Workshop: Projections and how we arrive at our conclusions (1 Viewer)

[Ministry of Pain]

I have downplayed the legitimacy of projections for the many past seasons. For a variety of reasons I think they are very dangerous. They get really dangerous when people get married to their work. Some projections are better than others. I typically like what FBG posts preseason as they typically err on the side of caution.



But if we are going to run projections I think it’s only right to try and do it right. I see people post player projections and I am never sure how they arrive at their numbers. When you ask a lot of people where they get their numbers form they can’t seem to tell you. I would propose that most owners pull the year end stats out for the team instead of the player. Instead of trying to figure out how many carries that DWill and JStew are going to each have, let’s first look at John Fox and the Carolina Panthers rushing attempts per year the past few seasons. Then when we determine how many attempts they will have, then we can start breaking it down to a 55/45 split or whatever. I do the same with the passing game.



It is early, only April and we don’t know what impact the rookies will have yet but we can start to determine if we see a surge in rushing attempts or pass attempts, etc…So I am going to start listing the teams as they finished 2009 and see if we can determine an increase or decrease based on trends or other info. I'll start with the #1 rushing team and work down form there, it just so happens to be the Jets.



New York Jets

Rushing: They rushed the ball 607 times last year. An increase of almost 200 attempts over a season ago. A lot of this I attribute to Rex Ryan but also the rookie Sanchez left them with few options at the moment. I see them rushing the ball a little less but I also see them winning more football games this year, and if that happens then they will be pounding the ball to finish games, not zipping it all over the place.

Passing: They were #1 in rush attempts, #32 in pass attempts. The NFL avg per team is roughly 532 attempts…The Jets threw it a total of 393, almost 50 less than the next team and they blew away anyone close to them in rush attempts…next one was 525 by Carolina.



They have to move back to the avg mean but I don’t think leaps and bounds. I would still give them about 550 rushing attempts…factor in 50 of those going to the QB and RBs you would never draft and you have about 500 carries to spread over Greene/LT/ and maybe Leon Washington but we don’t know what he will do. I would probably give Greene about 40-45% of the carries which is roughly 200-225 carries…maybe 240-250 but that still leaves 250 to split up between LT and then Leon…IMHO, Washington is the odd guy out here with his injury situation. I don’t see NY abusing any one RB and trying to ensure they have some juice left for the playoffs. If one back were to really distance themselves from the rest, I could see a scenario where you might have Green in the 250-275 range but honestly 300+ IMO is a pipe dream for him right now. And don’t be shocked if New York takes a RB in the draft at some point.



Sanchez was only able to complete 53% of his passes, and he threw 20 picks but he was a rookie and he does have talent around him. I would probably pencil Sanchez in for 425-450 pass attempts. I would increase his completion % to about 55-56%, and you end up with an increase from 196 completions to roughly 240 completions. That’s a lot more receptions to be gobbled up from the WR/TE/RB spots. Sanchez will get more and more confident.



I likely would spread those completions around but you have to assume both WRs can improve upon what they did form a year ago…maybe not huge numbers but about 5-10% increase across the board for guys like Cotch, Braylon, and Dustin Keller…this is still far from a dominant passing game but there are points to have.



My projections roughly for the Jets right now would look something like this...



Mark Sanchez- 240/450/3000/16TD/12Int

J.Cotchery-65/850/4TD

B.Edwards-50/700/5TD

D.Keller-56/620/4TD

Shonn Greene-240/960/10TD…not much of a receiver

LaDainian Tomlinson-180-200/700/10TD, and 40/300/2TD…The Jets scored 21 TDs rushing last year, I expect that continue.



I will eventually run thru all 32 teams but I wanted to gauge the SP interest in this and also see what others do for their stat projections. I think you really have to analyze what coaches have done the last 2 years with the personnel they have on the rosters. You can’t always gauge that however when you have an influx of new coaches. But many of the HC were former OC/DCs so you can look at what those teams did and get an idea for what the new coach is going to do.

Thanks,

MOP

 

[moleculo]

 

[Chase Stuart]

Here were my projections:

Player Pos Comp Att Perc Pyd Y/A PTD INT Rsh Ryd YPC TD Rec Ryd YPC TD M. SANCHEZ QB 224 385 58.2% 2705 7.0 16 12 30 85 2.8 1 K. CLEMENS QB 32 55 58.2% 325 5.9 3 2 7 18 2.6 0 S. GREENE RB 235 1011 4.3 8 6 30 5.0 0L. TOMLINSON RB 141 578 4.1 7 15 110 7.3 1L. WASHINGTON RB 94 425 4.5 3 30 233 7.8 2T. RICHARDSON FB 5 25 5.0 0 2 12 6.0 0D. KELLER TE 54 645 11.9 5B. HARTSOCK TE 5 45 9.0 1J. COTCHERY WR 1 5 5.0 0 70 910 13.0 4B. EDWARDS WR 1 5 5.0 0 50 740 14.8 5D. CLOWNEY WR 2 16 8.0 0 15 200 13.3 1B. SMITH WR 15 80 5.3 1 6 65 10.8 0 OTHER 3 40 13.3 0 TOTAL 256 440 58.2% 3030 6.9 19 14 531 2248 4.2 20 256 3030 11.8 192009 210 393 53.4% 2596 6.6 12 21 607 2756 4.5 21 210 2596 12.4 12
I agree on the Jets passing about 450 times for about 3000 yards; I think Sanchez is pretty aggressive, so I've got more TDs and INTs than you. We mostly agree on the receivers, although I trust Cotchery more than Braylon relative to you. For Greene, we're also almost right on the money there, too (both projecting about 150 FPs). I've got the Jets projected at 19 TDs; I think you need to recognize that Sanchez/Washington/WR will probably steal about 4-5 TDs. I also have more faith in Washington than you. But mostly our numbers are very similar.

 

[Crippler]

I dont do projections in the end but this is some excellent work in the end. Well written and thought out and how I might do it if I took the time.

Maybe more past rushing and passing history(3 years) would be nice(even if tough for the Jets with new coach in place).

These projections for Jets look good even if I like LT to score a few more TD`s than you have.

 

[David Yudkin]

You haven't posted the team totals so I reserve the right to change my opinion, but . . .

The Jets had 607 total rushing attempts last year. That's not just high, that's very, very high. They were only one of five teams since 1987 to hit 575 team carries.

2004 Steelers 618 carries . . . 2005 Steelers 549 carries

2009 Jets 607 carries . . . 2010 Jets ???

2008 Ravens 592 carries . . . 2009 Ravens 468 carries

2001 Steelers 580 carries . . . 2002 Steelers 512 carries

1988 Patriots 588 carries . . . 1989 Patriots 485 carries

If we go back one more year . . .

1986 Bears 606 carries . . . 1987 Bears 517 carries*

1986 Falcons 578 carries . . . 1987 Falcons 355 carries*

1986 Rams 578 carries . . . 1987 Rams 546 carries*

(*prorated for 16 games vs. 15 games played that year)

Bottom line, I don't think we really know that the Jets are going to run anywhere near as much as they did last year, and history tends to support that the Jet ran a lot and might not be expected to carry that over this year.

