Langford was not kept because we had equivalent and cheaper options on the team, not because he was a bad pick. Vontae was let go because the current staff didn't care for him and felt the return was better than having him as the 3rd or 4th CB. Your points here ignore the nature of NFL economics. But rather than your continuing to harp on the same tired old points with no comparison to how other GMs have performed, why not show how or why Ireland is anything less than an average GM. Show some comparisons to other GMs. And if he needs to be replaced, tell us with whom and why we should believe that such person would do any better.
How about we all agree that Ireland is better than most would like to admit, but even with that most still think Ireland should be fired so that the franchise truly gets a clean slate.
There are really a couple of big issues with this Career AV approach. The first is glaring. This study becomes overwhelmingly lopsided toward 2008 (the year Miami used the most draft value of any team in the Draft because they had the #1 overall pick). The problem is the use of Career AV sums rather than some sort of approach that controls for time. Think about this. Chad Henne has a Career AV of 18. Jason Pierre-Paul has a career AV of 16. ...Let me repeat that: Chad Henne has a Career AV of 18. Jason Pierre-Paul has a Career AV of 16. ... If you're trying to claim some sort of objective, scientific approach, and you're looking at that discrepancy, and you don't think there's some sort of obvious systemic problem in what you've constructed, then I don't know what to tell you. The best draft pick that Jeff Ireland ever made, from a pure "AV" standpoint was Jake Long. And lucky for him (as far as this study is concerned) that happened in 2008, where Jake gets to rack up 4 seasons of cumulative AV scores versus poor chaps like Jason Pierre-Paul and Von Miller who have only had 2 seasons and 1 season, respectively. One reason that is a curious design choice by the study's designer is because typically you find that those people arguing on behalf of Jeff Ireland prefer the exact opposite. They want more weight placed on later drafts than earlier drafts. In this case the 2011 Draft of which Jeff Ireland's full control is a matter of unanimous consensus, is weighted about a quarter of that of the 2008 Draft which many people actively argue was not in Jeff Ireland's control at all. This is not a matter of preference or one choice being as good as another, you prefer Red and I prefer Blue, etc. This is pure oversight on the part of the study's designer. You can see that best when you examine the further treatment of the Career AV numbers by the study's designer, where he essentially divides the Career AV production by the amount of "draft dollars" used. This is an admirable idea, though the execution of it is misguided. The problem is that the numerator and denominator are disconnected as far as time weighting is concerned. The numerator is the Career AVs which are subject to how many years the player has had the opportunity to accumulate AV. The denominator is Chart Value, where 1 point of Chart Value is worth 1 point in 2008 and also 1 point in 2011. This is pure design flaw, not a matter preference. It is not a logically sound design because it produces a bias not only toward the early years being more important in the comparison than the later years, but in particular if those early years were ones in which you exercised a lot of Chart Value (ahem, #1 overall pick) then that makes the study even MORE biased in your favor. Let me walk you through an example. Two teams, Team A and Team B. Team A used 1500 points of Chart Value in 2008, but only 1000 points in 2009, 2010 and 2011. Team B used 1000 points of Chart Value in 2008 through 2010, but 1500 points of Chart Value in 2011. Let's also pretend that Team A and Team B both have an identical tendency to produce 25 Career AV points per season, per 1000 Chart Value points. The teams have identical draft potency. However, they don't come out identical in the study posted by the original poster. Despite their identical draft potency, Team A will have producd 150 Career AV in 2008, followed by 75 Career AV in 2009, 50 Career AV in 2010 and 25 Career AV in 2011. Team B on the other hand will have produced 100 Career AV in 2008, 75 in 2009, 50 in 2010 and 37.5 in 2011. That gives Team A a 66.7 Career AV per $1000, whereas Team B has a 58.3 Career AV per $1000. Yet the *only* difference between the two is that Team A had its big allocation of Chart Value in 2008, whereas Team B had its big allocation of Chart Value in 2011. Again this is design flaw, not a matter of preference. It's a design flaw which is coincidentally beneficial to the Miami Dolphins in this study. I'm not accusing