ESPN offers up new Draft Pick value Chart

[DaBruinz]

Got the new ESPN The Magazine and they had 2 guys do up a "new draft chart". Some of their facts are sound, but there is one key factor that it doesn't take into consideration. It totally ignores the differences in the length of a contract that players can sign.

While I think that they are somewhat right in their thinking that there is less of a value difference between picks 1 and 16, and 17-32, I believe that they are wrong in that its only a meer 890 points. Its not the 2400 points of the first draft value chart, but its certainly not 890 points. Its probably closer to a 1800 point drop, in my honest opinion.


Here is how I see it working out:
http://www.geocities.com/nh_wynter/DraftValueChart.htm


This is just something I threw together in an excel worksheet. But I feel it takes into consideration the differences in the contract lengths.

But, like anything, its a guideline and not set in stone.

On other thing that I find hilarious about the article is that they totally blame Jimmy Johnson and "his flawed chart" for the wows of teams like the 49ers, Lions and Raiders.


P.S. If someone can tell me how to embed the spreadsheet into the message, I'd appreciate it.

 

[JoeSixPat]

Got the new ESPN The Magazine and they had 2 guys do up a "new draft chart". Some of their facts are sound, but there is one key factor that it doesn't take into consideration. It totally ignores the differences in the length of a contract that players can sign.

While I think that they are somewhat right in their thinking that there is less of a value difference between picks 1 and 16, and 17-32, I believe that they are wrong in that its only a meer 890 points. Its not the 2400 points of the first draft value chart, but its certainly not 890 points. Its probably closer to a 1800 point drop, in my honest opinion.


Here is how I see it working out:
http://www.geocities.com/nh_wynter/DraftValueChart.htm


This is just something I threw together in an excel worksheet. But I feel it takes into consideration the differences in the contract lengths.

But, like anything, its a guideline and not set in stone.

On other thing that I find hilarious about the article is that they totally blame Jimmy Johnson and "his flawed chart" for the wows of teams like the 49ers, Lions and Raiders.


P.S. If someone can tell me how to embed the spreadsheet into the message, I'd appreciate it.

Nice work - certainly better than anything I've seen prior (in particular Florio's on PFT)

A good barometer though is consider Dallas's picks at 22 and 28. What's the true trade value there?

Currently pairing them together is deemed slightly below the Patriots pick at #7. That's probably too low - though I'd hapilly make that trade.

Under this proposal trading 22 and 28 together will nearly get you the #1 pick.

That's probably too high - maybe not in this draft where I think the feeling is that any of the top 5 could or should be #1 - but a new chart isn't just about this draft.

I guess maybe that's the point - the draft value chart needs to have a degree of flexibility in each draft based on the consensus top 5 talent - where the risk/reward balance is greatest - and where perpetually bad teams can remain perpetually bad because the risk reward is increasingly skewed by the outrageously high salaries of the top 5.

Of course one can't solely blame Jimmy Johnson or the trade chart for the woes of bad teams that stay bad. It's their own fault for making bad picks and bad salary cap decisions.

If you ask me, it might be a fair thing to let the team with the #1 spot should get to choose where they pick. In this draft if I were the fins I might choose #3 or even #5. But hey - if they wanted to pick at #10 that'd be their call (and their risk).

In doing so the worst team isn't stuck in what some would deem the worst position - having to fork over #1 money to a guy who isn't worth it.

I suppose one could involve all the teams to allow them to select where they draft based on their cost/benefit analysis of the draft, selecting their draft position in order of how they finished.

Chances are this wouldn't result in a dramatic re-ordering of the draft -but conceivably the #32 team could find themselves "stuck" with the #1 pick in a bad draft year.

All of that would dramatically change how pre-draft trades go down, but I just throw it out as another concept.

 

[pats1]

P.S. If someone can tell me how to embed the spreadsheet into the message, I'd appreciate it.

If it's small enough to fit in your screen, highlight and copy it in excel, then paste it into MS Paint (or Photoshop, etc.). Then just save it and post it as an image -- either using the attachment feature here or imageshack.us.

 

[patchick]

FWIW, here's the version I did a year ago:

Here's how I described it at the time:

"I've multiplied the top 16 (6-year) slots by a factor weighted to their rank. OP = old point value, NP = new, RANK = draft slot (#1, #2, etc.):
NP = OP + (.015 X RANK)

For picks 17-32
NP = OP X 1.1

All old values still hold for round 2 on."


In retrospect it was too timid, I ought to take another stab. But I still think it's a little more realistic than either Florio's (which has fundamental structural flaws) or ESPN's (which bizarrely ignores the contract tiers).