As for the Jets 2010 rushing projections, I still think Leon Washington is the most explosive back on their roster, and if he's still around he will still get the ball some, so I would still pencil in him for 5-7 carries a game especially considering he's had a 5.2 ypc over the past 3 seasons.

 

[Ministry of Pain]

You haven't posted the team totals so I reserve the right to change my opinion, but . . .The Jets had 607 total rushing attempts last year. That's not just high, that's very, very high. They were only one of five teams since 1987 to hit 575 team carries.2004 Steelers 618 carries . . . 2005 Steelers 549 carries2009 Jets 607 carries . . . 2010 Jets ???2008 Ravens 592 carries . . . 2009 Ravens 468 carries2001 Steelers 580 carries . . . 2002 Steelers 512 carries1988 Patriots 588 carries . . . 1989 Patriots 485 carriesIf we go back one more year . . .1986 Bears 606 carries . . . 1987 Bears 517 carries*1986 Falcons 578 carries . . . 1987 Falcons 355 carries*1986 Rams 578 carries . . . 1987 Rams 546 carries*(*prorated for 16 games vs. 15 games played that year)Bottom line, I don't think we really know that the Jets are going to run anywhere near as much as they did last year, and history tends to support that the Jet ran a lot and might not be expected to carry that over this year.As for the Jets 2010 rushing projections, I still think Leon Washington is the most explosive back on their roster, and if he's still around he will still get the ball some, so I would still pencil in him for 5-7 carries a game especially considering he's had a 5.2 ypc over the past 3 seasons.

If the Jets ran it 550 times...figure 50 to the QB and other iffy backs...240 for Greene, 180+ for LT, It leaves about 130 carries on the table which fits within your 5-7 carries a game for Leon. Factors though for the Jets that might make them slightly different than other teams. They had a rookie HC and rookie QB, they kept it simple. but Ryan is the offspring of Buddy Ryan who would like to put a stop to the forward pass in the NFL, he's oldskool. The Jets felt they needed another RB on top of Greene and Washington so they obviously are a team that will emphasize the run. They have a top5 rushing OL, that mattersSanchez will still only be in year two and there is no reason to think they start zipping tha ball all over the place. They went 8-8 last year, this season many are projecting 10+ wins, and if that happens they will have the opp to pound the football even more late in football games. I feel when you mix in coaching philosophy, a 2nd year QB, a good strong defense we didn't mention which should set up short fields and plenty of scoring opps...Are you wanting to move New York to 500 pass attempts? That would be a big leap and I agree with you that their Rush attempts are skewed. Nice feedback DY.

 

[David Yudkin]

IIRC, they went 9-9 not 8-8. IMO, the Jest had two statical quirks that bug me. Their running attempts were off the charts -AND- their defense was off the charts.

I suspect that no matter how good they look on paper, the defense will not do as well this year. They ranked first in both points and yardage allowed. I doubt that will happen again.

I think if the Jets are to win 10-11 games this year (against what looks like a tougher schedule), then they will have to pass more. I don't think they can let the defense win games for them without more production from the passing game.

Here's their schedule for those that are interested . . .

Home: BUF, MIA, NE, BAL, CIN, GB, MIN, HOU

Away: BUF, MIA, NE, CLE, PIT, CHI, DET, DEN

Off the cuff, I think NYJ pass more, run less, give up more points, and score more points (but rushing totals go down and passing totals go up).

I'm not sold that the Jets are as good as some people are making them out to be, but we shall see . . .

 

[Chase Stuart]

IIRC, they went 9-9 not 8-8. IMO, the Jest had two statical quirks that bug me. Their running attempts were off the charts -AND- their defense was off the charts.I suspect that no matter how good they look on paper, the defense will not do as well this year. They ranked first in both points and yardage allowed. I doubt that will happen again.I think if the Jets are to win 10-11 games this year (against what looks like a tougher schedule), then they will have to pass more. I don't think they can let the defense win games for them without more production from the passing game.Here's their schedule for those that are interested . . .Home: BUF, MIA, NE, BAL, CIN, GB, MIN, HOUAway: BUF, MIA, NE, CLE, PIT, CHI, DET, DENOff the cuff, I think NYJ pass more, run less, give up more points, and score more points (but rushing totals go down and passing totals go up).I'm not sold that the Jets are as good as some people are making them out to be, but we shall see . . .

I pretty much agree with all of this. One thing to note, though, is that (as I said all of last December/January) the Jets were a lot better than their record. Sure they were "only" 9-7, but based on their points scored and points allowed numbers, they had a Pythagorean record of 11.4 and 4.6. Now you might say "who cares what a team's Pythagorean record was?" And I'd say "well, a team's Pythagorean record is more closely correlated with a team's actual record than the next season than a team's actual record in the initial season. So, if you want to project how many wins a team will have next year, you're much better off if you look at Pythagorean record."The Jets lost five really close games last year (last minute losses to Miami twice, missed 3 FGs in a 3-point loss to Atlanta, overtime loss to Buffalo, last minute loss to Jacksonville) but didn't have many close wins. Close games are mostly decided by luck, and the Jets had a lot of bad luck last year (which is why their Pythagorean record was a lot better than their actual record, and it's why the Jets did so well in the playoffs). But yes, more passes, more passing yards, fewer rushes, fewer rushing yards, scoring more points and allowing more points all seem like safe bets. Even with Cromartie and Jenkins and maybe a better pass rusher, it would still be tough for an improved Jets D to put up the sick numbers last year's team recorded.

 

[Sinrman]

Yeah, I don't see the Jets running as much as last year, but still a lot. 550-575 attempts, I'm thinking. And personally I think they will use LT a lot more in the passing attack then any of the projections I see here. I think he will be used a lot on 3rd down for screens and dumps. It's something that the Jets haven't been known for, and it would be a nice mix to their offense and not make them AS predictable.

 

[Chase Stuart]

Yeah, I don't see the Jets running as much as last year, but still a lot. 550-575 attempts, I'm thinking. And personally I think they will use LT a lot more in the passing attack then any of the projections I see here. I think he will be used a lot on 3rd down for screens and dumps. It's something that the Jets haven't been known for, and it would be a nice mix to their offense and not make them AS predictable.

If Washington is healthy, I don't see this happening. But if not, then yeah, especially with Greene having terrible hands.(With Washington, the Jets ran a bunch of screens and dumps; Washington was one of the best third down backs in the league.)

 

[Ministry of Pain]

IIRC, they went 9-9 not 8-8. IMO, the Jest had two statical quirks that bug me. Their running attempts were off the charts -AND- their defense was off the charts.I suspect that no matter how good they look on paper, the defense will not do as well this year. They ranked first in both points and yardage allowed. I doubt that will happen again.I think if the Jets are to win 10-11 games this year (against what looks like a tougher schedule), then they will have to pass more. I don't think they can let the defense win games for them without more production from the passing game.Here's their schedule for those that are interested . . .Home: BUF, MIA, NE, BAL, CIN, GB, MIN, HOUAway: BUF, MIA, NE, CLE, PIT, CHI, DET, DENOff the cuff, I think NYJ pass more, run less, give up more points, and score more points (but rushing totals go down and passing totals go up).I'm not sold that the Jets are as good as some people are making them out to be, but we shall see . . .