anyone of rigging anything. But the simple fact of the matter is that in 2008 the average team utilized 1897.2 points in Chart Value whereas the Miami Dolphins utilized 4298.7 points in Chart Value. The Dolphins essentially ARE "Team A" in the example above. They utilized 4298.7 points in Chart Value in 2008, 1782.2 points in Chart Value in 2009, 1507.6 points in Chart Value in 2010, and 1430.7 points in Chart Value in 2011. In other words, the way this study has been set up specifically rewards the Dolphins not for being better drafters, but because they used a TON of Chart Value in a year in the study which becomes weighted the highest. Like I said, pure design flaw. The question becomes, how do we adjust the study? Do we just turn Career AV into per season average, which would make it so that 2010 and 2011 mean as much as 2008 and 2009? Well, that's a start. It would be irresponsible and misguided not to at the very least start with that adjustment. But some limitations would remain even if you do that. The study would still have a de facto "weighting" toward the earlier drafts rather than the later drafts. Look at Jason Pierre-Paul as a for instance. As a rookie he only earned 2 "AV" from the PFR website. However, as a second year man he earned a nice 14 "AV" from them. Because he's only been in the league two seasons, an entire half of his career's "per season AV" would be weighted toward that rookie season. In other words, he's an 8 "AV per season" player according to the study, whereas in actuality we feel he's probably closer to that 14 "AV" number. Let's say that in the next two years that's exactly what he does...14 "AV" per year. That means his Career AV cumulation would be at 44, which averages out to 11 "AV" over his four year career. That would mean for the study's purposes, the GM that drafted Jason Pierre-Paul wouldn't get full credit for him purely because he took him in 2010 as opposed to 2008. And if your study was intended to show how GMs did over a full span of drafts, not with any weight given to one or two years where this GM happened to have "B+" drafts as opposed to later years where he had "C-" drafts...then this de facto weighting is a problem. The other criticism I would have for the study is in the "AV" statistic itself. From looking at it, I get a feel that the statistic is a little too dependent on whether a player played, as opposed to whether a player played well. The fact of the matter is Jeff Ireland has phased out just about every player that was on the roster prior to 2008, and replaced them with his own draft picks. This has meant plenty of opportunity for Jeff's own draft picks to go out there and play. The year that had a roster that was most leveraged toward pre-2008 players was that 2008 season when, coincidentally or no so coincidentally, the Dolphins went 11-5. That year, there were 16 of 24 "starters" that were pre-2008 players (Ronnie Brown, Greg Camarillo, Ted Ginn, David Martin, Samson Satele, Ikechuku Ndukwe, Vernon Carey, Vonnie Holliday, Matt Roth, Channing Crowder, Joey Porter, Will Allen, Andre Goodman, Yeremiah Bell, Renaldo Hill and Brandon Fields). By 2009, that was way down to only 9 of 24 (and that's counting Jason Taylor, who arguably shouldn't be counted because he was jettisoned and then brought back by this staff). In 2010, it was down to 5 of 24, then last year 4 of 24. This year I think the projection will be 2 of 24. Given that kind of phasing out, there is a lot of opportunity for your own draft picks to play and start games, and if this stat is a little over-leveraged toward who is starting rather than whether they're playing very well...well, you've got a little bit of a problem. The construction kind of assumes that a player has to be a certain requisite talent level across the league in order to play and get playing time. But getting playing time on a team like the Giants isn't exactly like getting playing time on a team like the Dolphins. Anyway, people are asking for the sources of disconnect between the study's results and the team's (lack of) winning prowess particularly in 2009, 2010 and 2011 when more of the Dolphins' draft picks were playing and starting (there were only 2 Ireland draft picks starting in 2008)...I think the above are probably some pretty good reasons for why that is the case.
I'm not "quibbling" over a few points, I'm pointing out that the system you employed has two players very close to each other who, in reality, are light years apart. As I've said, I don't have time to do a comparison for all 32 teams at the moment, but if you genuinely can't see how someone might have a problem with a scoring system that scores Koa Misi and JPP roughly the same, then I have to wonder how sincere your requests for a factual discussions actually are.
