Rankings vs Cheatsheets (1 Viewer)

Random Footballguys staff discussion today. Do people think there's a difference between "Cheatsheets" and "Rankings"?

And if they're different, how so?

You need a choice of "depends".

If you are talking about just straight lists of players with no other information then they are the same.  If you add narrative to rankings then they become different.

A cheatsheet is just a list of players.  By definition I see a cheataheet as a quick reference to be used as a reminder.  Cliff notes for your rankings 

Rankings (if done right) will elaborate on the players and give insight as to why players are ranked where they are ranked.

Another vote for “depends”. 
 

at FBG at least, a cheatsheet is a flat list that is driven by projections. Whereas the FBG rankings allow me to see multiple people’s rankings, filter some out if I want, sort by a single person if I want, and sort by median or adjusted average, etc. 

So, by the above, I’d say they’re pretty different but not exactly sure if that’s what you’re really asking. 

Joe Bryant said:

Random Footballguys staff discussion today. Do people think there's a difference between "Cheatsheets" and "Rankings"?

And if they're different, how so?

Click to expand...

Firmly in the "No Difference" camp. 

Unless they are backed up by full team projections, neither one is all that valuable to me. For instance I've seen plenty of places where MIN has a top10 WR with high upside, a top 15 WR with a very high floor, and a top 15 "sleeper" TE with high upside to outperform his ranking by a fairly wide margin. But somehow I almost never see Cousins ranked even in the top half of starting QB rankings. If more people in the industry actually put in the time to show their work as to how that happens it would be easier for me to take their rankings more seriously.... even if I disagreed with them. I'm not saying popular websites don't have full team projections rather I'm saying if there are staff members on those sites that are pushing out rankings without projections to back them up then why even bother? If they aren't willing to put in the work than just make a top-3 underated/overated article for each position and be done with it.

Joe Bryant said:

Random Footballguys staff discussion today. Do people think there's a difference between "Cheatsheets" and "Rankings"?

And if they're different, how so?

Click to expand...

I voted “much different”.

To me. It goes like this:

Rankings are what individual industry pros & individual FF managers create. It’s a comprehensive list of players at each position (& maybe overall) going however many deep I feel is appropriate for the league & scoring system. I won’t exclude any player in these, nor will i add any detail other than where I think they’re ranked. 

I see a “cheat sheet” as a composite of rankings by whatever combination of those experts (and my own) rankings. Often a “cheat sheet” will have a column for ADP, with early & late, & I’ll usually add notes field or strike through players I don’t want. 

I see them both as having value, but definitely different tools. 

Completely different for me. 
 

Simplest explanation for me. I see rankings as a compilation of projections. The end result of those projections if you will. 
 

Cheatsheets are how I apply those rankings to my draft. Bump players up or down based on the team dynamics of my league, owner tendencies, bye weeks, SOS, etc.  

I'll play....

There's no difference because they are both inferior to Tiers.

SeniorVBDStudent said:

I'll play....

There's no difference because they are both inferior to Tiers.

Click to expand...

I incorporate tiers into my cheat sheets. 

checkmate. 

BoltBacker said:

Firmly in the "No Difference" camp. 

Unless they are backed up by full team projections, neither one is all that valuable to me. For instance I've seen plenty of places where MIN has a top10 WR with high upside, a top 15 WR with a very high floor, and a top 15 "sleeper" TE with high upside to outperform his ranking by a fairly wide margin. But somehow I almost never see Cousins ranked even in the top half of starting QB rankings. If more people in the industry actually put in the time to show their work as to how that happens it would be easier for me to take their rankings more seriously.... even if I disagreed with them. I'm not saying popular websites don't have full team projections rather I'm saying if there are staff members on those sites that are pushing out rankings without projections to back them up then why even bother? If they aren't willing to put in the work than just make a top-3 underated/overated article for each position and be done with it.

Click to expand...

I am struggling with a similar situation....FBG has 3 PIT WR's ranked in the top 28 PPR.....but Ben is like QB22....it doesn't seem to add up.....when Manning made 3 WR's serviceable/startable....he was easily in the top 10 or top 5.....seems like either the PIT WR's are being overvalued as a group or Ben is being seriously undervalued.....

Stinkin Ref said:

I am struggling with a similar situation....FBG has 3 PIT WR's ranked in the top 28 PPR.....but Ben is like QB22....it doesn't seem to add up.....when Manning made 3 WR's serviceable/startable....he was easily in the top 10 or top 5.....seems like either the PIT WR's are being overvalued as a group or Ben is being seriously undervalued.....

