[S-Dubb]

Draft trades were big mistakes
Jets fail to play the percentages by moving up
By Michael Salfino



The Jets' Young Turks, GM Mike Tannenbaum and head coach Eric Mangini, made some old school mistakes last weekend at the NFL draft.


How the players the Jets traded up for ultimately turn out is irrelevant. As I stated in my draft preview, trading up is always a mistake given the way these picks are valued. To explain why, let's present all the points presented by the Jets and people outside the organization, who not only defended, but also generally lauded the decisions to trade up for Pittsburgh cornerback Darrelle Revis and Michigan linebacker David Harris.

Argument: According to the draft value chart -- used by teams to gauge the value of draft picks -- the Jets didn't pay too great a price to move up because the points for the draft slots they gave up was only slightly less than the points they acquired.

I love it when people with no scientific, economic or statistical background fabricate things like draft value charts out of whole cloth. Can you imagine Jimmy Johnson, the man who popularized the chart, sitting on his boat with bare feet, a Corona in one hand and a pencil in the other, coming up with this nonsense? But don't take my word for it.

Some real economists -- Duke's Cade Massey and the University of Chicago's Richard Thaler -- researched the 1999-2004 drafts. They concluded that this value chart is a farce that all teams should look to exploit by trading down whenever they find a partner willing to pay close to the chart price. Massey and Thaler determined that by measuring how first-round picks actually performed, assigning value to that performance and then subtracting what it cost to sign the draft pick, the assigned values are grossly inflated. So, according to this objective evidence, the Jets were very likely fleeced by essentially paying chart value in both trades.

Argument: Even if the chart is poorly conceived, the Jets, by all accounts, drafted two very good players.

If most everyone who claims to be an expert actually proved it by accurately predicting who the best players in each draft class were going to be, opinions like this would be meaningful. But even professional NFL GMs and scouts are not nearly as good at identifying amateur talent as they believe. That's why higher draft picks cost too much to trade up for and sign. If the NFL draft were a rational market, high draftees would have a high probability of earning their keep. They do not. In fact, Massey and Thaler concluded that there's about a 47 percent chance that a player drafted at a given position has about a 47 percent chance to outperform the previous player drafted as his position.

Argument: Who cares what the Jets are paying Revis? Woody Johnson can afford it. As a second-round pick, Harris will make barely more than the veteran minimum.

The salary cap makes the cost of signing players important. It's not just that the Jets are paying Revis more money than the player they would have taken had they stayed put at No. 25. The trades also cost the Jets second and third-round picks, which, if you are fortunate, are very cheap and locked up at a low cost for a minimum of four years. Saving money on the players selected with these picks gives you the opportunity to use your money on proven difference makers in the highly competitive free agent market.

Argument: The Jets had a need at cornerback and needed to upgrade their run defense with an inside linebacker. They took the best players on their board with these picks. So, give them credit for being aggressive.

Even if we accept the silly notion that this Jets team entered the draft with two holes, New York could have kept its extra picks and taken two corners and two run stuffers instead of one each. If you're trying to win the door prize, do you want four raffle tickets or two? It's at best 50/50 that Revis will be the best cornerback in the draft.

The first corners drafted in the last few years were DeAngelo Hall (2004), Terrence Newman (2003), Quentin Jammer (2002), Nate Clements (2001), Deltha O'Neal (2000). Only Hall and Clements made the Pro Bowl with the teams that drafted them. O'Neal eventually did after he was traded to Cincinnati by Denver, which drafted him at No. 15 and saw him struggle mightily. Let's include O'Neal anyway and say the chance of the first corner drafted turning into a top player is 50 percent.

As for Harris, he may well have been a value pick, but one of the reasons you play the 3-4 defense is because you're looking for different types of players than the crowd. Why not just wait and let these players come to you?

Argument: If the odds of finding a great corner in the first round and an inside linebacker in the second aren't great, imagine how much worse they are later on.

The odds are worse, but there's a better chance than most think of finding great players and even greater values later in the draft. The Jets hope that Revis turns into another Asante Samuel. Yet Samuel was drafted with the 23rd pick in the fourth round!

Jacksonville Pro Bowler Rashean Mathis was the ninth defensive back selected in 2003. When scouts were drooling over Jammer (drafted No. 5), the Eagles found Pro Bowler Lito Sheppard at No. 26. The Jets were slated to pick No. 25 last week. Cornerback Ronde Barber is probably going to the Hall of Fame and he was a third-round pick, the 17th defensive back taken in 1997.

Great inside linebackers can also be found in later rounds. Witness recent Pro Bowlers Zach Thomas (fifth round) and Jeremiah Trotter (third round). Throughout football history, inside backers have largely been products of their defensive systems. For example, 3-4 teams such as the Jets don't generally send inside linebackers to the Pro Bowl because these linebackers have to split tackles with another inside linebacker, while the middle linebackers in a 4-3 roam the middle alone.

