Kevin Williams (1 Viewer)

[houndirish]

If anyone has figured out how to effectively start Kevin Williams could you please share the secret?

I noticed he carrying a pretty lofty ranking in this week's cheatsheets. I was trying to figure out why.

I'm carrying KW, Kiwi and Charles Grant in my redraft league. Sacks are weighted very high. I was leaning towards Grant and Kiwi but then saw Williams top 10 ranking on the cheatsheets. Can anyone shine a light on this. While you're at it, why is Grant so high? Are the Bengals still minus a Tackle?

 

[Rozelle]

If anyone has figured out how to effectively start Kevin Williams could you please share the secret?I noticed he carrying a pretty lofty ranking in this week's cheatsheets. I was trying to figure out why. I'm carrying KW, Kiwi and Charles Grant in my redraft league. Sacks are weighted very high. I was leaning towards Grant and Kiwi but then saw Williams top 10 ranking on the cheatsheets. Can anyone shine a light on this. While you're at it, why is Grant so high? Are the Bengals still minus a Tackle?

Williams has been a crapshoot.Also not sure why Grant is ranked high, Bengals are suppose to be poor matchup for opposing D-lineman.

 

[houndirish]

I just read that Levi Jones is out for Cincy as is one of their interior lineman. Although I'm not sure why that would inflate Grant's ranking. Will Smith would be facing Jones's replacement.

 

[The Man With No Name]

I've been wondering every week why he has such good projected stats. Here's what Norton has projected for him this season so far(by adding each weeks projections):

Tackles- 52........actual is 25

Assists- 14.........actual is 4

Sacks- 13.5.......actual is 5

F. Fumble- 1.5....actual is 0

P. Defense- 6......actual is 6

He has 3 solo tackles the last 6 weeks, and 0 sacks while his projections for those 6 weeks are 23 tackles, 7 assists, 7 sacks. WOW.

Only one week has he projected him for 0 sacks, 2 weeks with 0.5 sacks. I've been benching him a lot lately and am probably not alone.

 

[Biabreakable]

I don't think Dodds does the IDP projections but perhaps I am mistaken. I thought the old Red Eye crew did them.

Green bay has inexperienced rookie LG Colledge playing yet he has played suprisingly well for a rookie. One of the things Green Bay has improved this season.

Last time KW faced them he had one solo tackle. Thats it.

Not sure where the optimism comes from.

As far as Grants matchup Levi Jones has been out almost the whole year so that should not matter. It did early on but Jone's replacement Whitworth (also a rookie) has 14 games of starting experience now.

 

[-OZ-]

Only one week has he projected him for 0 sacks, 2 weeks with 0.5 sacks. I've been benching him a lot lately and am probably not alone.

He's one of the most frustrating IDPs in the league. He has the talent, and in any game he should break out big. But, waiting for that big game sucks. I've been starting him, but probably shouldn't. I generally consider anything the DT gives to be a bonus, but I'm beginning to think a consistent 3 tackles or so is better than the possible big game.

 

[Islander]

I've been wondering every week why he has such good projected stats. Here's what Dodds has projected for him this season so far(by adding each weeks projections):Tackles- 52........actual is 25Assists- 14.........actual is 4Sacks- 13.5.......actual is 5F. Fumble- 1.5....actual is 0P. Defense- 6......actual is 6He has 3 solo tackles the last 6 weeks, and 0 sacks while his projections for those 6 weeks are 23 tackles, 7 assists, 7 sacks. WOW.Only one week has he projected him for 0 sacks, 2 weeks with 0.5 sacks. I've been benching him a lot lately and am probably not alone.

Norton does the IDP projections I think.There are several players for which the sum of weekly projections is completely out of whack with actual numbers. Especially on the high side like K Williams. I plan on doing a summary like this during the offseason (for several players). I think the IDP projections still have quite some room for improvement because they are not really "expected values" for the upcoming week. They are often biased one way or another. If I could pick and choose every week which player I thought would be over/under the projection, I think my success rate would be well over 60%, and it's not because I am an IDP god, it's because the projections are clearly biased.

 

[Rozelle]

Only one week has he projected him for 0 sacks, 2 weeks with 0.5 sacks. I've been benching him a lot lately and am probably not alone.