Click to expand...

Maybe Ben is not expected to play 16 games? 

Same difference.

Acrobat7 said:

Maybe Ben is not expected to play 16 games? 

Click to expand...

which is a whole nuther discussion its its own right.....I mean, people that predict say only 14-15 games or something....where exactly does that come from.....are they "predicting" an injury....or just basing it off what the guy "seems to average"....IDK....it seems kind of silly IMO to factor in injury in predictions.....I really don't want my "rankings experts" to somehow think they are "injury predictors".....IMO rankings should be based on playing every game in this case it is (17)....and then let me decide if I want to dial that back because of injury or something.....cause I am probably as "qualified" at predicting an injury as anybody else......TBH the only case might be if a ranker thinks a team will have locked up a bye and "sit" their guys in the last game (ala KC last year)....but even that is a reach, especially with only one bye now....

FBG does provide their projections by team broken down by player so you could see how the projections for say all PIT QBs was broken down and also the receiving projections for all of their players (RB/WR/TE), and also had a listing of their projections by position so one could see how that teams players stacked up against other players in that position.

In this case, and not to reveal too much of the paid content, the vast majority of Pittsburg passing yards are projected to go to the WRs, and it is spread very evenly. Ben is projected with about 90% of the Pit passing yards with Rudolph getting about 10%. The 3 PIT Wrs are all projected in the top 28, but what is not mentioned is that the highest rated one is at 24, the 3rd highest is at 28, as their projections are near identical.

Right or wrong, they do have the data out there in the open to look at to see how they are coming up with the rankings based on the projections. It's up to us to determine how much agreement/disagreement with the projections we have.

acarey50 said:

FBG does provide their projections by team broken down by player so you could see how the projections for say all PIT QBs was broken down and also the receiving projections for all of their players (RB/WR/TE), and also had a listing of their projections by position so one could see how that teams players stacked up against other players in that position.

In this case, and not to reveal too much of the paid content, the vast majority of Pittsburg passing yards are projected to go to the WRs, and it is spread very evenly. Ben is projected with about 90% of the Pit passing yards with Rudolph getting about 10%. The 3 PIT Wrs are all projected in the top 28, but what is not mentioned is that the highest rated one is at 24, the 3rd highest is at 28, as their projections are near identical.

Right or wrong, they do have the data out there in the open to look at to see how they are coming up with the rankings based on the projections. It's up to us to determine how much agreement/disagreement with the projections we have.

Click to expand...

Rudolph getting 10% of the passing yards basically predicts Ben to miss two games though.

Acrobat7 said:

Rudolph getting 10% of the passing yards basically predicts Ben to miss two games though.

Click to expand...

Is your assertion that injury history and likelihood to miss games should not be factored into projections/rankings? I personally think it should so long as it is identified, such as in this case. Roth can be flagged as an injury risk. Again, digging into paid content so I'll be a bit vague, but for Roth, of the 6 projections used to build up the consensus projection, 3 of the projectors show Roth playing less than 17 games this season, which given recent history with him, I believe is a fair expectation and a factor that needs to be considered.

Similarly, how would you then approach situations like say Chicago or the Niners where there is a high likelihood that while the veteran QB is likely to start the season, they are highly unlikely to finish the season with a highly drafted rookie likely to take over. Not easy to project for those scenarios, but it needs to be factored in.

Me personally, he's one of my top late round QB targets as I think there is a decent chance he outperforms the consensus projection this year, but he is a risky prospect, and thus even though I am one that tends to wait a long time on QB, I still would not want to rely on him being my QB1 at the start of the season - I think he is a great pick as a backup QB/streamer that is available at the end of drafts.

Stinkin Ref said:

which is a whole nuther discussion its its own right.....I mean, people that predict say only 14-15 games or something....where exactly does that come from.....are they "predicting" an injury....or just basing it off what the guy "seems to average"....IDK....it seems kind of silly IMO to factor in injury in predictions.....I really don't want my "rankings experts" to somehow think they are "injury predictors".....IMO rankings should be based on playing every game in this case it is (17)....and then let me decide if I want to dial that back because of injury or something.....cause I am probably as "qualified" at predicting an injury as anybody else......TBH the only case might be if a ranker thinks a team will have locked up a bye and "sit" their guys in the last game (ala KC last year)....but even that is a reach, especially with only one bye now....

Click to expand...

Mike Clay does some very good projections in my opinion and he pretty consistently has players missing a game or two in his numbers.