Home...Buf(W), MIA(W), NE(W), BAL(L), CIN(W), GB(W/L), MN(W-Favre payback), HOU(W)Away...Buf(W). MIA(L), NE(L), Cle(W), Pit(L), Chi(W), Det(W), Den(W)They can win 10 games...how quick we forget that Rex Ryan who had a monster in Baltimore is now running things in New York, and with his ego there is no way he is going to think he should handle things differently. I'm a Miami fan, born here, live here now, and I hate the Jets, however...you gotta be honest about what they are doing. I don't dismiss you DY, you are without a doubt one of the best when it comes to statistical anlysis, both you and Chase so I am not surprised both of you checked in here. I'm trying to approach it more form that side this season as well but I can't in good faith pencil in Sanchez for 500 attempts. Do you think Sanchez will be under valued in redrafts this season? It's an honest and curious question because in a year where there are likely 8-10 teams that you want almost no part of in redrafts for QBs, at least in New York you are drafting the starter...it's not like Sanchez is getting benched. And unlike Henne in the same division he actually has a WR1, WR2, and solid TE with which to work.

 

[Chase Stuart]

Do you think Sanchez will be under valued in redrafts this season? It's an honest and curious question because in a year where there are likely 8-10 teams that you want almost no part of in redrafts for QBs, at least in New York you are drafting the starter...it's not like Sanchez is getting benched. And unlike Henne in the same division he actually has a WR1, WR2, and solid TE with which to work.

Sanchez almost certainly will be undervalued; the question is, what does that get you? If he's drafted as QB24, and he puts up QB18 numbers, he's undervalued. But I don't think his upside, at least for 2010, is particularly high.

 

[Ministry of Pain]

Do you think Sanchez will be under valued in redrafts this season? It's an honest and curious question because in a year where there are likely 8-10 teams that you want almost no part of in redrafts for QBs, at least in New York you are drafting the starter...it's not like Sanchez is getting benched. And unlike Henne in the same division he actually has a WR1, WR2, and solid TE with which to work.

Sanchez almost certainly will be undervalued; the question is, what does that get you? If he's drafted as QB24, and he puts up QB18 numbers, he's undervalued. But I don't think his upside, at least for 2010, is particularly high.

Agree, but the fact is he is reliable to the point that he will be on the field...Cleve, AZ, SF, Sea, St L, Oak, Buff, Wash, you really want one of those guys on your roster. If you grabbed a projected top5 or top10 QB that you plan to use a lot, then Sanchez becomes a nice roster filler later in the draft. I understand he doesn't have a high ceiling but sometimes you just need a guy behind a star for a bye week or a 1-2 week injury. Good thoughts Chase, thanks.

 

[David Yudkin]

Chase answered pretty close to what I would have said. But as far as fantasy goes, either you have an elite scoring QB on your roster or you are losing points from the QB position. I don't think Sanchez will make a huge leap to be worth a starting spot in pretty much any 1 QB fantasy league. He might be a below average QB2 . . . but as Chase said what does that get you.

As for the Jets and/or the Ravens, BOTH teams have been a bit of a shotgun blast over the past few years. The Jets have seen their record jump up by up to 6 wins in a season . . . or fall back by 6 wins in a season. THe Ravens have improved by up to 7 wins in a season . . . and have dropped off by up to 8 wins in a season.

As for Chase's assertion that the Jets were better than their record indicated and that they "should" have been 11.4-5.6, the same would hold for New England, who "should" have been 11.6-5.4. The Pats were one play/stop away from being 15-1 with late game losses to NYJ, DEN, IND, MIA, and HOU.

Long story short, an NFL season can turn on a handful of plays, and a number of teams in any given season could have had 8-12 wins if things worked out.

The reason why I think the Jets will struggle some is that I am not a big Sanchez believer, but certainly if he plays well the Jets could have a very good team.

 

[Sabertooth]

I project that not one prognosticator gets more then 5 NFL players (not bench warmers) stats exactly correct for the 2010 season.

 

[Ministry of Pain]

I project that not one prognosticator gets more then 5 NFL players (not bench warmers) stats exactly correct for the 2010 season.

Hey Sabe,I was banging that drum for many seasons and I still understand completely what you are saying and that is why I also believe that once you have the stats set up or done then you start putting guys into tiers. This way you have 5-6 guys with somewhat typical expected output and then you can just grab whichever falls to you. I agree it is far form perfect but it does have merit and it can allow you to maneuver around in your draft with some peace of mind as to where you want to go. The 1st 5-7 rounds pretty much we all know who is going to go where, but it's the middle rounds and late rounds where you find the guy that comes form nowhere and helps propel you into the playoffs. But I understand what you are saying.

 

[BuckeyeArt]

I have downplayed the legitimacy of projections for the many past seasons. For a variety of reasons I think they are very dangerous. They get really dangerous when people get married to their work. Some projections are better than others. I typically like what FBG posts preseason as they typically err on the side of caution.



But if we are going to run projections I think it's only right to try and do it right. I see people post player projections and I am never sure how they arrive at their numbers. When you ask a lot of people where they get their numbers form they can't seem to tell you. I would propose that most owners pull the year end stats out for the team instead of the player. Instead of trying to figure out how many carries that DWill and JStew are going to each have, let's first look at John Fox and the Carolina Panthers rushing attempts per year the past few seasons. Then when we determine how many attempts they will have, then we can start breaking it down to a 55/45 split or whatever. I do the same with the passing game.

I don't see the value in projections unless you are projecting on a PPG basis (or just projecting 16 game totals). Projecting end of year totals is meaningless if you are including a few weeks of injury. Everbody gets to make changes to their lineups weekly so using PPG when making projections makes much more sense. I want their potential numbers if they stay healthy. I don't really care if my projections are wrong because of injury.A good example is looking at Steven Jackson projections last year. Most were projecting Jackson for about 280 carries, which is about what he averaged over the last 3 years. What people should have been looking at were his average carries per game, which were about 21. Over 16 games, this would give him around 330 carries. Not surprisingly, he finished averaging over 21 carries per game. The latter projections are much more useful when sorting players.

 

[Ministry of Pain]

I have downplayed the legitimacy of projections for the many past seasons. For a variety of reasons I think they are very dangerous. They get really dangerous when people get married to their work. Some projections are better than others. I typically like what FBG posts preseason as they typically err on the side of caution.



But if we are going to run projections I think it's only right to try and do it right. I see people post player projections and I am never sure how they arrive at their numbers. When you ask a lot of people where they get their numbers form they can't seem to tell you. I would propose that most owners pull the year end stats out for the team instead of the player. Instead of trying to figure out how many carries that DWill and JStew are going to each have, let's first look at John Fox and the Carolina Panthers rushing attempts per year the past few seasons. Then when we determine how many attempts they will have, then we can start breaking it down to a 55/45 split or whatever. I do the same with the passing game.

I don't see the value in projections unless you are projecting on a PPG basis (or just projecting 16 game totals). Projecting end of year totals is meaningless if you are including a few weeks of injury. Everbody gets to make changes to their lineups weekly so using PPG when making projections makes much more sense. I want their potential numbers if they stay healthy. I don't really care if my projections are wrong because of injury.A good example is looking at Steven Jackson projections last year. Most were projecting Jackson for about 280 carries, which is about what he averaged over the last 3 years. What people should have been looking at were his average carries per game, which were about 21. Over 16 games, this would give him around 330 carries. Not surprisingly, he finished averaging over 21 carries per game. The latter projections are much more useful when sorting players.