Argument: What you say may be true most years, but this was, by most accounts, a weak draft. Teams say drafts are weak or strong based on the solely on projected first-round talents. But all drafts generally produce the same number of really good players if you look beyond the first round. Half of the guys named to the Pro Bowl every year weren't drafted in the first round.

Argument: Even if I buy all your points, how can you say that how the players turn out doesn't matter? If the these two guys end up being stars, those were great draft moves. Period.

You hear teams talk all the time about playing the percentages. The trades for Revis and Harris both went against the percentages. If you play for an inside straight and get it, does that mean you played the hand well? No. You're also emboldened to repeat your error. The odds are against you at the casinos, too. But if you have a good day and beat them, you're only that much more likely to press your luck again and end up losing it all back, and then some.

I agree that the Jets didn't risk too much here. But I was hoping this regime would be smart enough to exploit the value chart like the Patriots shrewdly did. Jets fans should worry that Tannenbaum and Mangini will make a habit of trading up and end up paying steep prices that could haunt the franchise for years. Ask Giants fans how trading Philip Rivers and what turned out to be Shawne Merriman and Nate Kaeding for Eli Manning has worked out thus far. Michael Salfino is a nationally syndicated football and baseball newspaper columnist and regular contributor to SNY.tv

[cmart=the best]

Ask Giants fans how trading Philip Rivers and what turned out to be Shawne Merriman and Nate Kaeding for Eli Manning has worked out thus far.

Oh shit that better not happen to us

[kelticwizard]

Yeah, there's no sense trading up at all. None.

Like in 1991, when the Jets' first pick was the second round. There were two QB's who they figured might be on the board in the second round. Atlanta picked just before them, at pick #33, and took one. The Jets just played it cool and took the other one. Afterward, they said that although they would have picked the guy Atlanta got, they liked their pick.

The picks?

At #33, Atlanta got Brett Favre.
At #34, the Jets got Browning Nagle.

No sense trading up at all. No sir.

[Jetsfan0099]

I dont think it was bad move, we got a really good CB in this draft, and that was a huge need for us. I am happy to get Darrelle Revis, I didnt think it was possible for us to land him. But Mr.T found a way.

[nysackexchange]

This guy can show percentages all he wants, no one will be able to say whether this move was a good one until 3 years from now when we see how good these guys are playing.

[Chadforpresidentin08]

I actually have to agree, we gave up a lot for a team that really needed to fill up some plugs. The Revis trade was a good deal, but the David Harris trade was pretty pricey, too pricey for my tastes. We're still lacking a good FB, a solid OG, and a solid DE opposite Ellis and NT. We could've really used those picks we gave up. I suppose trading up once is okay, but twice is too much, especially for a team like us.

[NYJBlueHen]

Yeah, there's no sense trading up at all. None.

Like in 1991, when the Jets' first pick was the second round. There were two QB's who they figured might be on the board in the second round. Atlanta picked just before them, at pick #33, and took one. The Jets just played it cool and took the other one. Afterward, they said that although they would have picked the guy Atlanta got, they liked their pick.

The picks?

At #33, Atlanta got Brett Favre.
At #34, the Jets got Browning Nagle.

No sense trading up at all. No sir.

For every situation like that, there is probably tons more that don't work out. Hell look at Robertson for us. Either way the Jets gambled and it might pay off or it might not, no way to judge either way right now. I trust the front office and I like that they were aggressive. They so guys they wanted and went and got them, I can't knock that. But I would have to agree with the strategy of stock piling picks, particularly in this years draft.

[S-Dubb]

I'm going to say the say thing I said on draft day. In many other situations we wouldn't have done it this way but when you get Jones and move back further on the board in the second and have elite corners that fell down the draft board you make moves. We could have chilled and picked out someone we really didn't want. Our hand was forced and we need to roll the dice.

Good job by Mannenbaum!

[kelticwizard]

Exactly. Nobody is saying that we should do it this way EVERY year.

But we won 10 games with the following statistics: When the Jets ran the ball, they averaged 3.5 yards per carry. When the other teams ran the ball against the Jets, they averaged 4.6 yards per carry. That is a glaring deficiency, but we still got to the playoffs.

We are just a little bit away from being one of the real good teams in the NFL, so we rolled the dice to plug the huge holes we had. We shouldn't do that year after year, but there wasn't that much between us and the best teams, so we did it.

The author, Michael Salfino, forgets that the percentages only work out for the long run-not necessarily the short run.

[Pennington]

What this guy is trying to do is trivialize the draft process. Turning it into a numbers game, like a calculated game of poker allows a person like himself to feel like he may actually possess the skills necessary to be a GM.

The one factor that has been completely removed from his argument is scouting. He would like us to believe that all GMs are created equally. Hence on an even plain, we can formulate statistics about the draft process as a whole and the probabilities of success in certain rounds. This is simply UNTRUE and INCOMPLETE.