While I am not sure on the reasoning for this, you would have to ask Mike, I am guessing it is based off of an average number of games missed by players in a season. Maybe he is using specific data for each player in determining that as well. I don't know.

I feel similarly about predicting injuries as you do. At the same time I see some consistency in Mike Clays method with this, I mean he has almost every FF relevant player missing a game or two. So while his seasonal projections might look a lot lower than mine do because of the missed games, I have some confidence that he is applying the same method he is using to all of the players and I can deal with that.

FTR - FBGs has the backup getting 7.8% of the Steelers passing yards. The PA for Rudolph (49) suggest the consensus projections think Ben will miss one game + a couple of moo uo situations. 1/17 = 5.8%. Reasonable.

WRT “predicting injuries” I would submit no one is doing that. There’s no WR with say 11GP bc the dude has hammy issues every single year including college (hi Deebo.) But projecting 17GP universally isn’t at all realistic.

2020 GP:

QB - 11 of the top 15 QBs played every game. For this position, I generally want 17 GP projections bc that’s what happens IRL. Coaches don’t give QBs a week off bc they’re beat or dealing with a tweak, too much depends on continuity with the guy who is responsible for checking out of a play based on the Mike, front alignment or coverage. We shouldn’t project Tom Brady will have 15GP bc he never misses a game. Short of an ACL or suspension it’s never happened. OTOH, Big Ben has played every game 4 times in 17 years. Which is more accurate/likely? That he’ll play 17 or 15-16? That’s an obvious exception. There’s at least 3 QB situations where it seems reasonable to predict a 11/6 or 10/7 GS split: CHI, NE, SF. That’s not injury related, but we *think* the veteran will be supplanted at some point by the 1st Round rookie QB.

RB - of the top 30 RB in PPR, exactly six (6!) guys played 16 games. King Henry and five pass catchers (Hunt, McKissic, Hines, Edmonds, Bernard.) You could nitpick two of those but the point remains: guys who carry the load for their team are gonna miss 1-2 GS, year in and year out. In the previous two seasons (2018-19) the majority of the top 30 missed games.

WR - 18/30 played 16G

TE - 6/12 played every game.

If we were seeing projections with 3, 4, 5 games missed, that’s a serious error. I’m not seeing that anywhere. 1 game missed (maybe 2 for a workhorse RB) - perfectly reasonable, it would be less accurate to project 100% health across the board.

IMO

ok....soooo....I guess a solution would be to list the average number of "predicted games played" listed with the rankings by each player so we wouldn't have to go back and cross check with the projections (which is where this info usually is)....using Ben as the example, if the consensus thinks he is going to miss 2 games and that is why in the rankings he is say QB24 instead of QB16....I might want to know that quickly when looking at the rankings instead of having to go back back and look at projections....I have to admit, I am not a huge projections guy especially when it comes to games played so that is why the Ben example kind of stood out to me when 3 of his WR's are in the top 28 and he was ranked in the low 20's himself...

I realize this is kind of overkill and nitpicking.....but thought it was relevant to the thread about rankings vs. cheatsheets and how projections factor in and are part of that...

it just feels to me that when you start factoring in "predicting players to miss games because of injury", things get wonky and kind of sideways.....there should be the same baseline (17 games) for everybody and then leave it up to us to determine if we should factor in a player missing games because of injury.....

being replaced is maybe a little bit different story....but still I would want FBG to give me Dalton's ranking based on 17 games.....I will already be factoring in on my own the odds of him being replaced by week 10 or whatever.....but who knows....maybe it doesn't happen.....maybe Fields breaks a leg in preseason or whatever...saying you will "adjust Dalton's rankings" when/if that happens doesn't really help me with drafts that are happening now or have already happened...

Stinkin Ref said:

I realize this is kind of overkill and nitpicking.....but thought it was relevant to the thread about rankings vs. cheatsheets and how projections factor in and are part of that...

Click to expand...

Not necessarily overkill/nitpick, but, and I'm sure I'm preaching to the choir here and stating the obvious, but they all feed into each other. Here is how I view it:

Step 1 - Projections - Whether you do your own or use a trusted source (whether that is FBG or another), you first need a set of projections. This is the barebones raw data that is the foundation of what is to come.

Step 2 - Rankings - Now you must apply your individual leagues scoring system to those projections, thus converting the projection into a fantasy point prediction. Then simply sort each position by the projected points. This is also where I'd work up a "value" for each player which would be largely based on lineup rules, and there are many methods for doing this that would be an entirely separate discussion. Either way, we now have a value column.