I like it

 

[Chase Stuart]

Chase answered pretty close to what I would have said. But as far as fantasy goes, either you have an elite scoring QB on your roster or you are losing points from the QB position. I don't think Sanchez will make a huge leap to be worth a starting spot in pretty much any 1 QB fantasy league. He might be a below average QB2 . . . but as Chase said what does that get you.As for the Jets and/or the Ravens, BOTH teams have been a bit of a shotgun blast over the past few years. The Jets have seen their record jump up by up to 6 wins in a season . . . or fall back by 6 wins in a season. THe Ravens have improved by up to 7 wins in a season . . . and have dropped off by up to 8 wins in a season.As for Chase's assertion that the Jets were better than their record indicated and that they "should" have been 11.4-5.6, the same would hold for New England, who "should" have been 11.6-5.4. The Pats were one play/stop away from being 15-1 with late game losses to NYJ, DEN, IND, MIA, and HOU.Long story short, an NFL season can turn on a handful of plays, and a number of teams in any given season could have had 8-12 wins if things worked out.The reason why I think the Jets will struggle some is that I am not a big Sanchez believer, but certainly if he plays well the Jets could have a very good team.

You can certainly make those same arguments for the Patriots as especially those last four games all easily could have turned. OTOH, NE also had some wins like that, too: against Baltimore and against Buffalo on MNF were games NE was probably lucky to win. The Jets were one of the more imbalanced teams when it came to close games. But I agree that NE was arguably better than their record last year.

 

[Imagine]

MOP, Chase, David - I think this is a fascinating and educational thread, and really appreciate the work you do to get to the point where you are able to not only do the projections, but provide the basis for why you think that way. In this particular one, you have chosen the NYJ, and they are a great one to analyze for lots of reasons.

Another thing that was good about this thread was the agreement or disagreement by several people, most with reasons why they like or dislike the projections, or projections in general. Those opinions, especially the dissenting ones with reasons why, all provide excellent food for thought.

My question is whether a separate thread like this for each of the 32 teams would be worthwhile?

That's a lot of work, unless you already have the majority of your projections already completed. You’ll tweak them of course after the draft in a few weeks, but for the most part, they are probably in much the same shape as the NYJ ones that you three have each done. But it’s really fun to read what you guys think about each other’s projections, and where you agree or disagree. Kind of a peek inside the great minds!

So I guess that's my suggestion. Maybe do a couple teams a week, maybe by division? That means that in 3 to 4 months, we'd have 32 threads that talked specifically about projections for the Cardinals, the Seahawks, the Bears, and so on. Too much work? Maybe. But like I said earlier, this is a fascinating thread, and I for one would love to spend time reading not only what you three think, but also what the homers think about their team that they know more about than the average Joe out here.

For what it’s worth, there are a few teams out there that would provide extremely interesting analysis, because the results in 09 were so much different than expected. Injuries always play a huge role, but I can think of several players that didn’t come close to the consensus projections. Guys like Forte, Turner, LT, Jacobs, Lynch, and Slaton jump off the page when comparing their Actuals in 09 to the pretty much typical late summer projections. It would be fun as part of the 2010 projections to do an autopsy of why we were mislead last year on guys like this.

Thanks for a great thread guys.

 

[Ministry of Pain]

MOP, Chase, David - I think this is a fascinating and educational thread, and really appreciate the work you do to get to the point where you are able to not only do the projections, but provide the basis for why you think that way. In this particular one, you have chosen the NYJ, and they are a great one to analyze for lots of reasons. Another thing that was good about this thread was the agreement or disagreement by several people, most with reasons why they like or dislike the projections, or projections in general. Those opinions, especially the dissenting ones with reasons why, all provide excellent food for thought.My question is whether a separate thread like this for each of the 32 teams would be worthwhile? That's a lot of work, unless you already have the majority of your projections already completed. You'll tweak them of course after the draft in a few weeks, but for the most part, they are probably in much the same shape as the NYJ ones that you three have each done. But it's really fun to read what you guys think about each other's projections, and where you agree or disagree. Kind of a peek inside the great minds!So I guess that's my suggestion. Maybe do a couple teams a week, maybe by division? That means that in 3 to 4 months, we'd have 32 threads that talked specifically about projections for the Cardinals, the Seahawks, the Bears, and so on. Too much work? Maybe. But like I said earlier, this is a fascinating thread, and I for one would love to spend time reading not only what you three think, but also what the homers think about their team that they know more about than the average Joe out here.For what it's worth, there are a few teams out there that would provide extremely interesting analysis, because the results in 09 were so much different than expected. Injuries always play a huge role, but I can think of several players that didn't come close to the consensus projections. Guys like Forte, Turner, LT, Jacobs, Lynch, and Slaton jump off the page when comparing their Actuals in 09 to the pretty much typical late summer projections. It would be fun as part of the 2010 projections to do an autopsy of why we were mislead last year on guys like this. Thanks for a great thread guys.

Those kind words are much appreciated. Also your ideas for future threads. I like the 2009 autopsy you are talking of, FBG I'm guessing will do some of that in their preseason articles leading up to week 1. This is an excellent time of year for those that don't just open up in August getting ready for their draft, many of the posters in here right now follow football the way many follow the stock market. Also when you get your stats done early then you yourself have things in order as the summer rolls along and you can simply tweak as you were saying or make massive changes if need be. The real error that I see people make and I feel Saber was pointing out is the differences between what actually happens and what we "think" will happen. But the projections for me just help put guys into tiers...my tier3, 4, and 5 for WRs is pretty large, meaning guys you take in round 4, you can find others with similar numbers in rounds 7. But as another poster said, the PPG seems to be the key. All this is good feedback and now I can perhaps slant things to a PPG basis, or rushing attempts, all of this is excellent stuff.

 

[Sabertooth]

For what it’s worth, there are a few teams out there that would provide extremely interesting analysis, because the results in 09 were so much different than expected. Injuries always play a huge role, but I can think of several players that didn’t come close to the consensus projections. Guys like Forte, Turner, LT, Jacobs, Lynch, and Slaton jump off the page when comparing their Actuals in 09 to the pretty much typical late summer projections. It would be fun as part of the 2010 projections to do an autopsy of why we were mislead last year on guys like this.

Thanks for a great thread guys.

Turner was right on track when he got hurt. The others seemed to regress for one reason or another. But Turner was simply injured which goes to the theory of doing projections on a ppg basis.

 

[David Yudkin]

Prior results have no bearing on future results . . .

Teams that ranked #1 in points and yardage allowed and how they fared the following year:

2008 Steelers --> 2009 12 and 5

2006 Ravens --> 2007 22 and 6

2004 Steelers --> 2005 3 and 5

2002 Buccaneers --> 2003 4 and 5

1996 Packers --> 1997 5 and 7

1986 Bears --> 1987 4 and 2

1985 Bears --> 1986 1 and 1

1981 Eagles --> 1982 19 and 20

1979 Buccaneers --> 1980 20 and 20

1976 Steelers --> 1977 17 and 7

1972 Dolphins --> 1973 1 and 3

1970 Vikings --> 1971 1 and 2

1969 Chiefs --> 1970 8 and 6

 

[Iwannabeacowboybaby!]

