• Content count

  • Joined

  • Last visited

Community Reputation

491 Excellent


About ZWK

  • Rank

Profile Information

  • Gender
  • Location

Recent Profile Visitors

12,200 profile views
  1. Dayes had lousy stats at NC State and was one of the least athletic RBs at the combine. The only good thing I see in his statistical profile is that they threw him a fairly high number of passes. I'm paying attention to upside in my last tier, which means a lot of 1) guys who played better in previous seasons and might bounce back (McGuire, Conner, Hood, Breida), 2) guys who stood out on at least one aspect of athleticism (Breida, Carson, Joe Williams, Logan), and 3) guys with a strength who could wind up having fantasy value in the right role, like Theo Riddick or LeGarrette Blount (Mathers, Logan, Conner, Hood).
  2. A big part of the question with Fournette is what to do about players who have a big season whose last season in college is far from their best season. It could mean that their big season was something of a fluke, or it could mean that they were playing through nagging injuries or dealing with a struggling offense. My formula has a standard way of penalizing players for their worse production in their last season. But, in forming my overall personal opinion on when to draft players, I'm inclined to put some extra weight on the opinions of experts in situations like this, since they have extra information on what was going on. In this case, Fournette was dealing with an ankle sprain which seems like a pretty good excuse for worse production. If a player is rated highly by conventional wisdom but has never had a big season then I tend to be pretty skeptical of him, but Fournette's 2015 season matches his reputation so I'm more willing to go along with the experts. In 2015 he had good efficiency stats, huge totals, and great elusiveness numbers. He also has a good size-speed combo. Also, all of the other RBs have their own negatives; you can see that none of them reached the Gurley-Elliott tier in my ratings. Cook showed below average athleticism at the combine (with averageish size & 40 time and lousy jumps & agility drills), didn't have great rushing numbers in traffic, and has off-the-field concerns. Mixon has some concerns about his vision/decision-making, played behind Perine in 2015, and I think he benefited from getting the ball in space in Oklahoma's offense, along with his obvious off-the-field concerns. McCaffrey is undersized, didn't have great rushing numbers in traffic, and his rushing efficiency hasn't been at the same level as the other top RBs. Fournette had a mediocre 2016 season (including missed time), a lousy vertical, less involvement in the passing game, and weight fluctuations. I think it's a good year to trade down to pick 5 or so and take whoever is left out of the RBs and Corey Davis, although I'd wait till after the draft in case things change on draft day. Yes. They are both players whose last season was not their best - McGuire's 2016 production was bad and Breida's was atrocious. Breida's workout numbers plus his impressive 2015 production would have me excited about him as a prospect, but the awful 2016 season which came in between has tempered my enthusiasm. Still seems like a decent late-round flyer.
  3. Pre-draft RB update. Here's what my formula says: Eddie Lacy 2013 Todd Gurley 2015 Melvin Gordon 2015 Christine Michael 2013 Carlos Hyde 2014 Ezekiel Elliott 2016 Joe Mixon 2017 Lache Seastrunk 2014 Jay Ajayi 2015 Tre Mason 2014 Giovani Bernard 2013 Ameer Abdullah 2015 Knile Davis 2013 Derrick Henry 2016 Dalvin Cook 2017 Curtis Samuel 2017 Darius Jackson 2016 Jeremy Hill 2014 C.J. Prosise 2016 Christian McCaffrey 2017 Jerick McKinnon 2014 Kenneth Dixon 2016 Jonathan Franklin 2013 Tevin Coleman 2015 Jordan Howard 2016 Duke Johnson 2015 Alvin Kamara 2017 Samaje Perine 2017 Bishop Sankey 2014 DeAndre Washington 2016 David Johnson 2015 D’Onta Foreman 2017 Marcus Lattimore 2013 Jeremy McNichols 2017 Aaron Jones 2017 Jhurell Pressley 2016 Stephen Houston 2014 Daniel Lasco 2016 Henry Josey 2014 Zac Stacy 2013 D.J. Harper 2013 Le'Veon Bell 2013 Kareem Hunt 2017 Cierre Wood 2013 Kenjon Barner 2013 Paul Perkins 2016 Montee Ball 2013 Latavius Murray 2013 Isaiah Crowell 2014 Leonard Fournette 2017 Andre Williams 2014 Marlon Mack 2017 Charles Sims 2014 Joe Williams 2017 Dri Archer 2014 Elijah Hood 2017 Alex Collins 2016 James Conner 2017 Devonta Freeman 2014 David Cobb 2015 Devontae Booker 2016 Matt Breida 2017 T.J. Yeldon 2015 Cameron Artis-Payne 2015 Karlos Williams 2015 Josh Robinson 2015 I'Tavius Mathers 2017 Corey Grant 2015 Kenyan Drake 2016 David Fluellen 