The tale of the tape

[detlef]

In my thread about the Niners trading their pick to NE, I eluded to the fact that I think the current state of the unstructured rookie salaries is going to hurt teams picking near the top.

 

That, trading down for value might begin to mean more than just getting more picks, it might begin to mean not having to lay out so much cash for a guy who's not that much more likely to be a stud than a guy you can get later.

 

So, I went back through and checked 4 drafts from 2000-2003 to see how the picks fared. I was going to do more but the numbers seemed to be consistent enough that no further analysis was needed.

 

I established a scale of 0-4. Zero being somebody who never did squat, One being a guy who never spent any significant time as a starter, two being a part time starter, three being a consistent starter (after perhaps a year or two), and four being among the best at his position.

 

Unless the player was high profile, I pretty much just went by starts (since everyone's player page showed games played and started). In certain situations, I gave players the benefit of the doubt if they had major injuries or penalized them if they had massive off the field problems (like Vick (from 3 to 2) and Koren Robinson (from 2 to 1)).

 

If anyone cares, I'll post the highlights and lowlights of each section of each round. Of course, we could debate back and forth whether some guys are 2s or 3s, etc. However, I think it was pretty fair.

 

None the less, here are the results. The number posted is the average rank of players taken in that section of the round (broken into 1/3s):

2000 1st 2.3, 2nd 2.7, 3rd 1.4

2001 1st 1.9, 2nd 2.6, 3rd 2.1

2002 1st 2.0, 2nd 2.3, 3rd 2.4

2003 1st 2.4, 2nd 1.9, 3rd 2.7

 

Overall, the 1st and 3rd portions of the 1st round averaged out to be a dead heat (2.15 and 2.147 respectively) and the middle of the round was strongest at 2.375.

 

So, essentially, there is no reason to believe that players taken at the top of the round are going to be much better and certainly not worth what you're going to have to lay out to get them.

 

If nothing else, the numbers were all close enough that, even if I ranked a player here or there high or low, it was not going to have a massive effect. After all, to justify the massive difference in contracts that the top players get, they'd have to be much more likely to be studs.

[muck]

The incremental $$s paid, plus the incremental media attention, given to a very top pick increases the likelihood (i) that they don't pan out, and (ii) that they never really have to do anything else the rest of their life (excepting to avoid spending their $$ on silly things).

[detlef]

Keep in mind, I didn't hold higher picks to higher standards. If they only turned out so-so, they got as high a rating as if the 25th pick turned out so-so.

[detlef]

Just curious. Does nobody find this info surprising? Interesting? Illustrative?

 

I ask this because when I posted my bit about the Niners/NE trade, there were plenty of comments to the effect that "The best way to get great players is through the draft and the best place to do that is at the top of the 1st round", etc.

 

Thus, it seems people were implying that the mad jack that you have to lay out for these high picks was worth it. This would certainly illustrate that is simply not true. Perhaps a larger sample would show that the top of the round is at least better than later in the round but to justify the vast difference between what a top 10 and #25 pick are getting, you'd think it would need to be huge.

 

Considering that you can only spend so much money and you're going to need it to either attract free agents or resign any of your own players that you want to keep, why would you want to be put into a position where you're paying 5x more for a player when there's no concrete data to support why they should be much better? As it stands, that is exactly where the teams picking at the top of the draft are.

[Randall]

That, trading down for value might begin to mean more than just getting more picks, it might begin to mean not having to lay out so much cash for a guy who's not that much more likely to be a stud than a guy you can get later.

 

 

True for Ted Thompson. He is determined to stay out of the cap mess that hurt San Francisco and others for years. He looks for good but undervalued players that will do well. He often hits and misses but sticks with the plan. With that strategy busts are rare.

[detlef]

True for Ted Thompson. He is determined to stay out of the cap mess that hurt San Francisco and others for years. He looks for good but undervalued players that will do well. He often hits and misses but sticks with the plan. With that strategy busts are rare.

Yep. The funny thing is, guys like this always get screamed at by the fans for trading back and passing up on the "can't miss studs". Hell, look at how much crap the Philly fans were giving their team here as they kept trading back.

 

The interesting thing would be to see what the second round stats are.

[Savage Beatings]

This makses me wonder if there is a similar correlation with Dynasty rookie drafts. I would think so.

[Randall]

Yep. The funny thing is, guys like this always get screamed at by the fans for trading back and passing up on the "can't miss studs". Hell, look at how much crap the Philly fans were giving their team here as they kept trading back.

 

The interesting thing would be to see what the second round stats are.