He's one of the most frustrating IDPs in the league. He has the talent, and in any game he should break out big. But, waiting for that big game sucks. I've been starting him, but probably shouldn't. I generally consider anything the DT gives to be a bonus, but I'm beginning to think a consistent 3 tackles or so is better than the possible big game.

I feel your pain ... I gave up on him several weeks ago, he's buried on my depth chart.

 

[Islander]

... and he'll start in the Pro Bowl. Go figure.

 

[Biabreakable]

... and he'll start in the Pro Bowl. Go figure.

He is a good player. Don't make any mistakes about that. The numbers don't mean everything.

 

[Islander]

... and he'll start in the Pro Bowl. Go figure.

He is a good player. Don't make any mistakes about that. The numbers don't mean everything.

I agree he's a very good player and helped the Vikings run D reach top notch status this year. But I don't think he played to his 2004 level and did not see it as a Pro Bowl caliber year.

 

[jgb95]

I signed him this year to a 5 year deal at above market value.

I can cut him right now and take the salary cap hit now instead of in 2007.

What do you guys think??

I can also try and resign him to a more favorable contract in 2007.

He was the #8 DT this year in my league. We also have to start one next year every game (2 DE and 1 DT). My other DT's are Tommie Harris and Ellis Wyms.

 

[Jene Bramel]

Norton does the IDP projections I think.There are several players for which the sum of weekly projections is completely out of whack with actual numbers. Especially on the high side like K Williams. I plan on doing a summary like this during the offseason (for several players). I think the IDP projections still have quite some room for improvement because they are not really "expected values" for the upcoming week. They are often biased one way or another. If I could pick and choose every week which player I thought would be over/under the projection, I think my success rate would be well over 60%, and it's not because I am an IDP god, it's because the projections are clearly biased.

I'll be looking forward to the offseason summary. Thanks.

 

[John Norton]

Norton does the IDP projections I think.There are several players for which the sum of weekly projections is completely out of whack with actual numbers. Especially on the high side like K Williams. I plan on doing a summary like this during the offseason (for several players). I think the IDP projections still have quite some room for improvement because they are not really "expected values" for the upcoming week. They are often biased one way or another. If I could pick and choose every week which player I thought would be over/under the projection, I think my success rate would be well over 60%, and it's not because I am an IDP god, it's because the projections are clearly biased.

I'll be looking forward to the offseason summary. Thanks.

I will be looking foreword to that summary as well as I'm sure it will help me going foreword. Guys I can assure you that there is no "bias" involved in the projections. I don't know how you can even come up with such an assumption to be honest. I'm sure there are players that I am over/under on. If you were projecting over 400 players every week I'm sure we could sit back and pick apart your tendencies and mistakes as well. On the specific subject of Kevin Williams, when I look at the Vikings each week I try to gauge how many sacks, picks, tackles etc they will produce and then do my best to distribute those numbers among the players in a way that makes the most sense. Williams is incredibly inconsistent which makes him very difficult to predict. You never know when he will blow up for a big game. He's also one of the Vikings best defensive players. If you look at their team on a given week and say "I think they will have 3 sacks in this game", the chances are pretty good that Williams will be involved in one of them.If projections were the only thing I had to do each week it would be different but keep in mind that I do the weekly numbers, the going foreword numbers and the EOTG column each week on top of running a collision repair business, being a father etc. If I had an infinite amount of time to dedicate to just the weekly projections would they be more accurate? Sure they would. But I can't make a living doing projections alone. It's easy to sit back and pick them apart, try doing them yourself for an entire season before you begin casting stones. The other thing that most of you don't see is that the projections drive the weekly "rankings" so I have to be aware of how a player will rank based on what I project for him each week. If I put Williams down for a 1-1 in any give week I would have half the IDP communty calling me an idiot for "ranking" him outside of the top 150 DL. With thousands of readers looking at them from multiple points of view, I often find myself in a no win situation. All I can do is give my best effort. All aside, I do appreciate when you guys point out issues like this as all of us here at FBG strive to put out the best product possible. It's good to get other points of view. I most certainly do listen and when a point is brought to my attention I will always give it a closer look.Thanks for the comments and best of luck this week in the big game

 

[Islander]

If you were projecting over 400 players every week I'm sure we could sit back and pick apart your tendencies and mistakes as well.

