season. Since tight end scoring remains so largely unchanged for the lesser players, I am in no rush until I see the others filling up on them.

Kickers? Sorry - no way I squander an early pick on one of them. There is just not enough difference in them to make one single kicker be considered a 16 week starting stud. Defenses are something that looks like offer a small advantage in owning top ones, though I am more likely to wait until the run begins and then grab two before the position gets too light.

This looks at the most common scoring used in fantasy football though we know that all sorts of flavors exist. The reality in different scoring is that it only tends to change the relationship of the position to all others, there is surprising little change in how players rank out in different scoring systems. The biggest two scoring methods that change the look of how players fall in a ranking is when quarterbacks have all yardage considered the same (which devalues rushing quarterbacks and raises the pure passers) and if there are reception points used since that rewards possession receivers and third down backs.

This sort of analysis is equally as valid for auction leagues as well. You must know how players are valued in your league in order to know what they are truly worth. Don't go blowing a huge chunk of your salary on Tony Gonzalez, for example, if it only nets you about 35 points a year when you can pay more for QB, RB or WR that will yield much bigger dividends in points.

Perhaps this looks involved, but it really isn't. Once you are accustomed to taking prior season results and turning them into spreadsheets and graphs, you can tell more about a league in 15 minutes than any other sort of effort. You can learn more about the league than other team owners that have played for years and who have been blinded by the perceived importance of some individual player they once had. It is about positional value, the relation to other positions in scoring and only then about the individual players.

Remember - the fantasy numbers for each position in your league will be very similar from 2002 to 2003 (unless you change scoring rules). Those numbers remain relatively unchanged each season, it is just that the players scoring them change. That does go back to knowing your players and there is no escaping that picking good or bad players makes a difference. But picking good players that respect a scoring system and uses it to make draft or auction choices puts you in the driver's seat.

Save the Last Draft for Me - Decidedly unscientific and the least reliable, it doesn't hurt to take a gander at last year's draft (if available) in order to see not only which team took which player, but what positions each team took in the first five rounds. By the fifth round, the big battle plan is already done and the rest of the draft is filling in and backing up. Great teams are often made so later in the draft with solid picks of sleepers or undervalued players. But that is player-specific, and going into the draft we want to know more about what the early rounds might be.

How likely is Bob to wait on a QB? Will Theodore continue to snap up Randy Moss at the first chance again? The patterns that people establish in the past most often repeat themselves, though draft position also has a lot to do with their selections. Gaining knowledge about your fellow league mates can help you better understand what players may be available to you in different rounds. People certainly can change from year to year, but normally they follow the same pattern if only regarding position.

If you are in a new league, learning the habits of your league mates becomes much more difficult. If they are online, there is always the chance that you can access some other league they were in. It all starts to smack of spying, but reason and the CIA will tell you that it is merely information gathering. Learning about your opponents tendencies and player preferences can give you a better feel for your draft and what may be possible. But one caveat - do not rely too heavily on it and make it the last consideration you have in your plans. More than anything, do not allow it to affect any plans you have on acquiring a specific player. If you really want a player badly, just take him and do not play waiting games.

Last year's draft results and the final scoring numbers for players is something that many of you have - don't ignore another tool.

Victories By Design

Analyzing your league's scoring rules by graphing and then using 3-10-20 analysis can give you a great read on the true importance of a position in that league, both within itself and how it relates to the other positions which is critical because when you use a draft pick, or spend limited salary cap dollars, you need to be maximizing the payback. You can only do that by knowing how the positions score and how quickly you need to establish depth for a position.

You can go into whatever depth of individual player analysis that you want or you can just borrow a decent cheatsheet, but the fact is that you will not be able to build an optimized team without knowing where to take players and how much value they can bring. Understand the positional scoring in your league and you can best design your team.

Once you are armed with this knowledge, you are already ahead of some other drafters who rely more on prediction and projection than what actually always happens. The final step - and a big one at that - is taking all this wisdom into your draft and then creating that winning team. We'll take that final step when we look at the importance of your draft position in the companion article - "Your Draft Slot, Your Team".

Until then - fire up that Excel while you are at work and get something productive done!