I think most people thought the Jets heading into last season had a pretty good offensive line and stingy defense.

The big question was they were going with a rookie QB and with that comes a learning curve.

Will the Jets improve off last year? Well, they went to the AFC Championship game, to improve off that you need to at least make the Super Bowl. That's a tall order for anyone.

In terms of improving off their record from last year and just playing better football than last year, I think they can. Like David Yudkin, I'm still not totally sold on Sanchez becoming the next Joe Namath but I will say I saw some progression out of him and he wasn't the same guy in those playoff games as he was early on in the NFL season. You don't need to have an elite passing QB when you can play top notch defense and can run the ball. You need a guy who doesn't make costly mistakes and can make a play when you need him to. I do think Sanchez can be that guy, which makes the Jets dangerous with some of the additions they've made this offseason.

I think the addition of LT is a big help in terms of leadership and the guy can still play for what they'll be askinig of him. I think Jets fans have a lot to look forward to the next couple of seasons and should IMO unseat the Patriots in that division.

 

[Tau837]

My question is whether a separate thread like this for each of the 32 teams would be worthwhile?

I have suggested "Team Spotlight" threads to Jason Wood the past couple of years, but I gather that it hasn't been done because it is too much extra work on top of Player Spotlights and all the other preseason content that is being done around that time. I suppose we could do such threads for all teams now, but that would be of limited utility since the draft and all preseason developments are still to come... but waiting to make those threads more timely has presumably been too much extra work.The reason I suggested Team Spotlight threads is because I think too many people throw out projections in Player Spotlight threads without considering a lot of team-based factors that should be incorporated into player projections. I think this tends to cause people to overestimate passing attempts and touches for other skill positions.I don't personally do projections for all teams and players. But when I do projections for Player Spotlight threads, I tend to try to start by addressing the number of offensive plays I project the team to have... then break those down into passing attempts and rushing attempts... then, if I am projecting RBs, WRs, or TEs, break down targets at a team level, or if I am projecting RBs, break rushing attempts down at a team level.Then I'm ready to project an individual player's numbers, so I look at his past performance, going back up to a few years if the situation has remained similar (e.g., same coach/philosophy, similar teammates, etc.); try to assess what has changed on both offense and defense and how that will affect the player; take into account whether the schedule looks significantly better or worse; and try to assess whether I expect the player to be getting better, holding steady, or declining for the upcoming season... then I project his numbers accordingly.One issue that comes up is whether to project 16 games. To me, the worth of projections is for sorting/ranking players for drafting and trading purposes... I don't care much about correctly projecting end of season numbers. I also don't think it is useful for ranking purposes to try to project games missed, whether based on historical positional averages or even a player's own history... I prefer assuming 16 games for all projections, and then just incorporating injury risk for risky players at draft/trade time rather than incorporating that into the rankings.

 

[BuckeyeArt]

Prior results have no bearing on future results . . .Teams that ranked #1 in points and yardage allowed and how they fared the following year:2008 Steelers --> 2009 12 and 52006 Ravens --> 2007 22 and 62004 Steelers --> 2005 3 and 52002 Buccaneers --> 2003 4 and 51996 Packers --> 1997 5 and 71986 Bears --> 1987 4 and 21985 Bears --> 1986 1 and 11981 Eagles --> 1982 19 and 201979 Buccaneers --> 1980 20 and 201976 Steelers --> 1977 17 and 71972 Dolphins --> 1973 1 and 31970 Vikings --> 1971 1 and 21969 Chiefs --> 1970 8 and 6

Instead of just using the #1 teams, use all the teams and run correlations. It starts to become meaningful.

 

[Sinrman]

I have downplayed the legitimacy of projections for the many past seasons. For a variety of reasons I think they are very dangerous. They get really dangerous when people get married to their work. Some projections are better than others. I typically like what FBG posts preseason as they typically err on the side of caution.



But if we are going to run projections I think it's only right to try and do it right. I see people post player projections and I am never sure how they arrive at their numbers. When you ask a lot of people where they get their numbers form they can't seem to tell you. I would propose that most owners pull the year end stats out for the team instead of the player. Instead of trying to figure out how many carries that DWill and JStew are going to each have, let's first look at John Fox and the Carolina Panthers rushing attempts per year the past few seasons. Then when we determine how many attempts they will have, then we can start breaking it down to a 55/45 split or whatever. I do the same with the passing game.

I don't see the value in projections unless you are projecting on a PPG basis (or just projecting 16 game totals). Projecting end of year totals is meaningless if you are including a few weeks of injury. Everbody gets to make changes to their lineups weekly so using PPG when making projections makes much more sense. I want their potential numbers if they stay healthy. I don't really care if my projections are wrong because of injury.A good example is looking at Steven Jackson projections last year. Most were projecting Jackson for about 280 carries, which is about what he averaged over the last 3 years. What people should have been looking at were his average carries per game, which were about 21. Over 16 games, this would give him around 330 carries. Not surprisingly, he finished averaging over 21 carries per game. The latter projections are much more useful when sorting players.

Projections are fun. Easy way to pass time during the looooong offseason. Doesn't matter the scoring system.

 

[BuckeyeArt]

I have downplayed the legitimacy of projections for the many past seasons. For a variety of reasons I think they are very dangerous. They get really dangerous when people get married to their work. Some projections are better than others. I typically like what FBG posts preseason as they typically err on the side of caution.



But if we are going to run projections I think it's only right to try and do it right. I see people post player projections and I am never sure how they arrive at their numbers. When you ask a lot of people where they get their numbers form they can't seem to tell you. I would propose that most owners pull the year end stats out for the team instead of the player. Instead of trying to figure out how many carries that DWill and JStew are going to each have, let's first look at John Fox and the Carolina Panthers rushing attempts per year the past few seasons. Then when we determine how many attempts they will have, then we can start breaking it down to a 55/45 split or whatever. I do the same with the passing game.

I don't see the value in projections unless you are projecting on a PPG basis (or just projecting 16 game totals). Projecting end of year totals is meaningless if you are including a few weeks of injury. Everbody gets to make changes to their lineups weekly so using PPG when making projections makes much more sense. I want their potential numbers if they stay healthy. I don't really care if my projections are wrong because of injury.A good example is looking at Steven Jackson projections last year. Most were projecting Jackson for about 280 carries, which is about what he averaged over the last 3 years. What people should have been looking at were his average carries per game, which were about 21. Over 16 games, this would give him around 330 carries. Not surprisingly, he finished averaging over 21 carries per game. The latter projections are much more useful when sorting players.

Projections are fun. Easy way to pass time during the looooong offseason. Doesn't matter the scoring system.

I'm not questioning a particular scoring system. Just saying the only thing that matters is PPG (NOT PPR), not points per season.

 

[Cassius]

they had a Pythagorean record of 11.4 and 4.6.

starting to think you're all shtick, like the Dodds of the SP.