 

 

 

Thompson was criticized for trading down a few years ago in the 2nd. He passed on Chad Jackson who was highly regarded after the combine. He traded that pick(the Javon Walker trade) for what turned out to be 5 players. Then he picked Greg Jennings later in the 2nd with his own pick( a much better player). He was criticized for taking him too early. This year Jordy Nelson is the guy he took too early.

 

This year he traded down 6 times. Too many probably but he thinks several years ahead.

[Easy n Dirty]

Very interesting analysis, certainly seems to support the notion that in a salary cap world, high first round picks are overrated and may even be detrimental.

 

One other facet maybe worth thinking about - on your scale of 1 to 4, are you perhaps more likely to get someone rated a a 5 (superstud, think Peyton Manning) at the top of the first round? Off the top of my head I'm thinking maybe not - but I do think the analysis would be even more valid if you rated the players into more individual strata (maybe 1 to 10 instead of 1 to 4). No way of taking the subjectivity out of it, but if I understand the way you rated players, there's not enough of a spread between say a Peyton Manning versus a Roethlisberger or an Eli Manning.

[muck]

Right, EnD.

 

Also, there is a difference between, say, LT who was productive the first week he was on the field and Larry Johnson who rode the pine for a couple of years before playing ...

 

Interesting analysis, however, like EnD noted, you could take it a bit farther to see if the stratification is amplified any. More granularity, please, sir.

[muck]

Another thing to note is the record of the team the players went to in the previous year (i.e., the year before they were drafted) and their first year in the league.

 

Why? Occasionally, a very good team is picking high (see NE this year), and I would bet that a player that goes to a good team is more likely to play at a high level than a player that goes to a lesser team.

 

Another thing that could explain why the middle part of the 3rd round is a high performing area is that these are very good players who slipped out of the first round, and a playoff team from the previous year moved up a couple of spots to pick them...

[detlef]

Very interesting analysis, certainly seems to support the notion that in a salary cap world, high first round picks are overrated and may even be detrimental.

 

One other facet maybe worth thinking about - on your scale of 1 to 4, are you perhaps more likely to get someone rated a a 5 (superstud, think Peyton Manning) at the top of the first round? Off the top of my head I'm thinking maybe not - but I do think the analysis would be even more valid if you rated the players into more individual strata (maybe 1 to 10 instead of 1 to 4). No way of taking the subjectivity out of it, but if I understand the way you rated players, there's not enough of a spread between say a Peyton Manning versus a Roethlisberger or an Eli Manning.

 

 

Right, EnD.

 

Also, there is a difference between, say, LT who was productive the first week he was on the field and Larry Johnson who rode the pine for a couple of years before playing ...

 

Interesting analysis, however, like EnD noted, you could take it a bit farther to see if the stratification is amplified any. More granularity, please, sir.

Good points to be sure. Unfortunately, my work is at home so I can 100% confirm what I'm going to say here. There's certainly a spread within each level and likely more so among those that I gave 4s to. I mean, I gave both LT and Nick Barnett 4s and I doubt there's any real debate that LT is a bigger stud. However, overall, I didn't give that many 4s out (or zeros for that matter). Certainly 2s dominated the list followed by 1s and 3s. So, of the 120+ players in the study, maybe 10 or so were 4s. Let's say 5 of those 4s were all in the top 3rd and all of those 4s would really be 5s. That increases the average rating of that section from 2.15 to 2.275 which still doesn't even make it the most fruitful section and certainly not enough better than the end of the round to support the massive difference in money you're going to have to lay out. Of course, it also bears mention that this is football where you need 22 guys to start and plenty more to contribute a lot. Unlike hoops, one stud is not going to make a huge difference. Take LT. There's really no question that he is far and away the best RB in the game for the last several years. That has hardly delivered SD to the promised land.

 

Keep in mind, I'm not suggesting teams adopt a football version of moneyball, though that's not entirely far off from what seems to be working very, very well for NE. I don't think you can succeed without opening up the checkbook. I'm just saying that being forced to do so on a regular basis in the draft may not be a very good way to improve.

 

Of course, as muck suggests, there's a number of other elements at play here that support why players selected at the top have a harder row to hoe. Who knows, maybe if David Carr hadn't spent his first few years getting pummeled behind a crappy line, he wouldn't have turned out so poorly. Maybe if he toted a clipboard around behind a solid QB on a good team. Who really knows. It does bear mention that one of the few QBs taken at the top in the years I studied who turned out good, Carson Palmer, was not thrown to the wolves right away.

 

I suppose I figured that these outside factors would work themselves out and that the end of the round really only needed to be close in terms of turning out productive players to substantiate the argument. Thus, even if all the extraneous factors all piled on to help the case of the top 10, the fact that the very raw data makes it tied with the end means it would not likely pull far enough ahead.

Edited April 29, 2008 by detlef