Tim Couch Pulls Out Posted April 5, 2016 Report Posted April 5, 2016 http://coldomaha.com/2016/vikings/combing-the-combine-to-uncover-pass-catchers/ Long story short, using a statistical model that places quality scores on receivers based on an age-adjusted formula created by rotoviz and then calculating how many "adjusted" yards a receiver will earn based on their quality score, this author tried to find value picks in this year's draft while also uncovering a few players that scream "bust" - i.e. won't reach their projected level of adjusted yards based on draft position. Notable top level receivers - Coleman Thomas Shepard Carroo Louis Horn Johnson Notable mid-level receivers - Doctson Miller Listenbee Garrett Notable busts - TreadwellFullerBoydCooperHiggins Outside of Boyd, who I believe could be a good utility player and decent #2, I'm inclined to agree with most of these findings.
Tour2ma Posted April 5, 2016 Report Posted April 5, 2016 Interesting read... also enjoyed the similar LB article.
The Gipper Posted April 5, 2016 Report Posted April 5, 2016 http://coldomaha.com/2016/vikings/combing-the-combine-to-uncover-pass-catchers/ Long story short, using a statistical model that places quality scores on receivers based on an age-adjusted formula created by rotoviz and then calculating how many "adjusted" yards a receiver will earn based on their quality score, this author tried to find value picks in this year's draft while also uncovering a few players that scream "bust" - i.e. won't reach their projected level of adjusted yards based on draft position. Notable top level receivers - Coleman Thomas Shepard Carroo Louis Horn Johnson Notable mid-level receivers - Doctson Miller Listenbee Garrett Notable busts - Treadwell Fuller Boyd Cooper Higgins Outside of Boyd, who I believe could be a good utility player and decent #2, I'm inclined to agree with most of these findings. Interesting as these things may be....I doubt they have validity in terms of specifically predicting the future of any particular player. No more so than saying that 38% of the first of any position drafted has turned out to be a good player or not.
Tim Couch Pulls Out Posted April 5, 2016 Author Report Posted April 5, 2016 Interesting as these things may be....I doubt they have validity in terms of specifically predicting the future of any particular player. No more so than saying that 38% of the first of any position drafted has turned out to be a good player or not. Well that's something we obviously won't know for a few years - however, it purports about a 70% accuracy rate based upon prior drafts IIRC. This is the stuff I HOPE we're placing some emphasis on in the pre draft process (knowing full well that it hurts the argument for Wentz). Obviously I don't want our FO drafting players strictly based off of random algo's, but it SHOULD at least create some sort of roundtable discussion among the brain trust. There are value picks to be had at a number of positions - I'd love to get a few steals, as it were.
darren15 Posted April 5, 2016 Report Posted April 5, 2016 Glad I got Shepard in the 3rd for the Browns on our fake draft. Would rather grab him at 32 instead of some of the others mentioned...like Fuller and Smith My fave receiver in this draft
gumby73 Posted April 5, 2016 Report Posted April 5, 2016 Interesting as these things may be....I doubt they have validity in terms of specifically predicting the future of any particular player. No more so than saying that 38% of the first of any position drafted has turned out to be a good player or not. These kind of list fry me. First put them on a team, with a depth chart that may have a immediate need at position, than tell me who's their QB.. Crazy guessing nonsense in progress..
The Gipper Posted April 5, 2016 Report Posted April 5, 2016 These kind of list fry me. First put them on a team, with a depth chart that may have a immediate need at position, than tell me who's their QB.. Crazy guessing nonsense in progress.. See my list of WRs on the First vs. Best thread. On there are about 3-5 WRs who could be considered the best of their class though they were lower picks.....and one thing they had in common was that they had Ben Roethlisberger or Tom Brady or Peyton Manning throwing to them. So there is that factor.
Tour2ma Posted April 6, 2016 Report Posted April 6, 2016 None of which helps predict the future... The difference between what the article's micro analytics are trying to do and what a macro, statistical rehash of the past is restricted to doing could not be greater.