 

[Dinsy Ejotuz]

Now you might say "who cares what a team's Pythagorean record was?" And I'd say "well, a team's Pythagorean record is more closely correlated with a team's actual record than the next season than a team's actual record in the initial season. So, if you want to project how many wins a team will have next year, you're much better off if you look at Pythagorean record."

Chase states a fact, and you think it's schtick?

 

[Joe Bryant]

they had a Pythagorean record of 11.4 and 4.6.

starting to think you're all shtick, like the Dodds of the SP.

Trust me he's not. We're trying to keep the discussion respectful so please do that. Be it staffers or anyone else.J

 

[Mark Wimer]

they had a Pythagorean record of 11.4 and 4.6.

starting to think you're all shtick, like the Dodds of the SP.

Trust me he's not. We're trying to keep the discussion respectful so please do that. Be it staffers or anyone else.

J

Chase has schooled me in statistical analysis since he was a teen-ager. And, before someone gets snarky about the preceding, I'm not ashamed to admit that. Chase (or David Dodds/Jason Wood/Maurile Tremblay/John Norton/Bob Henry/Doug Drinen/Bruce Henderson and many others on staff, for that matter) is/are the best in this business. We as a community all benefit from the expertise/perspective that each member of the board brings to these topics, IMHO. I know that I have grown in many ways thanks to the collective wisdom of the other staffers and posters here at Footballguys.com (poster/member list is far too vast to name - many of the posters I've competed against in our Ultimate Survivor leagues would lay waste to 99% of the 12-team local leagues played here in the USA, IMO).

MW

 

[Cassius]

they had a Pythagorean record of 11.4 and 4.6.

starting to think you're all shtick, like the Dodds of the SP.

Trust me he's not. We're trying to keep the discussion respectful so please do that. Be it staffers or anyone else.J

Well at least we've got confirmation about Dodds.

 

[THE UNDERCOVER BROTHA]

Good read but i think after the draft and more cutdowns will throw alot of projections off.

But we dynasty freaks can appreciate these early projections.

You are higher than me on LT MOP, i know he on the Jets but he looked terrible this past season.

 

[Michael Fox Fan]

Prior results have no bearing on future results . . .Teams that ranked #1 in points and yardage allowed and how they fared the following year:2008 Steelers --> 2009 12 and 52006 Ravens --> 2007 22 and 62004 Steelers --> 2005 3 and 52002 Buccaneers --> 2003 4 and 51996 Packers --> 1997 5 and 71986 Bears --> 1987 4 and 21985 Bears --> 1986 1 and 11981 Eagles --> 1982 19 and 201979 Buccaneers --> 1980 20 and 201976 Steelers --> 1977 17 and 71972 Dolphins --> 1973 1 and 31970 Vikings --> 1971 1 and 21969 Chiefs --> 1970 8 and 6

I wouldn't exactly say that is the same as "prior results have no bearing on future results". 117/13 = 989/13 = 6.8Teams that finished 1/1 in pts and yardage followed that up by on average finishing (in year N+1):9th in points7th in yardageThat is significantly above average. It's not like these teams are immediately reverting to the mean.It's more accurate to say that finishing 1st in points and yardage doesn't guarantee a team will finish top-5 in both categories the following year, but they are quite likely to finish above average.

 

[Michael Fox Fan]

Prior results have no bearing on future results . . .Teams that ranked #1 in points and yardage allowed and how they fared the following year:2008 Steelers --> 2009 12 and 52006 Ravens --> 2007 22 and 62004 Steelers --> 2005 3 and 52002 Buccaneers --> 2003 4 and 51996 Packers --> 1997 5 and 71986 Bears --> 1987 4 and 21985 Bears --> 1986 1 and 11981 Eagles --> 1982 19 and 201979 Buccaneers --> 1980 20 and 201976 Steelers --> 1977 17 and 71972 Dolphins --> 1973 1 and 31970 Vikings --> 1971 1 and 21969 Chiefs --> 1970 8 and 6

Instead of just using the #1 teams, use all the teams and run correlations. It starts to become meaningful.

I've done this and posted it before in the SP. The basic takeaway was that defenses revert to the mean, but top-tier defenses are likely to remain above average and terrible defenses are likely to remain below average. Pretty sure I looked at it from the perspective of fantasy points allowed, rather than NFL points or yards allowed.

 

[Pictus Cat]

The anti-projection stance gives me a chuckle.

If you have rankings, you are making a projection.

If you tier them, you are getting closer to a projection.

If you adjust for scoring, you are making a projection.

Think of it as a list of 12 players you can slide up and down a board.

You have three at the top.

space

You tier two a little below.

space

space

space (because you have a big tier dropoff here)

5 players here

space

2 players

Congratulations. What you percieve in space can be assigned mathmatical numbers which can be turned into projections.

Move a couple of players up because they catch more balls as possession recievers, you've projected receptions without knowing it.

I think a combination of mathmatical minds and talent evaluations is a great way to go.

If you love a player and think he will be the best this season by far, wouldn't it be nice to have a mathmatical and historical referrence to find out you are valueing that player 30% more than has historically ever happened.

Now throw in VBD...

 

[Pictus Cat]

...and furthermore, why can't projections be entered for all staffers into the Dominators just based of rankings. A general formula could place players into projections based off of difference of ranking?

A little review and tweek afterwards and it should be pretty close to each staffers thinking.

Let the computer do the math for the space staffers give in they're tiers.

 

[BuckeyeArt]

Prior results have no bearing on future results . . .Teams that ranked #1 in points and yardage allowed and how they fared the following year:2008 Steelers --> 2009 12 and 52006 Ravens --> 2007 22 and 62004 Steelers --> 2005 3 and 52002 Buccaneers --> 2003 4 and 51996 Packers --> 1997 5 and 71986 Bears --> 1987 4 and 21985 Bears --> 1986 1 and 11981 Eagles --> 1982 19 and 201979 Buccaneers --> 1980 20 and 201976 Steelers --> 1977 17 and 71972 Dolphins --> 1973 1 and 31970 Vikings --> 1971 1 and 21969 Chiefs --> 1970 8 and 6

Instead of just using the #1 teams, use all the teams and run correlations. It starts to become meaningful.

I've done this and posted it before in the SP. The basic takeaway was that defenses revert to the mean, but top-tier defenses are likely to remain above average and terrible defenses are likely to remain below average. Pretty sure I looked at it from the perspective of fantasy points allowed, rather than NFL points or yards allowed.

I looked at it using yards per carry. It was pretty consistent.

 

[H.K.]

LaDainian Tomlinson-180-200/700/10TD, and 40/300/2TD

I don't see LT2 catching 40 passes next year for several reasons:1) The Jets' RBs had 25 receptions combined last year (28 if you count FB Richardson's 3 receps at FB). That would be quite bump for one RB to get 40.

2) Washington will eat into the reception total:

Jets | Ryan wants L. Washington to go back to old role Tue Mar 23, 05:58 PM

Bert Hubbach, of The New York Post, reports the New York Jets head coach Rex Ryan said the team should put RB Leon Washington in his old role as a third-down back and returner. 'What we should do ... is put [Washington] back to his role, which he had initially, which is third-down back, a change-of-pace type back and a Pro Bowl returner,' Ryan said. 'And there's nothing wrong with that.'