And by the way, I somehow REALLY doubt that a famed statitician like Richard Thaler would waste his time on something like this...and if he were to waste his time, I dont think that he would waste much of it. He probably put this together in one afternoon.

What I would like to see is a GM by GM analysis.
E.g. take your Savage's, Millen's, Heckert's, and Tanny's of the world and analyze each and every pick they have made over the course of their general management careers. If the sample set wasnt so small, you could even take a position by position analysis (i.e. what has been Heckert's success rate of drafting CBs or LBs...) Once you have your personalized ratings you can then make a statement such as...Tannenbaum, who has a 75% success rate of drafting starters in the first round, 25% in the second round, 15% in the third round, 100% in the fourth round and so on..made a good trade because he played HIS OWN probabilities well by trading out of unsuccesful rounds.

Obviously the above numbers are made up, but I am trying to illustrate a point. But by no means am I advocating that this kind of analysis should be placed on the NFL draft...because there are simply too many variables. Not all draft classes are equal. Some years are very top heavy in terms of talent, while others dont have many studs at the top, but have a lot of value all throughout. Some years are very strong at a particular position, while others are not. Additional some years there is far less demand for certain positions than in other years allowing more talented players to slip later into the draft.

I can keep going, but it should be pretty clear that Mike Salfino is wrong. He strikes me as somebody who is going against the grain simply to garner attention, and people like that really roil me up.

[Smedsthejet]

Whilst I think we paid a heavy price to move up and could have done with a couple more picks to add depth, I like the fact Tannenbaum and Mangini went out and got the players they wanted rather wait and hope they might fall.

[Pennington]

Furthermore, Id like to add that Salvino is implying that these moves indicate a pattern of behavior. Lets make something clear..up until now, there is clearly no pattern. In fact the Tangini duo's decisions have been a little chaotic...meaning hard to predict. This is not a negative statement, in fact its a very good thing. Just because they traded up this year does not preclude them from trading down next year and stock piling picks in a frenzy.

[kelticwizard]

Just because they traded up this year does not preclude them from trading down next year and stock piling picks in a frenzy.

Absolutely. Just because they did it this way this year does not mean that they will do it that way next year.

IF they do it year after year, then Salfino might have a point. But if they only do it one year, based on the situation that one year, then he does not.

[Swissle]

Absolutely. Just because they did it this way this year does not mean that they will do it that way next year.

IF they do it year after year, then Salfino might have a point. But if they only do it one year, based on the situation that one year, then he does not.

This dude is just trying to stir the pot a little bit, which to me, is retarded to do just for the hell of it. Which is obviously the case. A lot of writers do this to try and go against the grain and make some bold ass statements praying that they unfold that way, just so they could say " I told you so".

I believe the Jets will have the last laugh. Revis and Harris will be solid. A tie for defensive rookie of the year? I SAY YES!

but only to say I told you so.

[Pennington]

I suspect that Harris will have a better year, simply because of the scrutiny of the CB position. But hey, you never know.

[jets0n]

Who the hell is Michael Salfino?

[Kirby JUICE]

The simple fact that this guy uses Pro-bowl selections as a meter for success makes him lose credibility in my eyes. He condemns the Harris move because 3-4 ILBs rarely make the Pro-bowl as a result of sharing tackles? What a joke!! The guy wants to gauge players based on FAN VOTING? I was much happier with Kerry Rhodes' play last year than John Lynch's and I know you all will agree with me on that.

[SecondHandJets]

I'm a huge sports fan and I'm also an MS economics student so I've read a shit load of studies that use statistics to try and predict sports behavior. Some make sense, for instance the MoneyBall/Billy Beane logic of using OBP. as opposed to AVG. The problem comes in when you take theoretical economics and try to apply them to real world situations. In economics, we operate under ceteris perebis when we solve problems and look for trends. However, in the real world that is never the case. Take for instance the market for DB. With Clemens getting $10 million, it is clear that the most efficient way to get a franchise DB is through the draft. Keeping in mind that the draft is a crap shoot in the sense that no one knows who will be a sure thing and who will be in the trainers room teams rely on the information available to make the best reasonable assumption. This is your classic case of applied game theory and if the author of the article bothered to look at it that way, he'd soon see that the Jets were right for moving up. In game theory, you are expected to move not only on what your need is but also on what you determine your opponents (other teams) needs are. The Jets anticipated a run on defensive players in the middle of the first and moved up to get the guy who they reasonably believe wil be the best DB in this year's crop.

Also, across the board studies like these are useful if you're determining long term trends. They completely ignore individual team situations. For instance, according to this study, the Jets should try to draft 10 player per year. But in reality, the Jets have 22 set starters, plus some key backups and special team players. They face a constraint in the form of a mandated roster and bringing in 10 fresh faces when you have no room for them on your team is highly inefficient. Again, these studies are very interesting and some of them give key insights on a number of factors, however they are not in any means a blue print for successful roster building.