2014 Jamaal Williams 2017 Treavor Scales 2013 Jeremy Langford 2015 Elijah McGuire 2017 Michael Ford 2013 Robert Godhigh 2014 Matthew Tucker 2013 Mike Davis 2015 George Atkinson III 2014 Keith Marshall 2016 Terrance West 2014 Anthony Wales 2017 Teriyon Gipson 2017 Brian Hill 2017 Michael Dyer 2015 Wendell Smallwood 2016 Wayne Gallman 2017 Tim Cornett 2014 Jonathan Williams 2016 C.J. Anderson 2013 Andre Ellington 2013 James White 2014 Stanley Boom Williams 2017 De'Anthony Thomas 2014 Lorenzo Taliaferro 2014 Christopher Carson 2017 Tyler Ervin 2016 And here's the order that I'd draft them in: Leonard Fournette Dalvin Cook Joe Mixon Christian McCaffrey Alvin Kamara Curtis Samuel D'Onta Foreman Kareem Hunt Samaje Perine Jeremy McNichols Marlon Mack Aaron Jones Brian Hill Jamaal Williams Wayne Gallman Elijah McGuire James Conner Elijah Hood I'Tavius Mathers Matt Breida Christopher Carson Joe Williams TJ Logan The top 4 RBs and Corey Davis are my top 5 rookies. After that I generally like the WRs (or TEs) more than the RBs where they seem to be going in rookie drafts. I rate Kamara closer to the Foreman-Hunt tier than to the Mixon-McCaffrey tier.
  4. Football Outsiders recently posted their RB projections, using a stat-based projection system which they call BackCAST. They see this as an extremely strong RB class, with Leonard Fournette, Dalvin Cook, and Joe Mixon all rating among the 10 best RB prospects of the past 20 years. They also give relatively strong ratings to Brian Hill, Marlon Mack, Christian McCaffrey, Samaje Perine, and Jeremy McNichols. They're down on Alvin Kamara, though, mainly because of his low workload and mediocre 40 time (which is the only combine drill that they include in their formula). I've posted before about some of the strengths and weaknesses of FO's approach.
  5. nfldraftscout has Zamora's pro day numbers as 4.53 40, 34" vert, 10'0" broad. That tweet had him at 4.49 40, 40" vert, 11'1" broad. A 0.04 second disagreement on 40 times isn't unusual, but disagreements of 6" and 13" on the jumps is.
  6. He's short rather than thin - his BMI is actually the same as Mike Williams (26.7). I generally take BMI below 26.0 to be a warning sign for a WR, and anything above 26.5 to be not worth worrying about. I currently have Ross as the clear #7 pick, although he could get passed by Kamara or a TE depending how the draft goes.
  7. He has been a 1067/7.4 guy so far (per 16 games, for the 37 games that he has played).
  8. Never have 8 te, never have 6 qbs (assuming start 1). Most leagues roster around 24, use those roster spots for wr/rb flyers. Don't carry a kicker or defense in offseason unless you're required to, use those spots for skill players. Depending on who I have at qb I try to carry 3, I'm not rostering a guy simply because he has a job, although I'll sometimes pick up a guy off waivers if I think I can move one. There always seems to be a team that is thin at qb that will part with a 2nd/3rd for a decent prospect (like osweiler last year). That said I hopefully have 2 decent qbs and a 3rd prospect that can stream or be injury insurance. Its hard to find a difference maker at te, so don't roster a bunch of guys there hoping that eric swope suddenly becomes gronk. One way that I think of it: prospects have no position. If I'm trying to decide between TE Gerald Everett and WR Josh Malone, I'll think about how likely each of them is to wind up panning out, how much they're likely to score if they do pan out, how long it will probably take for me to find out if they'll pan out & how long I can afford to use a roster spot on them while I wait, and how much trade value they'll have at various points in time. But I won't be thinking about whether I'm stacked or thin at TE and WR, because 1) by the time that either of them pans out (if they ever do) my roster will probably look pretty different from how it looks now and 2) if I have to eventually make some trades to rebalance my roster I can do that. This could wind up in me having a 24-man roster that includes Everett, Hodges, Hooper, Higbee, and a few other TEs, and I'm fine with that. Apparently OP only has 15-man rosters over the offseason, which makes Everett and Malone both pretty close to worthless and means that he's very unlikely to have 8 TEs that are worth keeping over the offseason. The cutdown to 15 means that non-elite prospects are less valuable and studs are more valuable compared to a dynasty league where you can keep all 24 guys. It also makes rd 3-5 picks in the annual 10-round rookie/veteran draft more valuable, because there should be some pretty good veterans in there. I think I'd mostly target veterans in that draft from rd3 onward, treating it like a keeper league rather than a full dynasty league (in other words, like a redraft league but with some bonus for guys with upside who might emerge this season, especially if they're young).