3) Expanding on point #2, LT2 averaged 7.7 YPR on 20 catches last year, versus Sproles 11.0 YPR on 45 catches. LT2 is not what he once was as a player, and he gave up his receiving role in SD to Sproles, who clearly outperformed him. If Washington is healthy, one should expect him to be the Jet RB used in the passing game.Good thread, btw, like the concept of a team breakdown for justifying projections.

 

[MAC_32]

The anti-projection stance gives me a chuckle.

I would rather spend the hours upon hours it'd take to create projections researching, I think it's a more effective and efficient use of time.

 

[Ministry of Pain]

LaDainian Tomlinson-180-200/700/10TD, and 40/300/2TD

I don't see LT2 catching 40 passes next year for several reasons:1) The Jets' RBs had 25 receptions combined last year (28 if you count FB Richardson's 3 receps at FB). That would be quite bump for one RB to get 40.

2) Washington will eat into the reception total:

Jets | Ryan wants L. Washington to go back to old role Tue Mar 23, 05:58 PM

Bert Hubbach, of The New York Post, reports the New York Jets head coach Rex Ryan said the team should put RB Leon Washington in his old role as a third-down back and returner. 'What we should do ... is put [Washington] back to his role, which he had initially, which is third-down back, a change-of-pace type back and a Pro Bowl returner,' Ryan said. 'And there's nothing wrong with that.'

3) Expanding on point #2, LT2 averaged 7.7 YPR on 20 catches last year, versus Sproles 11.0 YPR on 45 catches. LT2 is not what he once was as a player, and he gave up his receiving role in SD to Sproles, who clearly outperformed him. If Washington is healthy, one should expect him to be the Jet RB used in the passing game.Good thread, btw, like the concept of a team breakdown for justifying projections.

Nice post HK, nice post

 

[Pictus Cat]

The anti-projection stance gives me a chuckle.

I would rather spend the hours upon hours it'd take to create projections researching, I think it's a more effective and efficient use of time.

I agree, but you don't have to start from scratch with FBG. They do projections, which is ranking with details. The difference in thier projections for your point system can be translated into tiers.You can compare them to your own tiers.You can assume projections for your tiers.You can ask why do your tiers differ and how do you think the projections will work out to make the player you like be the better scorer.You can compare them to historical statistics to realize if the player you love is being ranked by you to be the best of all time or how they fall into the teams breakdown for your expectations.It is quantifying your expectations due to your research.

 

[Imagine]

Chase answered pretty close to what I would have said. But as far as fantasy goes, either you have an elite scoring QB on your roster or you are losing points from the QB position. I don't think Sanchez will make a huge leap to be worth a starting spot in pretty much any 1 QB fantasy league. He might be a below average QB2 . . . but as Chase said what does that get you.

As for the Jets and/or the Ravens, BOTH teams have been a bit of a shotgun blast over the past few years. The Jets have seen their record jump up by up to 6 wins in a season . . . or fall back by 6 wins in a season. THe Ravens have improved by up to 7 wins in a season . . . and have dropped off by up to 8 wins in a season.

As for Chase's assertion that the Jets were better than their record indicated and that they "should" have been 11.4-5.6, the same would hold for New England, who "should" have been 11.6-5.4. The Pats were one play/stop away from being 15-1 with late game losses to NYJ, DEN, IND, MIA, and HOU.

Long story short, an NFL season can turn on a handful of plays, and a number of teams in any given season could have had 8-12 wins if things worked out.

The reason why I think the Jets will struggle some is that I am not a big Sanchez believer, but certainly if he plays well the Jets could have a very good team.

Hi David,I wanted to follow-up on a comment you made that I think is one of the key decisions to a successful non-PPR redraft season. “When do you take your QB?”

Your statement that “either you have an elite scoring QB on your roster or you are losing points from the QB position” is so key.

My QB draft philosophy has been that I can wait, because there are QBs in the 5th or 6th rounds that will carry you well enough to allow you to draft RBs and WRs in those first 4 or 5 rounds. Last year I waited a bit too long when I was targeting Schaub in the 6th, and he went right before me. That left me with guys like Garrard and Hasselbeck where you needed to guess right regarding who to start each week…if you guessed wrong, it was usually a very long weekend.

So having a weak QB last year, combined with the huge numbers that those top QBs put up, make me think two things.

• One is that other owners may be drafting elite QBs even earlier than they typically do, which makes waiting even more “dangerous”.

• Two is that since you can be losing HUGE points each week from the QB position, I wonder if taking a QB early might be something to consider this year.

So finally a three part philosophical question.

1. Have Rodgers, Brees, and Manning so separated themselves from the pack that they are worth an early pick, probably in the 2nd or 3rd round?

2. If you don’t get one of those top three early, are the players in the next tier, probably Brady, Schaub, Rivers, Romo ones you feel you must get in the 3rd or 4th round?

3. And finally, if you sit tight, and assume those 7 are gone by the 5th, what do you see as a reasonable draft plan to keep from totally getting mashed at the QB position each week?

As I look at this question, it’s pretty far removed from the point of this thread, so if I’m way off base, let me know and I’ll find another way to ask you the question.

Thanks!

 

[phthalatemagic]

Chase has schooled me in statistical analysis since he was a teen-ager. And, before someone gets snarky about the preceding, I'm not ashamed to admit that. Chase (or David Dodds/Jason Wood/Maurile Tremblay/John Norton/Bob Henry/Doug Drinen/Bruce Henderson and many others on staff, for that matter) is/are the best in this business.

He has? I want to read the schooling. Chase is a guy with an opinion, not an expert. But if he is, I'd like to read about his theories in an abstract, unapplied to FF sense to get a better idea of this expertise.

 

[David Yudkin]

Chase answered pretty close to what I would have said. But as far as fantasy goes, either you have an elite scoring QB on your roster or you are losing points from the QB position. I don't think Sanchez will make a huge leap to be worth a starting spot in pretty much any 1 QB fantasy league. He might be a below average QB2 . . . but as Chase said what does that get you.

As for the Jets and/or the Ravens, BOTH teams have been a bit of a shotgun blast over the past few years. The Jets have seen their record jump up by up to 6 wins in a season . . . or fall back by 6 wins in a season. THe Ravens have improved by up to 7 wins in a season . . . and have dropped off by up to 8 wins in a season.

As for Chase's assertion that the Jets were better than their record indicated and that they "should" have been 11.4-5.6, the same would hold for New England, who "should" have been 11.6-5.4. The Pats were one play/stop away from being 15-1 with late game losses to NYJ, DEN, IND, MIA, and HOU.

Long story short, an NFL season can turn on a handful of plays, and a number of teams in any given season could have had 8-12 wins if things worked out.

The reason why I think the Jets will struggle some is that I am not a big Sanchez believer, but certainly if he plays well the Jets could have a very good team.

Hi David,I wanted to follow-up on a comment you made that I think is one of the key decisions to a successful non-PPR redraft season. “When do you take your QB?”

Your statement that “either you have an elite scoring QB on your roster or you are losing points from the QB position” is so key.

My QB draft philosophy has been that I can wait, because there are QBs in the 5th or 6th rounds that will carry you well enough to allow you to draft RBs and WRs in those first 4 or 5 rounds. Last year I waited a bit too long when I was targeting Schaub in the 6th, and he went right before me. That left me with guys like Garrard and Hasselbeck where you needed to guess right regarding who to start each week…if you guessed wrong, it was usually a very long weekend.