Forget Brady then. How about the incredibly easy schedule we had where we faced something like 6 backup QBs, including Seneca Wallace, whoever was QBing STL at that time, JP Losman, etc. All of that factored in to a mirage of a season. I believe we also only turned the ball over 10 times all year, which is a good deal of luck rather than skill. And other than Penny's injury in 2009, when we were already 0-3 by the way, losing Henne made no real difference in terms of wins or losses. Fair enough, but Rome ought to be more than a wood frame house and a big billboard saying "Coming Soon!" 5 years after you begin work on the project Difference is, those other guys you mentioned have all been instrumental parts of playoff teams whereas Long has not. None of this is to suggest that Long is not great or that he wasn't worth the #1 pick, but he is a complementary player. A LT is not the centerpiece of a SB team, particularly in today's era. Never said Long was a abd choice but he was not the BEST choice, and good GMs, should make the best out of aseries of good chocies most of the time. And again, Matt Ryan's numbers are comparable to Peyton Manning's over their first four seasons. I posted it earlier. Under your metric, a team passing on Manning would have been justifed after the first 4 years b/c Manning would not have been a franchise QB. And once again your entire analysis misses the point. I don't care what other teams may have done with a particular pick. I care about what Ireland did. He liked Jimmy Graham and passed to take lazy, fat, out of shape OL. And you again try and make it seem like Jason Worilds is horrible. He is not. Point being, an argument in support of Ireland that amounts to "these other guys screwed up too" is a horrible argument. Espescially considering most of the other GMs you mentioned have a rather long history of success and screwing up leass than most others. You seem to excuse Ireland's lack of vision on Graham because other teams also lacked the vision. Wouldn't it be nice to actually have the guy with vision for once?
I guess I didn't see them as very valid and kinda lost them in the flow of this thread, but I'll address them now. Post #110 This simply misses the point. Nobody is suggesting that it is the be all and and end all stat. But it is an objective one and I don't see anyone offering a better one. What are the positions that Ireland has ignored in his 4 years here? QB? Nope. Used a 1st and 2 seconds on the position. Brought in Pennington, Moore and Garrard. WR? Nope. Used 2 seconds on Marshall, a third on Turner and 4ths on Hartline and Gates. Marshall was plenty talented but the new coach didn't want him. So what was Ireland supposed to do with him? The new coach is on record saying he doesn't need or want the proverbial no. 1 receiver. Pass rusher? Cam Wake may be the best in the game. Brought Jason Taylor back. Drafted Odrick (who in one season has more sacks than Brandon Graham and Jerry Hughes, the popular pass rush choices, in their entire careers combined)
I've never said he should be. I've said that I don't believe Ireland is good enough to turn a talent depleted team into a contender; in order to do that, you need to hit more and harder than he has done. Would I fire him if I was Stephen Ross? Yes I would - if I was convinced I could get a guy I believe to be better. That's not the same at all as shouting "fire Ireland!"
But the problem is that it may not be a valid stat. A bathroom scale is objective, but if it tells you the wrong weight, it isn't valid.
I don't know if this has been addressed in the thread but it's a different passing environment in today's NFL. Directly comparing passer ratings from two different era's is an apples to oranges comparison. The more accurate way of comparing the two would be to look at where they were ranked among their peers. In Marino's first four seasons, he was ranked: 3rd, 1st, 5th, and 2nd among his peers in passer rating; giving him an average ranking of 2.75. In Ryan's first four seasons, he was ranked: 11th, 20th, 11th, and 8th among his peers in passer rating; giving him an average ranking of 12.5. As you can see, Ryan isn't comparable to Marino at all.
I didn't weight the years differently because the question I was trying to get at was how much total value did a team get from their draft picks per unit of draft currency during a particular time period. So that is what I looked at. I have no personal love for or bias in favor of Jeff Ireland. I have never met the guy and have relationship with him. And I have no interest in having a bad GM, if he is really bad. Nor was there any oversight on my part. I used the most straightforward approach to answer the question I posed, which was which team has gotten the most value out of their draft picks from 2008-2011 per unit of draft currency. One could absolutely try to do a similar analysis on a per year played basis and I invite anyone interested in that to do the work. The results may not differ as much as you think. The Falcons 2008 draft has a similar effect on its numbers, as does the Ravens' 2008 draft. Etc. But I too would like to see how such a study turns out. Henne has a higher CarAVthan JPP because he contributed at a more important position than JPP for a longer period of time. I haven't said Henne is a better player than JPP nor has PFR. But he played for 2+ years at QB compared to JPP's 1+ year at DE.