I always thought this to be true. And I believe this would be true for almost anybody who would be doing the projections, even if it was their full time job. There are so many players and numbers that there will be some flaws. Sorry if my post sounded like I was casting stones, but my goal in my offseason summary will be to point out where improvements can be made. Weekly projections will probably be imperfect forever, but in the spirit of continuous improvement, if we can point out how to make them better, then why not. I don't know of a better source of IDP projections anyway so it's not like I will stop subscribing if the projections don't improve I think no matter how good the projections become, we will always find feedback on how to improve them.In the ideal world, I think if you do your projections, and there are 50 FBGs here that point out where your numbers are off and then you agree based on the feedback and adjust them, the end result would be better. If I did my weekly projections and they were subject to scrutiny of 50 FBGs pointing out where I am too high or too low and I adjust based on the feedback, again the end result is better. So maybe next year we should have a thread in the IDP forum where people list which projections they believe are flawed and John decides whether to adjust or not. I would be in this thread weekly - I am not sure if many others would. It would only work if several people posted in that thread because no individual focuses on the same things. One thing I will suggest (and did so in the pinned "complaints" thread in the shark pool) is I would like weekly projections from more than one person, including offense. Giving us the ability to weight a few sets of projections will smooth out the biases each set of projections may have.

 

[John Norton]

If you were projecting over 400 players every week I'm sure we could sit back and pick apart your tendencies and mistakes as well.

I always thought this to be true. And I believe this would be true for almost anybody who would be doing the projections, even if it was their full time job. There are so many players and numbers that there will be some flaws. Sorry if my post sounded like I was casting stones, but my goal in my offseason summary will be to point out where improvements can be made. Weekly projections will probably be imperfect forever, but in the spirit of continuous improvement, if we can point out how to make them better, then why not. I don't know of a better source of IDP projections anyway so it's not like I will stop subscribing if the projections don't improve I think no matter how good the projections become, we will always find feedback on how to improve them.In the ideal world, I think if you do your projections, and there are 50 FBGs here that point out where your numbers are off and then you agree based on the feedback and adjust them, the end result would be better. If I did my weekly projections and they were subject to scrutiny of 50 FBGs pointing out where I am too high or too low and I adjust based on the feedback, again the end result is better. So maybe next year we should have a thread in the IDP forum where people list which projections they believe are flawed and John decides whether to adjust or not. I would be in this thread weekly - I am not sure if many others would. It would only work if several people posted in that thread because no individual focuses on the same things. One thing I will suggest (and did so in the pinned "complaints" thread in the shark pool) is I would like weekly projections from more than one person, including offense. Giving us the ability to weight a few sets of projections will smooth out the biases each set of projections may have.

I didn't mean to sound defensive. Part of the job is accepting the scrutiny of several thousand very knowledgeable football people. I absolutely agree that feedback like yours is a good thing and constructive criticism is always welcome. As I said I/we are always looking at ways to improve the product. It's something that we discuss at length during our off-season brainstorming session each year. I just wanted to explain the process a little. There are a lot of people looking at the projections from different perspectives and the better everyone understands the process and what it is expected to provide, the more help you are likely to be when it comes to making improvements or when you see things like the Williams situation. When you complete your summary please pass along the results (norton@footballguys.com). I am very interested in seeing them. This is something I would love to do myself if only time would allow

 

[cr8f]

... and he'll start in the Pro Bowl. Go figure.

I know. That was a shock.

 

[Buckle]

I'm starting to think the projections are a little overrated in most circles. If I look at some projections at a certain position like DB or WR, the difference between the 20th ranked player and 40th ranked may not be that much. In some cases, it may be within a point or 1.5 points.

One time I did projections for a site and doing them for WRs was tough. After the 20th guy and on, everyone starts looking the same!

Maybe a different approach should be used like classifying players in a starter slots like LB1, LB2 or maybe in groups like studs (must start), solid starters, matchup starters, tempered expectations (good use of past production data on team facing that week), or bench. I would imagine it would be easier to rank players this way on a weekly basis, but I know this would not work for some people who may have an unusual scoring system (i.e. heavily weighted towards sacks for example).

 

[Islander]

I started working on the summary I mentioned above and ran into a problem. I do not have the week #5 stats. If you go in the week #6 link, and then click on "Final Week 5 Stats - Dodds", it gives you the week #6 stats rather than the week #5 stats.

Can someone fix this? Thanks!

 

[Islander]

Year-to-date stats as of Week #5 are also missing.

"Final YTD PPG Stats - Dodds"