Tim Couch Pulls Out Posted April 6, 2016 Author Report Posted April 6, 2016 None of which helps predict the future... The difference between what the article's micro analytics are trying to do and what a macro, statistical rehash of the past is restricted to doing could not be greater. Nothing predicts the future. But given that the analytics, in a bubble, is quite literally the study of data to discover trends, I'd say this fits quite in line with what our FO is likely doing. Reaching 100% certainty of a future event is impossible, especially in a game where there are limitless uncontrollable variables.
Tour2ma Posted April 6, 2016 Report Posted April 6, 2016 Thus the use of "helps"... If a tool can improve upon a coin-flip like (or worse) probability of future success, then it has value.
gftChris Posted April 6, 2016 Report Posted April 6, 2016 Nothing predicts the future. But given that the analytics, in a bubble, is quite literally the study of data to discover trends, I'd say this fits quite in line with what our FO is likely doing. Reaching 100% certainty of a future event is impossible, especially in a game where there are limitless uncontrollable variables. Yup. You take an 'absolute lock' perennial all pro receiver, one bad tackle, multiple torn ligaments and you have a bust. You take a guy projected to do nothing, give him Peyton as his OC and Rodgers as his QB and hey presto, 1000 yard season.
The Gipper Posted April 6, 2016 Report Posted April 6, 2016 None of which helps predict the future... The difference between what the article's micro analytics are trying to do and what a macro, statistical rehash of the past is restricted to doing could not be greater. Sure....both efforts could not be different.....and neither are worth a cabbage fart when it comes to predicting the future when it comes to any particular player.
The Gipper Posted April 6, 2016 Report Posted April 6, 2016 Nothing predicts the future. But given that the analytics, in a bubble, is quite literally the study of data to discover trends, I'd say this fits quite in line with what our FO is likely doing. Reaching 100% certainty of a future event is impossible, especially in a game where there are limitless uncontrollable variables. Well, I hope they know what they are doing and can by use of anal-ytics tell which 5th round WR is going to turn into a superstar. We don't need any more Vincent Mayles.
The Gipper Posted April 6, 2016 Report Posted April 6, 2016 Thus the use of "helps"... If a tool can improve upon a coin-flip like (or worse) probability of future success, then it has value. But none of that can ever be know until that future arises. Its like science fiction: I just got done doing a binge watch on the Battlestar Galactica series. Here is a scenario where they were so advanced that they had like fold space engines and had "robots" that were indistinguishable from humans.....but all their data and communications was pretty much done on paper and cardboard files. This from a series that ran from like 2003-2008. They were not able to "imagine" the use of like Ipad computers or smart phones. Had that series been made a few years later I doubt they would have been using paper files in space. Same here. No one could have probably imagined that the best WR in the game may have been a guy picked in the 6th round by the Steelers.
calfoxwc Posted April 6, 2016 Report Posted April 6, 2016 might help. these kind of analyses are only as legit as the amount of relevant information put into them. a wr playing vs solid db's, with a weak armed, so-so qb, inadequate oline... won't have as good of stats as a wr playing against so-so dbs with an excellent qb with a solid oline.... natural grass vs artificial turf.... good weather vs bad weather.... seems more of an exercise in futility....
The Gipper Posted April 6, 2016 Report Posted April 6, 2016 might help. these kind of analyses are only as legit as the amount of relevant information put into them. a wr playing vs solid db's, with a weak armed, so-so qb, inadequate oline... won't have as good of stats as a wr playing against so-so dbs with an excellent qb with a solid oline.... natural grass vs artificial turf.... good weather vs bad weather.... seems more of an exercise in futility.... Sounds more like stuff a bookie would want to know.
gftChris Posted April 6, 2016 Report Posted April 6, 2016 might help. these kind of analyses are only as legit as the amount of relevant information put into them. a wr playing vs solid db's, with a weak armed, so-so qb, inadequate oline... won't have as good of stats as a wr playing against so-so dbs with an excellent qb with a solid oline.... natural grass vs artificial turf.... good weather vs bad weather.... seems more of an exercise in futility.... With a sufficiently large dataset you can correct for those factors.
Recommended Posts
Archived
This topic is now archived and is closed to further replies.