So having a weak QB last year, combined with the huge numbers that those top QBs put up, make me think two things.

• One is that other owners may be drafting elite QBs even earlier than they typically do, which makes waiting even more “dangerous”.

• Two is that since you can be losing HUGE points each week from the QB position, I wonder if taking a QB early might be something to consider this year.

So finally a three part philosophical question.

1. Have Rodgers, Brees, and Manning so separated themselves from the pack that they are worth an early pick, probably in the 2nd or 3rd round?

2. If you don’t get one of those top three early, are the players in the next tier, probably Brady, Schaub, Rivers, Romo ones you feel you must get in the 3rd or 4th round?

3. And finally, if you sit tight, and assume those 7 are gone by the 5th, what do you see as a reasonable draft plan to keep from totally getting mashed at the QB position each week?

As I look at this question, it’s pretty far removed from the point of this thread, so if I’m way off base, let me know and I’ll find another way to ask you the question.

Thanks!

I agree that this probably should be in its own thread, but that being said . . .Part of the issue involving when to draft a QB is that even though QB scoring has gone up over the last few years, the overall VBD of QBs hasn't changed all that much. But as we indicated, that's all well and good as long as you have one of the guys worth starting, but ifg you don't you could be in big trouble.

Last year in particular, there were weeks where the rest of my weekly roster did very well but I ended up getting killed by a QB that went nuts that particular week.

Over the past two seasons, here are the PPG averages for the Top 25 QBs:

Drew Brees 24.9

Aaron Rodgers 24.3

Philip Rivers 22.0

Tony Romo 21.8

Matt Schaub 21.7

Peyton Manning 21.1

Tom Brady 20.5

Donovan McNabb 20.3

Jay Cutler 19.9

Brett Favre 19.1

Ben Roethlisberger 18.9

Eli Manning 17.8

David Garrard 17.5

Matt Cassel 16.8

Shaun Hill 16.7

Kyle Orton 16.6

Jason Campbell 16.5

Matt Ryan 16.2

Matthew Stafford 16.1

Tyler Thigpen 15.8

Chad Pennington 15.7

Alex Smith 15.7

Joe Flacco 15.5

Carson Palmer 14.7

Matt Hasselbeck 13.7

I would put Rodgers and Brees in the first tier and the next 5 guys in the second tier. If McNabb stays in PHI (doubtful at this point), I would group him with Cutler, Favre, and Roethlisberger in the third tier.

On paper, that almost leaves everyone else as fighting to have any real startable fantasy value. As I mentioned above, if you play a guy that ends up 15 or 16 ppg, you could be on average losing out on 8-9 ppg compared to the elite QBs.

I'm not sure that's something easily made up each week with a slightly better stable at RB and WR.

This year, I am looking to get the last guy of the front 7. If I don't fall into someone there, then I REALLY would want one of the next 4. After that, I'd be looking to stock up on several potential breakout guys (which for me I haven't figured out who those might be yet), but as you said that could be playing with fire.

So to summarize, with the top third of NFL starters scoring more, that doesn't make the STARTABLE fantasy QBs worth more . . . but IMO the rest of the league is now too far behind to automatically just wait on drafting QBs like was popular for many years. That was a decent strategy when it only cost you a couple of points each week, but that's probably no longer the case . . .

 

[Mark Wimer]

Chase has schooled me in statistical analysis since he was a teen-ager. And, before someone gets snarky about the preceding, I'm not ashamed to admit that. Chase (or David Dodds/Jason Wood/Maurile Tremblay/John Norton/Bob Henry/Doug Drinen/Bruce Henderson and many others on staff, for that matter) is/are the best in this business.

He has? I want to read the schooling. Chase is a guy with an opinion, not an expert. But if he is, I'd like to read about his theories in an abstract, unapplied to FF sense to get a better idea of this expertise.

Go Here: PFR Blog Look for entries by Chase. Read. Learn. MW

 

[Chase Stuart]

I would think it's no longer controversial that last year's record is not a very good predictor of next year's record. Wins are not very granular; points and points allowed (while not perfect) offer a much better look. You can show this in a ton of ways, but I'll just do a simple study.

I looked at all team seasons from '99 to '08, and recorded each team's winning percentage and it's Pythagorean winning percentage. The correlation coefficient between winning percentage in Year N and winning percentage in Year N+1 was 0.29; the CC between Pythagorean winning percentage in Year N and winning percentage in Year N+1 was 0.31. This means, unsurprisingly, that Pythagorean winning percentage is more closely correlated with future winning percentage than just plain old regular winning percentage. Again, unsurprising, given the paucity of information a win or a loss gives you.

The difference between the CCs is pretty small -- 0.29 and 0.31. But that's just because the two numbers themselves are highly correlated; the CC between winning percentage and Pythagorean winning percentage in the same season is 0.91; when two numbers are so highly correlated, both will appear to be similarly correlated with other things. That's just simple math.

There's an easy way around that, though. Take a look at the teams with the largest differentials in Pythagorean wins and actual wins. There were 31 teams like the 2008 Green Bay Packers who won at least 1.5 fewer games than they "should" have won based on Pythagorean record. They averaged 5.18 wins in Year N, but had 7.42 Pythagorean wins, i.e., they underperformed. In year N+1, they won an average of 7.84 games. So, quite clearly, looking at their Pythagorean record in Year N would have been much more helpful in figuring out how many wins they'd have the next year than looking at just their actual win totals. This is, what I think, pretty noncontroversial to most people.

On the other side, there were 38 teams who won 1.5 more games than they "should" have won; on average, they had 10.79 win in Year N, but just 8.77 Pythagorean wins. The next season, those 38 teams won an average of 8.71 wins, almost exactly matching their Pythagorean number.

Pythagorean winning percentage and regular winning percentage are usually very similar; teams tend to win the number of games they should, or in Parcells-lingo, you are what your record says you are. That's why the CCs between Win% and Future win % and PythWin% and Future win % were similar; Win % and PythWin% are highly correlated. But when they differ -- which is when we care most about Pythagorean winning percentage -- the numbers overwhelmingly show why you should use Pythagorean winning percentage. Of course, you should use something even more granular than points and points scored data if you really want to predict the future, but it's obviously better than just using last year's record.

 

[Mark Wimer]

Chase has schooled me in statistical analysis since he was a teen-ager. And, before someone gets snarky about the preceding, I'm not ashamed to admit that. Chase (or David Dodds/Jason Wood/Maurile Tremblay/John Norton/Bob Henry/Doug Drinen/Bruce Henderson and many others on staff, for that matter) is/are the best in this business.

He has? I want to read the schooling. Chase is a guy with an opinion, not an expert. But if he is, I'd like to read about his theories in an abstract, unapplied to FF sense to get a better idea of this expertise.

Go Here: PFR Blog Look for entries by Chase. Read. Learn. MW

FYI, here's a recent post on Chris Johnson with some of Chase's methodology laid out for you.BTW, I'm a Footballguy. I don't care about "theories in an abstract, unapplied to FF sense". I care about expert fantasy football analysis.

MW