  9. If Taywan Taylor counts as "late round" then put him at the top of that list.
  10. Understand your league's scoring system, and the relative value of players with those scoring rules. Dynasty rankings that you find on the web (by folks like FBG, DLF, and ZWK) can be a pretty good starting point for within-position rankings, but depending on your league's setup they might be wildly off for comparing positions (e.g. the value of a TE relative to a RB). You can look at last season's VBD totals for your league scoring to get a sense of relative value (or, even better, cumulative VBD over the past 3 years). Draft for best player available (rather than for your roster's needs) much more than you would in other formats, especially in the rookie draft (and also to some extent in the startup draft). You don't really know what your roster needs will be 2 years from now when those players finally hit (if they ever do), and you'll have plenty of opportunities to make trades. I also focus heavily on "BPA" in trades - are the players that I'm getting more valuable in this league's scoring system than the players that I'm giving up - with much less concern about my roster's needs. If you have a good sense of player values in your league's scoring system, then try to be one of the more active owners in your league in terms of making trades because every trade makes your team better (in expectation). It depends on the league, but often the most undervalued assets in dynasty are injured players and future first round (and perhaps second round) picks. So it's good to try to trade for them at a discount, and to be wary of trading yours away unless you really know what you're doing. If you ever reach the point during (or before) the season where it's clear that your team is not really in contention for the title that year, focus on building your roster for next year. That's an especially good time to try to trade for injured players and future rd 1-2 picks. It's also a good time to try to trade away your vets who only have a couple years left, and any likely one-year wonders, though try to get good market value for them rather than just ditching them for cheap. And you should be especially unconcerned about having an unbalanced roster if you're out of contention - it's okay if you have 8 TEs and a hole at RB if you're just trying to build roster value for the future.
  11. Hand size is related to drop rate, but it's not that strong a relationship. And John Ross had a better than average drop rate over the past 2 years, with just 6 drops on 105 catches+drops.
  12. John Brown 8.5", TY Hilton 8.5", DeSean Jackson 8.63", Mike Wallace 9", Antonio Brown 9", Greg Jennings 9".
  13. Let's talk about arms. Last year, information on prospects' arm length (and hand size) started to get more publicity, and I made the decision to treat a receiver's arm length as being just as valuable as height, inch-for-inch. I didn't have any data to back this up, but it seemed to me that the main benefit of a receiver's height is the increased catch radius, and an inch of arm length should increase their catch radius by at least as much as an inch of height. Arms probably increase catch radius by more than height does, since arms can reach in all directions, though height has the added benefit in helping to block out defenders. This week, I looked up the arm lengths of all 51 receivers in my successful/promising lists, and found the data for all but 2 of them (I haven't found Julian Edelman or USC/NYG Steve Smith's arm lengths). And, as I showed in my last post, they do tend to have longer-than-usual arms; their arm length stands out even more than their height. Here's another way of looking at that trend. There is a pretty close relationship between a WR's height and his arm length; generally taller people have longer arms. But you can look to see which people have arms that are longer than you'd expect given their height, or shorter than you'd expect. Out of the 49 successful/promising WRs that I have data on, 12 had arms that were more than an inch longer than you'd expect given their height, while only 4 had shorter-than-expected arms by over an inch: Arm Length Minus Expected (inches) 2.08 Michael Crabtree 1.61 Hakeem Nicks 1.56 Dez Bryant 1.52 Odell Beckham Jr. 1.48 T.Y. Hilton 1.48 Mike Evans 1.38 DeAndre Hopkins 1.22 A.J. Green 1.19 Dwayne Bowe 1.17 Kenny Britt 1.12 Kelvin Benjamin 