Mr. Norton's Rankings vs. Projections (1 Viewer)

Not trying to be anal since a big fan of Messrs. Norton and Bramel but having a hard time reconciling these.

Just taking the LB's for example, why wouldn't the rankings would fall out directly from the projections? I could see differences in scoring systems perhaps being the cause but it would seem everything should tie to FBG's default scoring and people could customize from there.

The discrepancies are most egregious where the potential for value picks is greatest. For example:

A. Ogletree - Rank = 11, Proj. = 22

B. Spikes - Rank = 23, Proj. = 15

B. Cushing - Rank = 32, Proj. =18

For guys like Burfict and Kuechly, it doesn't really matter if they are swapped (2) vs. (3) since they are both studs and it's a matter of personal preferences.

But the natural process is to read the exceptional EOTG commentary and get excited about a few non-obvious players. The next step is to re-value the player in your own rankings/projections using FBG's rankings/projections (especially Mr. Norton's) as a guide. Then cross-check with Bramel's tiers to learn even more.

The rankings are awesome because of the individual notes explaining differences with consensus. But the projections are crucial to translating to your own league's parameters and also determining tier break points.

So when there are large discrepancies it's difficult to have confidence taking a risk on the value plays.

If it's just timing then maybe they could be updated simultaneously.

TIA.

Bump.

If you look at the current Dodds/Bryant rankings and projections for offense they may vary slightly but for example Sammy Watkins is ranked WR36 and his projected stats have him WR36. There are a few guys who differ by one or two slots but for the most part they line up.

I suppose risk could play a factor but unless there is some type of rationale for the wide variance of certain IDP players it is difficult to find either the rankings or projections useful once you get past the top ten LB's.

it seems like guys have different philosophies of projections. Some will calculate upside into their projections, creating a sort of "mean outcome" projection, meaning that if they think there is a small probability of a rookie totally going off in their first year, they will cook that into the projection. Other projections are more of a "Most likely scenario" projection, where upside is not cooked in or represented in any way. If the "most likely scenario" for Ogletree isn't very good, but there is a lot of upside to be had with him, I can see why he would be ranked way ahead of his projection. The different might simply be that Dodds cooks in his upside and Norton doesn't.

But I'm with you, it would be nice to have all of these philosophies stated clearly for the sake of interpreting the results.

Rankings and projections are two totally different animals to me. I agree that they should look similar and believe that mine do, but they are not the same thing in my eyes. In the projections I am looking at strictly overall numbers. There is no way to relate consideration of things like week to week consistency or how much confidence I may have in the numbers that I have projected, injury risks and so forth.

I guess the simplest way to say it is that my approach to projections is that they are based on overall statistical expectations with some consideration for potential, while my rankings will take all aspects into consideration and more closely follow how I would actually draft. If they were exactly the same thing there would be no need for me to even do both.

Hope that gives a little better perspective and thanks for the feed back.

TheMathNinja said:

it seems like guys have different philosophies of projections. Some will calculate upside into their projections, creating a sort of "mean outcome" projection, meaning that if they think there is a small probability of a rookie totally going off in their first year, they will cook that into the projection. Other projections are more of a "Most likely scenario" projection, where upside is not cooked in or represented in any way. If the "most likely scenario" for Ogletree isn't very good, but there is a lot of upside to be had with him, I can see why he would be ranked way ahead of his projection. The different might simply be that Dodds cooks in his upside and Norton doesn't.

But I'm with you, it would be nice to have all of these philosophies stated clearly for the sake of interpreting the results.

Click to expand...

There are most certainly different approaches. I for example, will have the sleeper players that I am most high on ranked higher than they are projected. This has a lot to do with my personal drafting philosophy. When drafting I will put a higher value on the guy with upside than a player I may have projected higher but is less likely to out perform those projections or more likely to fall short of them. This is especially true when it comes to those players outside the top 15-20 at their positions, which seems to be the players being called into question. So again, I look at projections as a base line or starting point while my rankings are more like my personal draft list.

I'm not sure how Dodds approaches this but I would have to believe that he lays out his projections and then does his ranking based off of that ranking order. To be completely honest I have never been a big fan of projections as they don't allow for so many variables. The best way to approach using my contributions would probably be to read my column and compare it to my rankings, using my projections as a way to remove my personal drafting strategy and all the variables that could come into play.

I will add that I am due for an overhaul of my rankings now that the first round of preseason games is in the books. The currently posted projections are about 5-6 days newer than my last rankings. Both will be updated in the next 48 hours, just as soon as I finish scouting the late Saturday games.

Awesome.

Completely agree rankings and projections do not need to / should not line up exactly. The key is to understand why and with the explanation on your projections philosophy, plus by reading the EOTG columns and notes to the rankings that can be inferred without too much trouble.

A prime example is Cushing who is ranked significantly below his projection I would presume due to injury risk. Personally, I like to rank Laurinaitis higher than most have him projected for due to consistency.

Thanks for making thoughtful projections a key part of the FBG content. IMO they by no means are taken literally but are essential to understanding what variables are driving a guy's expected performance (big plays, solos, assists) so you can assess upside/downside risk.

In addition, they point out the break points between tiers. As we know rankings then become much less important as long as you have the tiers nailed down.

But as MathNinja points out there can be potentially some confusion if there are large discrepancies unless at least some insight is given on philosophy.

Appreciate the transparency.

John,

Could you help me with one question in particular here? I think it will help me understand your philosophy on this. I'm in the process of making my Defensive Tackle auction sheet, and I'm noticing a very large discrepancy in your rating of Kyle Williams.

In your projections, he is the top DT in standard FBG IDP scoring. In your rankings, he doesn't crack the top 12 DT's. How are you viewing Kyle Williams in your drafts? What is your mindset in making decisions on where to draft him or how much to pay in an auction? I have a similar question with regards to Jason Pierre Paul (7th in projections, 15th in rankings), but I think I have a clearer sense of what your answer would be there, and it's not the outlier Williams is.