 

[JoeSixPat]

FWIW, here's the version I did a year ago:

Here's how I described it at the time:

"I've multiplied the top 16 (6-year) slots by a factor weighted to their rank. OP = old point value, NP = new, RANK = draft slot (#1, #2, etc.):
NP = OP + (.015 X RANK)

For picks 17-32
NP = OP X 1.1

All old values still hold for round 2 on."


In retrospect it was too timid, I ought to take another stab. But I still think it's a little more realistic than either Florio's (which has fundamental structural flaws) or ESPN's (which bizarrely ignores the contract tiers).

Interesting approach - although for comparisons sake, under this trade value chart, Dallas would get better value trading 22 & 28 under the CURRENT chart.

The current trade chart gives Dallas 1440 points for pairing both picks putting them just about between rounds 7 (1500) and 8th (1400).

Under this proposal they have about the same ammunition - maybe a little less.

Really, the thing that needs to be reflected from the current trade value system and a new one is a factor by which the high cost of the top salaries could be reflected.

As such I'd suggest maintaining the current trade value chart - but somehow subtracting a number from each of the top picks, that would get lower with each lower pick

As such #1 would have a high value but have a higher relative "deduction" given the high cost of the #1 salary. #2 would have a high value, and a slightly smaller deduction, given the slightly lower salary.

At some point the deduction would phase out towards the middle of the first round, as those players salaries represent much better value.

The net effect would be that trading 22 & 28 should be able to buy Dallas better than the current #7.5 pick or so... maybe that gets them as high as 5.

What can't be ignored is that the ability to choose any player one wants at #1 - and nearly any player at #2 etc, still has a significant value attached to it.

The bottom line is that there are many factors to consider.

Note also the difference of 400 points per the top 4 draft picks in the current value chart.

http://www.nfldraftcountdown.com/features/valuechart.html

That drops all the way to a 100 point difference at #5 and maintains that through #8.

8-18 drop by 50 points. 19 & 20 drop by 25 - but then for some reason it jumps back to a 50 point drop at #21 and then down to 20 after that.

So the old one itself is making certain assumptions even without factoring in high salaries at the top of the draft (i.e. the drop from 20 to 21 is higher than from 21 to 22 for some reason).

 

[Metaphors]

The logic that the value chart is supposed to represent the relative ability of the player (JaMarcus Russell is 2x more valuable than AP) is flawed. Having the #1 pick last year meant you had the option of picking Russell or Joe Thomas or AP or anybody else. That option provides value beyond just the ability of the player selected at that position.

Sometimes that option provides little value for a team, like if your top team need is a CB this year. Sometimes having that option is critical if your top need position will be picked clean of quality by the time you pick (like DT for the Saints this year).

Is there a significant difference in ability between the players selected at #7 and #10? Probably not. Is there a significant different between lining up Sedrick Ellis at DT for New Orleans and their other options at DT (draft and otherwise)? Almost certainly. Enough to part with a 3rd round pick? That is for them to decide.

A static value chart can't take these factors into account. They are different from year to year and from team to team. I find the existing value chart to be fairly useful (with adjustments for contract length benefits). Sometimes it makes sense for a team, sometimes it doesn't.

Newer iterations of the value chart seem to underestimate the value of getting to the top of the draft for truly special talent like a Reggie White or Elway or PManning or more recently a Joe Thomas or AP. I say tweak the existing chart but don't throw the baby out with the bath water.

 

[Metaphors]

What can't be ignored is that the ability to choose any player one wants at #1 - and nearly any player at #2 etc, still has a significant value attached to it.

Sorry, I didn't read this part of your post before posting myself. My post repeated some of the main parts of yours.

 

[DaBruinz]

I took JSP's comments into consideration and tweaked my chart a bit.

 

[rookBoston]

The value chart needs to change, year to year.

In some years, there are two or three legitimate stand-out players who could go #1 overall. And there are some years where the #1 overall pick is Alex Smith because there was literally no one to pick, and things only get worse after that.

In 2004, the top of the draft was so strong, you have to read down into the teens before you find a player who doesn't belong in the NFL. Wilfork went #21 that year, and its hard to criticize too many teams who passed him up. The team that drafts Glenn Dorsey in the top-5 this year can only hope and pray that Dorsey becomes what Wilfork has turned into.

