Players On the Rise (1 Viewer)

[Tau837]

Players On the Rise

Ok I am a bit biased, but this is a MUST READ this fantasy season. You want to find those hidden gems in your draft? I present the method I have used year after year to great results. This article first appeared on the mrfootball website and was last published in 2001 (with the help of the Stats, Inc books). Thanks to Doug Drinen who extracted the data for me to write this piece.

Looking quickly at pro-football-reference and going back to 1931... there were 2544 instances where the 3years criteria (only that one - without looking at the age criteria) was met and 715 times the FFpts total in year N+1 was higher or equal to the FFpts in year N and 1829 times where the Fpts total was lower in year N+1... thus, a 28% 'sucess rate'...Unless I'm mistaken, looking only at the last 10 years (from '97 and up) and active players... the 'triple-criteria' (3years, age, same team) was met 283 times - and breaking out 62 times (FFpts going up in year N+1)... 22%...

Some notorious drops are, in my humble opinion:

T. Barber (69-98-226-168 in '98-'01);

P. Burress (27-136-174-109 in '00-'03);

K. Curtis (1-56-122-72 in '03-'06);

B. Franks (42-86-91-48 in '00-'03);

D. Givens (15-87-105-87 in '02-'05);

T. Gonzalez (74-150-174-129 in '98-'01);

A. Green (12-251-264-217 in '99-'02);

J. Harrington (124-173-195-122 in '02-'05);

L. Jordan (53-71-224-62 in '03-'06);

A. Lelie (68-79-150-91 in '02-'05);

S. Moss (3-71-177-115 in '01-'04);

J. Shockey (65-102-131-104 in '03-'06);

J. Stevens (7-52-85-47 in '03-'06);

H. Ward (96-132-219-182 in '00-'03).

Interesting info. I suspect that the 3 year criteria works best as a predictor when the third year total was still relatively low, like perhaps outside the top 20 at a player's position. I have no idea if that is true. To take an extreme example, LT meets the 3 year criteria... but how many people predict his fantasy points to go up?

 

[the UFFRM]

As far as the separate board, I don't think we will ever do that. We hear the arguments every year, but prefer having a consolidated board to talk about football.

I personally don't care about a separate board, I just want to be able to discuss some of the articles in detail. How can we do that?

How can we not do that? If you get some subscriber tidbits from a free board then it is serving its purpose. Remember when the internet was eventually going to be free? People with money will use it and people without money will have to read between the lines and look at the extra numbers. If anyone is scared about top secret classified fantasy football conjecture leaking out to the plebes then they are just insecure, paranoid, and stingy.

 

[BuckeyeArt]

Unless I'm mistaken, looking only at the last 10 years (from '97 and up) and active players... the 'triple-criteria' (3years, age, same team) was met 283 times - and breaking out 62 times (FFpts going up in year N+1)... 22%...

Some notorious drops are, in my humble opinion:

T. Barber (69-98-226-168 in '98-'01);

P. Burress (27-136-174-109 in '00-'03);

K. Curtis (1-56-122-72 in '03-'06);

B. Franks (42-86-91-48 in '00-'03);

D. Givens (15-87-105-87 in '02-'05);

T. Gonzalez (74-150-174-129 in '98-'01);

A. Green (12-251-264-217 in '99-'02);

J. Harrington (124-173-195-122 in '02-'05);

L. Jordan (53-71-224-62 in '03-'06);

A. Lelie (68-79-150-91 in '02-'05);

S. Moss (3-71-177-115 in '01-'04);

J. Shockey (65-102-131-104 in '03-'06);

J. Stevens (7-52-85-47 in '03-'06);

H. Ward (96-132-219-182 in '00-'03).

Great stuff. And quick. The next question would be "Is 22% meaningful?". You'd probably need to look at all other players (in the '97 and up data) who met the age, same team, and 3 years in the league (but not increasing each year) - and also apply some minimum FP's filter since someone who increases 3 straight years is likely not completely riding the bench - to see what the N+1 year was for that set (a control set). That would tell you whether this holds any predictive value at all.

 

[JayMan]

Interesting info. I suspect that the 3 year criteria works best as a predictor when the third year total was still relatively low, like perhaps outside the top 20 at a player's position. I have no idea if that is true. To take an extreme example, LT meets the 3 year criteria... but how many people predict his fantasy points to go up?

Great stuff. And quick. The next question would be "Is 22% meaningful?". You'd probably need to look at all other players (in the '97 and up data) who met the age, same team, and 3 years in the league (but not increasing each year) - and also apply some minimum FP's filter since someone who increases 3 straight years is likely not completely riding the bench - to see what the N+1 year was for that set (a control set). That would tell you whether this holds any predictive value at all.

Those are the two important points in this discussion...I only looked at the 'raw data' - to find the 'success rate'... one of the best example of miss (using only the raw data) is Shaun Alexander: 47-262-271 ('00-'02) and then 269 in '03... not exactly a 'breakout miss'... Obvisouly, some filters would have to be added to look more closely at the information to only check useful data... but I just wanted to point out that we can't blindly say that if a player meets all criteria - he 'most evidently' will breakout this coming season... this is an obvious exageration...

The second quote is very interesting since it would be useful to figure out if the 22% 'going up' is statistically meaningful - compared to other guys that meets the criteria excepts for the '3straight year going up' criteria and figure out if those that have been going up have a statistically better chance of breaking out in year N+1 than the others...

 

[Tau837]

Interesting info. I suspect that the 3 year criteria works best as a predictor when the third year total was still relatively low, like perhaps outside the top 20 at a player's position. I have no idea if that is true. To take an extreme example, LT meets the 3 year criteria... but how many people predict his fantasy points to go up?

Great stuff. And quick. The next question would be "Is 22% meaningful?". You'd probably need to look at all other players (in the '97 and up data) who met the age, same team, and 3 years in the league (but not increasing each year) - and also apply some minimum FP's filter since someone who increases 3 straight years is likely not completely riding the bench - to see what the N+1 year was for that set (a control set). That would tell you whether this holds any predictive value at all.

Those are the two important points in this discussion...I only looked at the 'raw data' - to find the 'success rate'... one of the best example of miss (using only the raw data) is Shaun Alexander: 47-262-271 ('00-'02) and then 269 in '03... not exactly a 'breakout miss'... Obvisouly, some filters would have to be added to look more closely at the information to only check useful data... but I just wanted to point out that we can't blindly say that if a player meets all criteria - he 'most evidently' will breakout this coming season... this is an obvious exageration...

The second quote is very interesting since it would be useful to figure out if the 22% 'going up' is statistically meaningful - compared to other guys that meets the criteria excepts for the '3straight year going up' criteria and figure out if those that have been going up have a statistically better chance of breaking out in year N+1 than the others...

 

[Lehigh98]

As far as the separate board, I don't think we will ever do that. We hear the arguments every year, but prefer having a consolidated board to talk about football.

I personally don't care about a separate board, I just want to be able to discuss some of the articles in detail. How can we do that?

How can we not do that? If you get some subscriber tidbits from a free board then it is serving its purpose. Remember when the internet was eventually going to be free? People with money will use it and people without money will have to read between the lines and look at the extra numbers. If anyone is scared about top secret classified fantasy football conjecture leaking out to the plebes then they are just insecure, paranoid, and stingy.

Dude, WTF are you talking about?