1.04 Santonio Holmes -1.12 Greg Jennings -1.44 Miles Austin -1.94 Eric Decker -3.57 Doug Baldwin Crabtree, Nicks, Bryant, Beckham, and Hopkins are/were all "plays bigger than his height" type receivers (the tallest of them is Bryant at 6'2.0"), and this looks like part of the reason why. Evans, Green, and Benjamin are just huge, with each of them measuring at least 110" in height+arm length (Calvin Johnson & Marques Colston are also in that club; they are joined by only one of the 101 WRs from the 2016 & 2017 combines - De'Runnya Wilson). At the other extreme, Doug Baldwin is reported as having 27" arms by Player Profiler, but that might be an error - I couldn't find data on his arms in any of the other sources that had data on most of the players (like or cbssports), and 27" is a full inch shorter than the shortest arms among the 330 players at this year's combine (WR Ryan Switzer at 28"). But I've decided to treat 27" as accurate and include Baldwin in all of my analyses. I have added arm length to my WR spreadsheet, though I only have data for the 2 most recent draft classes and the 49 successful NFL WRs who I looked up. For this year's draft class, here are the receivers whose arms are longer or shorter than expected given their height, by at least 0.98 inches: 1.55 DeAngelo Yancey 1.52 Taywan Taylor 1.47 Keevan Lucas 1.33 ArDarius Stewart 0.99 Shelton Gibson -1.12 Trent Taylor -1.16 Jerome Lane -1.16 Malachi Dupre -1.23 Josh Reynolds -1.32 Kenny Golladay -1.40 Josh Malone -1.76 Noel Thomas, Jr. -1.92 Ryan Switzer And for last year's draft class: 1.65 Charone Peake 1.54 Malcolm Mitchell 1.46 Keyarris Garrett 1.08 Pharoh Cooper 1.04 Johnny Holton 1.00 Chris Moore -1.06 Trevor Davis -2.06 Nelson Spruce -2.28 Mekale McKay -2.34 Robby Anderson
  14. Last year I took a look at which college stats are the strongest predictors of NFL success, by seeing on which college stats the successful NFL WRs tended to stand out the most. One rough way to see which combine numbers are important is to do the same thing with them. For example, if BMI is related to a WR's success, then we should expect to see that the average successful NFL WR is above the 50th percentile in BMI, compared to the other WRs who show up at the combine. This isn't a perfect measure - my view is that low BMI is a negative but high BMI isn't helpful, and that won't be captured in this analysis - but it's a decent starting point which makes it easy to compare different player features. I ran the numbers for the 32 successful WRs who are in the Elite NFL WRs tab of my spreadsheet - that is every WR who entered the NFL since 2006 and has at least 100 career VBD (though some WRs are missing from some of the measures, e.g. not everyone did the 3 cone drill). I also re-ran the numbers on a larger group of 51 WRs, which includes the 19 guys in the Promising NFL WRs tab. As the baseline to compare them to, I used the WRs from the 2016 & 2017 NFL combine, since I am missing data on some measurements from previous years (especially arm length and non-rounded height). Here are the numbers for the Elite NFL WRs who I have each measurement for (which ranges from all 32 for ht/wt/bmi down to 17 who did the bench press), with the numbers for the larger set of WRs in parentheses: 0.75 Vertical (n=22, 0.69 for all 39 WRs) 0.73 40 Time (n=29, 0.70 for all 47 WRs) 0.61 BMI (n=32, 0.58 for all 51 WRs) 0.60 Arm Length (n=31, 0.60 for all 49 WRs) 0.60 Weight (n=32, 0.60 for all 51 WRs) 0.60 3 Cone Drill (n=20, 0.55 for all 35 WRs) 0.60 Bench Reps (n=17, 0.66 for all 30 WRs) 0.56 Short Shuttle (n=21, 0.53 for all 37 WRs) 0.53 Height (n=32, 0.54 for all 51 WRs) 0.52 Broad Jump (n=20, 0.51 for all 37 WRs) It's interesting that all of these come out above 0.50 - the successful NFL WRs were better than average on every metric (though sometimes not by much). The vertical and 40 are on top, and the similar-seeming broad jump comes in last. BMI and arm length are the top body size measures. It is worth noting that all of these numbers are somewhat noisy estimates and the order could easily change with a larger sample of WRs. I would especially put less stock in the numbers with a smaller sample size, because those both have more noise and may be misleading if a nonrandom subset of WRs chose to skip that drill (e.g. if the WRs who are bad at bench press tended to choose to skip the bench). Update: I reran the numbers using all WRs from the 2006-2016 combines as the baseline to compare to. I left out Arm Length (which I don't have that data for) and Bench (because my source of data has a bunch of errors for that; I have corrected a couple errors that I found