In contrast, 2005 was so bad, there is only one pro-bowler from among the top-10 (Braylon Edwards) That's the year Pacman Jones went #6 overall. Whoever the Jets draft #6 this year is CERTAIN to be an improvement over Jones... with the possible exception of McFadden, who could flame out, too.

This year, where there are four or five players who could legitimately go #1 overall (Long, Long, Ryan, Gholston, Dorsey), the #1 pick should be worth relatively less than years where there is a consensus #1, like 2003 where Carson Palmer was a no-brainer. The value chart should dropoff at #6 and #7, when the available players will be in a different echelon of talent. The value of the #6 or #7 pick comes in situations where one of the blue chippers drop unexpectedly... speculative value, more than anything. And then the #9 and #10 picks should be far lower value, because the chance that McFadden or Ryan drops to them is almost zero. At #10, they'll be looking down at the top cut of the next cohort of talent (Mendenhall, Rivers, McKelvin, etc.) instead of looking up at the top cohort.

This year I think St. Louis and Oakland have the most valuable picks in the draft, at #2 and #3. They will get #1 overall talent at terrific value. Contrast that with 1998, when Indianapolis had the choice of Manning or Leaf, and San Diego was left to take "the other guy" at #2. There was no talk, that year, about Indy looking to trade down.

 

[JoeSixPat]

I took JSP's comments into consideration and tweaked my chart a bit.

Thanks - of course in reality I think your original probably was pretty close to the true value THIS year. And maybe that's not a bad idea... i.e. to tweak the draft value chart each year, taking into account a year like this when there's no consensus 1,2, or 3.

But now if my math is right, trading 22 and 28 together only gives a team 2210 points - not quite enough to move them up to #9.

That's now lower than it is under the current value chart - and I think Dallas' point about needing to discount the top picks given the high salary is valid.

Typical Monday Morning QB I know because you and Patschick are to be commended for giving this a go - but I my gut says that by trading into the higher salaries Dallas' picks should give them enough to get up to say, #5, regardless of whether that's achieved by weighing the lower picks more or discounting the higher picks (which would be my preferred way)

Overall I think there can be one "high salary" discount to the value of picks 1-4, another tier of discounts for 5-8, and then phase it out beginning at #9 - all based on the current trade value chart.

After that the salaries are much less outrageous and I think the current trade value chart works well.

And should the NFL ever get the rookie salaries more under control, or adjust them, the "discount factor" can be adjusted accordingly.

 

[JoeSixPat]

OK - decided to put my money where my mouth is and see if my concept might be workable.

Here's what I propose (too lazy to work it into a grid though)

The high salary of the very top pick is what lowers the value of that top pick - yet the top pick no doubt retains value as, by definition, the top pick.

So I've worked a "discount factor" into the draft trade chart.

The discount factor is a multiplyer, subtracted from the current trade chart value.
http://www.nfldraftcountdown.com/features/valuechart.html

The top pick is reduced by 10% (i.e. multiply by .10, and subtract that from the 3000 - giving you a value of 2700)

After the top pick the multipliers begin to go down, eventually weaning out by pick 13, when the salaries somewhat normalize.

Picks #2-4 are discounted by 8%

Picks 5-6 by 6%

Picks 7-8 by 5%

Pick 9 by 4%
Pick 10 by 3%
Pick 11 by 2%
Pick 12 by 1%

Pick 13 reverts to the existing trade chart as salaries don't appear to be so outrageously inflated there and further back in the draft.


So in terms of actual trade value - here's what I get for the top 13 picks (assuming my math is right:

1. 2700
2.2392
3. 2024
4. 1656
5. 1598
6. 1504
7. 1425
8. 1330
9. 1296
10. 1261
11. 1225
12. 1188
13. 1150 (which is the existing trade value chart)

The high salary discount factors can be played around with a bit, but you get the concept. This retains the value for the higher picks, discounts heavier the higher in the draft, and eventually weans out at 13.

Looking at the Dallas picks for example, where they currently don't quite make it to being equal to #7 (1440 total - to #15's 1500 points) under the "High Salary Discount Trade Chart" they have just OVER what it would take to get to #7 and begin to appropach #6.

Tweak this how you like - I think a lot depends on how over-inflated you deem rookie salaries to be. Personally I think its mainly the top 4 picks that are most onerous - after that, value comes back into play and the salaries are high, but not outrageous (top picks have ALWAYS been relatively expensive anyways).

 

[DaBruinz]

bumpity bump bump

 

[spacecrime]

Newer iterations of the value chart seem to underestimate the value of getting to the top of the draft for truly special talent like a Reggie White

Reggie "homophobe" White wasn't a first pick in the draft