Then tell me why it is invalid. Again, it is not intended to show precisely how good an individual player is, but it does do a pretty good job of showing the relative value of different players at different positions. If you disagree, then please explain why. And then please explain how it skews things in favor of Jeff Ireland.
Sigh. You say being a GM is not about numbers but say Ireland shouldn't be fired but isn't good enough and should be replaced, based on numbers. I'm fairly certain it would take considerable effort to contradict yourself more.
I think CK already did a pretty swell job. I don't know what you'd like people to add. The numbers obviously favor Miami by default due to the unusually high draft currency they had in the most "valuable" season, and they also obviously skew things in favor of players with a lot of starts rather than a lot of talent and/or production. I really admire the effort you put into it but I genuinely see no merit in these numbers in regard to draft day success.
Do you actually bother to read a post before you hit "reply" or are you on some kind of let's see how often I can pointlessly disagree spree?
No matter how you change the numbers, it still won't account for the fact that Ireland hasn't done a good job of drafting players that fit into a cohesive team that can produce results. It would still be flawed. Still missing skilled positions. You pointed out that he didn't technically "ignore" the positions and you're right. He just either didn't spend high enough draft picks (thus resulting in failed picks), ignored viable free agents, or jettisoned / traded away effective players. You could use team metrics to evaluate Ireland, but then you would run into a real problem eliminating the role the coaches played in the team metrics. As much as I love statistics, sometimes you just have to take a real close look at the product and judge it for what it is. And the product Jeff Ireland has put together hasn't been very good.
Some years the schedule is favorable and some years it is not. That si true for every team. If we are going to compare the Dolphins record to other teams we'd need to similarly discount the seasons when they had easy schedules. Take the Falcons for example, as they have been front and center with the Matt Ryan issue. In 2010, when they went 13-3, they had a very easy schedule. They got 2-14 Carolina twice, played the pathetic NFC West that Seattle won at 7-9, got Cincy, Cleveland, etc. So do discount that season too, and if so, by how much? Long was an instrumental part of a playoff team in 2008. Matt Ryan hasn't been the centerpiece of a SB team either. But you are holding Ireland to a standard that no other GM can be expected to meet either. None of them make the best choice most of the time. None of them even come close. Ryan's numbers are nice. My concerns with him have nothing to do with his overall numbers. I just don't think he is the guy to carry a team when the other pieces break down. The QBs that are worthy ofthe No. 1 overall pick can do that. Ryan really hasn't shown that he can. He's had opportunities, but hasn't done it. No, it is you that misses the point. You want to criticize Ireland for something that all of the other supposed top GMs did too. You chose to focus on a single choice and all of the other guys made a similar choice to pass on Graham. So why should Ireland be fired for making the same choice that the best in the league made too? Why should he be held to a different standard? Loomis is the only guy who we know had that vision on Graham. I don't think he is or will be available, but if he is, I'd be OK with replacing Ireland with Loomis. Of course, there's no certainty that he'll have the same "vision" when the next Jimmy Graham comes along, but ....
Fine, how did I misrepresent your statements? Best I can tell you said the following contradictions: - You can't define a GM by numbers but say Ireland hasn't gotten enough talent = 1 contradiction - Ireland shouldn't be fired, but replaced if you can find someone better, but you can't know if a GM prospect will be better = 2 contradictions That's a total of 3 contradictions.
If that's really the best you can tell (and I don't believe you, but never mind), then let's just leave it at that. I really don't feel like killing an army of straw man today.
Once again, the results aren't entirely on Ireland. When Sparano was here he put together a cohesive group of players to play the style that Sparano wanted to play. That style has now changed under Philbin and the early indications are that he is choosing players to Philbin's specifications. Philbin didn't want Marshall so off he went. Vontae too, although in slightly different context. Ireland has compiled a decent group of skill position players -- Tannehill, Bush, LMiller, Dthomas, Fasano, Bess, Marshall, etc. Again, Marshall isn't here largely because Philbin didn't want him and felt he didn't fit what Philbin wanted to do, not because Marshall isn't a talented skill position player.