which influenced the Bench numbers above and it might still have some errors). My source's numbers for Height are rounded and may have some errors too but I've included Height & BMI anyways. Here are those numbers: 0.65 Vertical (n=22, 0.59 for all 39 WRs) 0.65 BMI (n=32, 0.62 for all 51 WRs) 0.62 40 Time (n=29, 0.60 for all 47 WRs) 0.61 Weight (n=32, 0.60 for all 51 WRs) 0.61 Broad Jump (n=20, 0.60 for all 37 WRs) 0.56 3 Cone Drill (n=20, 0.52 for all 35 WRs) 0.54 Short Shuttle (n=21, 0.50 for all 37 WRs) 0.52 Height (n=32, 0.52 for all 51 WRs) This means that (for example), of the 22 successful NFL WRs who did the vertical, on average they would be in the 65th percentile relative to all of the WRs who did the vertical at the combine in 2006-2016. This is fairly similar to the first set of results. The main difference is that the 3 athleticism measures have converged to the 0.60-0.65 range: 40 Time & Vertical are lower and Broad Jump is higher. The agility drills (3 cone and short shuttle) have dropped a bit.
  15. Some of this data is readily available - for example, you can see the BMI of all WRs drafted in the first 3 rounds in 2006-2016 in the By Draft Pick tab of my WR spreadsheet. Low BMI WRs are less likely to be drafted early (AJ Green and Ted Ginn are the only sub 26.0 guys taken in the top 20 picks), but there does also seem to be a pattern of low-BMI WRs being less successful than other WRs drafted around the same spot. If we compare all WRs who weighed in at the combine in 2016 or 2017 vs. all WRs who were drafted in the first 3 rounds in 2006-2016 vs. the 51 elite/promising WRs (who have 100+ career VBD, or seem likely to get there, or seem like they would've gotten there if not for injury or character issues), here are the fraction that were below various BMI cutoffs: BMI Combine Rd 1-3 Successful <25.5 16% 9% 4% <26.0 24% 18% 10% <26.5 36% 33% 22% Only 10% of the Successful WRs had a BMI below 26.0, compared with 18% of the WRs who were drafted in the first 3 rounds and 24% of the WRs who weighed in at the combine in 2015-16. (I showed multiple cutoffs in this chart, and used round numbers as cutoffs, to try to avoid stacking the deck in favor of my hypothesis with carefully selected cutoffs.) The two main things that I'm interested in are 1) predicting what a player's NFL career will be like based only on my evaluation of the player (ignoring what scouts are saying, NFL draft position, etc.) and 2) predicting what a player's NFL career will be like while incorporating all relevant information (including where he's drafted, what experts say about him, etc.). It's true that I don't necessarily know at which step in the process a particular receiver failed to develop into a successful NFL WR (was he just not given an opportunity, or used in a role that didn't play to his strengths, or ...), and figuring out which step is an interesting question, but I'm more interested in the bottom line question of whether he will succeed or not. The 180-189 lb. WRs were just an example; you can also pick another weight range and see similar things. For example, here are the data for 193-202 lb. WRs (I chose that weight range because it means that anyone 6'2" or taller has low BMI and anyone 6'0" or shorter does not). Or you can play around with the PFR search and see what you fine. And the chart I gave earlier this comment (about combine vs. rd 1-3 vs. successful) is another way of looking at the pattern. It's true that there have been successful WRs of various shapes and sizes - AJ Green, Carolina's Steve Smith, Anquan Boldin, etc. But some shapes/sizes are more common than others, and we're trying to play the odds. If you can pick a prospect who has a 30% chance of succeeding rather than one with a 15% chance of succeeding, then you'll end up drafting twice as many good players. And if you compare successful NFL WRs to other WR prospects, you'll find many factors on which there's a trend (even though there are exceptions to pretty much any trend, sometimes lots of exceptions) - the successful WRs tend to be taller, heavier, faster, more athletic at jumping, and have longer arms. In college, the successful WRs tended to have more TDs, more long receptions, a higher yards per target, more big players in the running game or return game, a larger market share of their team's passing yardage, etc. (I summarized some of these differences in this post last year.) A player might wind up succeeding even if he doesn't measure out very well on many (or any) of these metrics, but you'll have a higher hit rate if you pick players who look good on these metrics rather than players who look bad at them.