Step 3 - Cheatsheet - For me this is simply a nice layout of my rankings, broken out by position. Ideally I like to include columns for player name, team/bye, projected points, value, ADP data if available and, if I have room for it a small note field or at least symbols to note players that I think are injury risks, have higher upside or have other risks (ie a 1st round QB ready to take over). This is where I also start marking off tiers in each position. Again, how that is derived can be a whole other discussion, but I usually group by similar values within a position, say within 1 or 1.5 PPG of each other, as in my opinion, no projections are that pinpoint accurate that if I have guys projected that closely together, and all else being equal, then I really view them as equal. For me, tiers is where the premade/Custom cheatsheets struggle the most, as most that I have seen are tiered by a preset number of players as opposed to by a grouping of value. By grouping by value, some tiers may only have 2-3 players, others may have 14-20 players.

Step 4 - Applying it - Personally, I'm not a big fan of overall top 50-60 or whatever rankings, but I see how they can be useful. When my pick comes up, I look at my top player at each position, then I try and make a best guess at which players at each position will be available at my next pick (or next two picks) to come to a best guess on which combination of those next 2-3 picks will give me the greatest overall value. Sometimes this means I will take a player that I have rated as a lower value over a higher value player at a different position, because I think that the likely value drop is less in that position and I will end up with the higher overall value for the team. More art than science at this point.

Acrobat7 said:

Maybe Ben is not expected to play 16 games? 

Click to expand...

That is a good thought, but if that's the case and PIT is rolling with Rudolph at QB I don't think many of those PIT receivers are going to hit their ADP. If I'm being completely honest I would admit @Stinkin Ref used a great example that I hadn't considered because I think most people are over projecting the entire PIT passing game for 2021. Maybe I'm underestimating how badly the losses in the secondary will hit them, but to me it's going to be all Defense and Harris.

One staff member(Henry) projects Ben for the most realistic number of pass attempts imo, and he has Ben playing the fewest games(15.0) as you allude to. The most unrealistic projection to my mind is Bloom who projects Ben to play 17.0 games but that may be a philosophical choice on his part. I realize there is an extra regular season game this year but Ben has only had as many attempts as Bloom is projecting once in his career, and what we are witnessing right now is not 2018 Ben. Not by a long shot. There used to be a guy that said teams really show their hand on draft day how they really feel about players/positions. PIT drafting Harris with the 24th pick in first round tells me all I need to know about how comfortable PIT feels letting their season ride on the arm of Ben again.

If you put me on the spot to choose ONE PIT receiver to meet expectation it would probably be one of the TE's given their "Low TE Rank" numbers. Obviously that's only if you play in very deep leagues.

acarey50 said:

Step 4 - Applying it - Personally, I'm not a big fan of overall top 50-60 or whatever rankings, but I see how they can be useful. When my pick comes up, I look at my top player at each position, then I try and make a best guess at which players at each position will be available at my next pick (or next two picks) to come to a best guess on which combination of those next 2-3 picks will give me the greatest overall value. Sometimes this means I will take a player that I have rated as a lower value over a higher value player at a different position, because I think that the likely value drop is less in that position and I will end up with the higher overall value for the team. More art than science at this point.

Click to expand...

Same here. Rankings and ADP are only useful to me as a tool to let me know how long I can wait on the guy I really want. It works better at the start of September than it does in July, though. Maybe it's because there are fewer data points and they disagree more in July. By September a lot of group think and popular narratives have set in across the hobby as a whole.

Stinkin Ref said:

it just feels to me that when you start factoring in "predicting players to miss games because of injury", things get wonky and kind of sideways.....there should be the same baseline (17 games) for everybody and then leave it up to us to determine if we should factor in a player missing games because of injury.....

being replaced is maybe a little bit different story....but still I would want FBG to give me Dalton's ranking based on 17 games.....I will already be factoring in on my own the odds of him being replaced by week 10 or whatever.....but who knows....maybe it doesn't happen.....maybe Fields breaks a leg in preseason or whatever...saying you will "adjust Dalton's rankings" when/if that happens doesn't really help me with drafts that are happening now or have already happened...

Click to expand...

I understand where you are coming from.

There is some evaluation of projections post season and so folks doing projections are tying to make their prognostication as accurate as possible, so projecting missed games may be a way to better achieve that goal. As BL pointed out in his post, players do miss a lot of games every year.