I'm not arguing with you to argue with you. I actually never do that. I'm simply stating things as I see them. Show me where I'm misrepresenting you.
Is there actually any kind of objective evidence to that fact? Serious question, maybe I missed something completely.
To me, CarAV seems like more of a team stat, rather, a players approximate value to a team, than it does a generic individual stat. I think it would be a better gauge of a GM's success if this were baseball and we used WAR (Wins Above Replacement). Basically, if I take a team that has a collective WAR of 60, it should finish with a better team record than that of a team whose WAR is 50. For whatever reason, metrics in football haven't caught up with those of baseball and basketball. My guess is it's because, IMO, it's the ultimate team game with many more variables.
You are responding to my statement out of context. It was in response to a statement by Shouright that was different from the issues raised by CK. But again, the objective was to determine how much value a team's drafted players provided from 2008-2011 per until of draft currency. The 2008 draftees generally provided more value because they played more than the 2011 draftees. They have provided 4 seasons of value. And the Dolphins' relatively high amount of draft currency that year was talen into account by doing the analysis per unit of draft currency. Other teams, such as the Falcons, Ravens, Broncos and Eagles also got a similarly high proportion of their CarAV from that draft. I'd be very interested in seeing it broken down on a "per player season" basis as CK suggests. Unfortunately, I don't have the time to do that now. I suspect we would still see Ireland come out as an above-average GM.
Okay then: I said you couldn't put a number on GM success. That doesn't mean that you can't evaluate a GM based on what he did. It means that you can't really evaluate him in comparison to 31 other GMs. There's no stat like GM rating (even if Fineas tried to establish one). You can only evaluate him based on whether or not you liked what he did and are comfortable going forward with the "production" you got from him. Basically, I'm saying that you can only evaluate a GM on an island, so to speak. For instance, it's tough to compare Jeff Ireland to Ted Thompson, when both picked at extremely different draft positions with an extremely different player pool, extremely different needs and extremely different systems in place. Do I think Jeff Ireland is good enough to get us out of the mess I think we're in? No, I don't. We're so talent depleted, we need more than a couple of solid drafts; we need to hit one or two hard. Has Ireland hit hard on a draft yet? No, he didn't. So if I had someone I was convinced could do it, yeah, I'd hire him.
Let me give you some rubber-meets-the-road examples of how the flawed construction of this study can skew the results. Let's say that every team in the entire National Football League has the SAME draft potency. That is to say each team produced 20 CarAV per 1000 points of Chart Value. The #9 team on the study's original list produced about 22 or 23 CarAV per 1000 points of Chart Value, so this hypothetical is loosely centered around whichever teams would have been #10 to #15, guestimated. The actual number doesn't matte, what matters is that all teams have the same exact potency. What effect would the pure TIMING aspect of the distribution of Chart Value have on the study? All 32 teams would have literally the same draft prowess, but the flaws in the structure of the study itself would produce the following results: #1 . Dolphins 23.9 #2 . Jets 23.5 #3 . Falcons 23.3 #4 . Bears 22.8 #5 . Chiefs 22.5 #6 . Raiders 21.7 #7 . Rams 21.5 #8 . Ravens 21.4 #9 . Packers 21.3 #10 . Jaguars 20.8 #11 . Saints 20.7 #12 . Lions 20.5 #13 . Steelers 20.3 #14 . Bengals 19.7 #15 . Seahawks 19.7 #16 . Patriots 19.5 #17 . Redskins 19.2 #18 . Cowboys 19.1 #19 . Giants 19.1 #20 . Eagles 19.1 #21 . Bills 19.0 #22 . Buccaneers 18.6 #23 . Panthers 18.6 #24 . Texans 18.4 #25 . Broncos 18.3 #26 . Titans 18.3 #27 . Cardinals 18.2 #28 . 49ers 18.1 #29 . Chargers 17.9 #30 . Colts 17.8 #31 . Vikings 17.0 #32 . Browns 16.9 Without a timing bias, pretending that every team executed the same amount of Chart Value every year, every one of the above readings would have been "20.0"...a 32-way tie. The Dolphins were not the only team to benefit from the flaws in the structuring of the study, but they were the team that benefited the MOST from those flaws.
