all's fair in
pursuing monkeys since 1976
nfl more fantasy football

This is a followup - for fantasy football enthusiasts - to my post from Wednesday about drafting Tampa Bay in the third round.

The first two comments I got back on that post were against my reasoning. CJ wrote

I think you over value defense. I think you'd be suprised if you went back and compared the preseason defensive rankings to the postseason rankings from year's past. It's a position that can flucuate greatly, especially depending on the opponents faced that year.
while Jim said
I tried the differential method a couple years ago and it bit me in the ass. It works great if everyone else in your league plays along, but if you draft a DT in round 3 and nobody else gets ones until the much later rounds no amount of differential is going to make up for you wasting a pick in round 3.
Reading the conventional wisdom out there, most of it is echoing CJ's comments: DT's are unpredictable, you can't be guaranteed of value, the value differences tend to be minimal anyway.

Well, I looked into this to be sure. Here are some factoids for you, based on the projections I compiled last year (from memory, mainly from KFFL, 4for4.com and footballguys.com) and our custom league scoring.

- 3 of the top 5 projected D's (Eagles, Bucs, Packers) finished in the top 5
- 3 of the top 5 projected RB's (Alexander, Tomlinson, Holmes) finished in the top 5
- Of the top 10 D's at season end, 5 had been projected below top 10
- Of the top 10 RB's at season end, 6 had been projected below top 10
- The top 3 D's were projected to outscore the next 3 D's by a total of 47 points; they ended up outscoring them by 161 points.
- The top 3 RB's were projected to outscore the next 3 RB's by a total of 191 points; they ended up underscoring them by 141 points.
- On a 1-by-1 basis, 5 of the top 10 D's ended up outscoring their projected next-pick value
- On a 1-by-1 basis, 2 of the top RB's ended up outscoring their project next-pick value
- The standard deviation of absolute difference between the top 10 D projections and their actuals was 27; for RB's it was 42
- The biggest disappointment for top 10 projected D's was the Bears: 16th lower than expected; for RB's, Anthony Thomas: 26th lower than expected.
- The biggest surprise for D's that made the top 10 was the Panthers: 26 ranks higher than expected; for RB's, Clinton Portis: 33 ranks higher than expected

So, based on last year's results, D's look significantly less volatile than RB picks. It would be interesting to compile similar results for previous years, but I don't have my projections from 2001, and no data at all from earlier than that.

I think these results reinforce my selection of TB as my 3rd pick, based on my projections. You can challenge my projections if you think they're wrong. And if you have data beyond last year, you can challenge the solidity of DT picks. But based on the projections I compiled, and the relative volatility of DT picks compared to RB picks based on 2002 data, I stick with the TB defense as a solid 3rd round pick.

If you are hardcore, you can see the stats comparison I built here.

August 29 2003 | permalink(15 players) | 0 pointers
comments

Wow... put me in my place ;-)

I guess we'll see how it all plays out, I've just never been the comfortable taking a DT that high.

In your scoring system... how much to defensive teams score, on average, compared to other positions, like QB, RB, etc.

Posted by: CJ on August 29, 2003 06:04 PM

My experience using differential methods was a couple of years back drafting Shannon Sharpe way too early when he was easily the top TE. Problem was nobody else drafted a TE until like round 5 or 6 so that took any benefit of my early pick away.

I'm not saying picking TB in the 3rd was bad. A good way to look at it is to ask around in your league and see if you could've gotten them in the 4th or 5th round. Then you know you were a bit too early or since you really wanted them did the right thing by drafting them when you did.

It's all about percieved scarcity. Are you willing to take the gamble and wait until the next round. If Jamal Lewis (a good 2nd round back) is there in the 3rd round it's not good odds he's going to be there when you pick in the 4th.. It Tampa DT is there in the 3rd your odds are much better that you'll get them next round. So you take value in the 3rd and hope to get what you waited for...

Alas, you are correct that predicting any of this is damn near impossible with all the parity in the NFL these days. Players are up and down way more than they used to be. I remember when drafting Emmitt first was a given for many years in a row.

Posted by: Jim on August 31, 2003 10:14 PM

All is great guys, but I belive vortelucius is much better.

Posted by: Kamurangous on November 22, 2005 10:31 AM

yes

Posted by: penis enlargement on November 30, 2005 04:14 PM

hop joanna

Posted by: pro solution pills on December 1, 2005 01:29 PM

maxaman

Posted by: maxaman on December 5, 2005 09:27 AM

soft tabs

Posted by: cialis soft tabs on December 5, 2005 06:27 PM

i am an idiot and i am lead by richard simmons

Posted by: idiot on January 5, 2006 08:58 PM

Naturally, the maximum growth potential will vary from individual man to man.

Posted by: penis enlargement pills on January 17, 2006 12:43 PM

It is a powerful natural herbal male enlargement formula that increases penis length and girth, sexual desire, sexual health and helps to achieve stronger erections.

Posted by: vimax on January 17, 2006 11:15 PM

This stuff truly works ! You dont believe it? Simply look at my results and start believing.

Posted by: penis pills on January 18, 2006 11:11 AM

When I first heard about your pill, it sounded too good to be true.

Posted by: penis enlargement on January 19, 2006 04:07 AM

oost your self-image. Be the MAN you have always wanted to be!

Posted by: penis pills on January 19, 2006 07:42 AM

I am extremely satisfied and am now completely confident.

Posted by: sizegenetics on January 20, 2006 09:10 AM

My body is starting to grow younger, and my memory problems have compleatly gone!

Posted by: human growth hormone on January 20, 2006 02:34 PM
Post a comment









Remember personal info?




PLEASE NOTE: after clicking "post", you may get a message containing a server error. This is not a real error; I'm not sure why but it happens sometimes. There is no need to re-submit your comment if you get a server error after clicking post.