There are a myriad of metrics like that. EPA (expected pts added), WPA (win probability added), VOA (value above replacement), etc. The only problem with those is that offensive lineman are grouped together (entire line is measured, not individual player).
Armando Salguero "Parcells takes full responsibility for 2008 draft, but gives 2009 decisions to Ireland. Unhappy with 2009 draft, Parcells took back 2010." "In 2010 Parcells drafted and left in October,Ireland has run the show ever since" Armando has a lot if other stuff in his recent tweets pertaining to this subject. Smith, Hartline, Clemons, Gardner, 2nd round pick acquired for future draft, still on team., if Smith plays like I think he's gonna play, I would say slightly above average draft including that extra pick.
Philbin's statements at the time make that fairly clear, although he was not explicit about it. The circumstances also strongly suggest that. I think it is fairly clear that Brandon Marshall would not have been a subject Philbin was ambivalent about. If he wanted Marshall here and Ireland traded him out from under Philbin against Philbin's knowledge or wishes, we certainly would have heard about it. We do know that Philbin and Ireland work together and discuss major personnel moves. We know Philbin said he doesn't believe you need to have a No. 1 receiver. We know that the team was planning, or at least thinking, about drafting a first round QB and that Marshall can be very hard on QBs. We know that Philbin is a bit "prudish" for lack of a better term and is extremely concerned about appearances and how the players conduct themselves in public and in the locker room. I think it all paints a reasonably clear picture that Philbin would not have been a big Brandon Marshall fan. Dave Hyde reported that Philbin was the one who wanted Marshall traded: http://articles.sun-sentinel.com/20...826-1_1_philbin-vontae-davis-brandon-marshall Barry Jackson similarly reported that Philbin didn't see Marshall as a good fit. http://www.miamiherald.com/2012/03/14/2691697/miami-dolphins-trade-receiver.html#storylink=cpy
This is basically it. Baseball, and to a lesser extent basketball, are individual sports masqueraded as team sports. Put Miguel Cabrera on Seattle and very little changes for him individually. LeBron on the Cavs is a 60+ win team. Not really the case in football.
Not sure I understand where those numbers came from, but again it is not a flaw in the structure. The thesis was "Which teams got the most approximate value from their draft choices from 2008-2011 per untit of draft currency. If you want to do a different, but similar study, to determine whose draft picks had the highest average AV per season played per unti of draft currency, that would be great. I'd like to see the results of that.
I didn't misrepresent you then, you just weren't clear. A GM does two main things. Acquires talent and makes deals (draft & FA). You can absolutely gauge a GM against other GMs on those tasks. Now this all goes back to evaluating Ireland's performance with uncorrupted data. No one can do that for the period priors to years ago.
It is true that an NFL player's performance depends on lot on his environment. But this isn't being used to show the value of individual players -- it is being used to show the value of a collective group of draftees on the very team for which they were drafted. Again, AV is not the ultimate or perfect stat, but it provides the best measure I am aware of for comparing the value of players at different positions. If you know of a better one, then let me know. If you are aware of another objective measure (whethewr better or not) that shows that Ireland has been a below-average GM, I'd be very interested in hearing about it.
How does the notion that Ireland had control of the 2009 draft mesh with Parcells' statements that he took responsibility for the Pat White pick (in 2009)? That 2009 draft was looking pretty good between then and the 2010 draft, with the exception of White and Turner. What was it that Parcells was supposedly so unhappy with Ireland about? Turner? I'm not buying it, but . . .
To be honest, if there's a stat that could accurately measure this, I don't think it's been invented yet. The main problem I have is, a stat should somewhat mirror team success. WAR is great indicator of team success, as is Win Shares, etc. Because the Dolphins aren't the 11th best team in the league, which CarAV would lead us to believe when looked at in a vacuum, then something doesn't jive. Either the stat is incredibly flawed OR the Dolphins are the unluckiest team in the league.
I was only commenting on his post in a vacuum. I agree with your overall premise and appreciate the time you put into researching the original post.
And if he did, then the most parsimonious explanation for the Dolphins' relatively mediocre to poor record during his tenure is that our quarterback play has not been adequate, and that gets you back to the Matt Ryan decision.
