In your opinion, what is fair value for Orlando Brown Jr?

[L. G. Dupre]

I think that Harvard Trade Value chart more closely fits how GMs will make decisions than the old JJ chart and that a package worth worth between 270 and 300 points would be fair value That’s the Harvard chart equivalent of a pick between 10 and 14, but that’s not the deal I’d want.

I’d prefer a package of OBJ and one of our 5ths for two picks in that with 270 and 300 points net value. I’d then use one of our two high picks to replace OBJ and treat the lower pick as BPA vig. The 24th picks in first and third rounds would be worth 330 Harvard points. Our top 5th is worth about 60, so that purely hypothetical package would be at the low end of my fair net value range.

https://www.bloggingtheboys.com/page...de-value-chart


Sent from my iPad using Tapatalk Pro

[WNCRavensFan]

I think that Harvard Trade Value chart more closely fits how GMs will make decisions than the old JJ chart and that a package worth worth between 270 and 300 points would be fair value That’s the Harvard chart equivalent of a pick between 10 and 14, but that’s not the deal I’d want.

I’d prefer a package of OBJ and one of our 5ths for two picks in that with 270 and 300 points net value. I’d then use one of our two high picks to replace OBJ and treat the lower pick as BPA vig. The 24th picks in first and third rounds would be worth 330 Harvard points. Our top 5th is worth about 60, so that purely hypothetical package would be at the low end of my fair net value range.

https://www.bloggingtheboys.com/page...de-value-chart


Sent from my iPad using Tapatalk Pro

Uh-oh... it uses Career AV to assign value. There are definitive flaws in Career AV. But it's still likely better than the old chart.

The most striking thing is the curve is much less extreme than the JJ Chart. Pick 179 has one tenth the value of Pick 1. On the JJ Chart, it's pick 60.

[BcRaven]

Agreed. I will take it one step further though. It would have to be a top 20 pick in the first round for me. For example, I wouldn’t trade him to the Chiefs for their late first round pick. OBJ would have more value on the team this season than such a late first round pick would bring.

It's not gonna happen, but if the Jaguars were to offer one of their 2nd round picks (they have #33 and #45), plus their 3rd (#65), I'd be satisfied... Bc

[JimZipCode]

Uh-oh – it uses Career AV to assign value. There are definitive flaws in Career AV.

What are the definitive flaws in Career AV?

I'm a huge admirer of PFR's AV stat. It attempts an absolutely impossible task, and does a decent job of it.

The most striking thing is the curve is much less extreme than the JJ Chart. Pick 179 has one tenth the value of Pick 1. On the JJ Chart, it's pick 60.

That makes more sense. (I mean the Harvard chart.)



I just downloaded the chart from that link to a spreadsheet, and took a look at the graph. There's a weird discontinuity between picks 32 & 33 – the value drops 25 points, from 200 down to 175. Clearly that's not in the Career AV data. Ya think maybe they're assigning a points value to the 5th-year option for first-rounders?

[WNCRavensFan]

What are the definitive flaws in Career AV?

I'm a huge admirer of PFR's AV stat. It attempts an absolutely impossible task, and does a decent job of it.

I went over it a little bit in a podcast on Filmstudy. AV is about the best we've got as an all-encompassing stat. However it penalizes people that excel in roles and sub packages. It rewards games started, snap counts, base stats and unit success. Also it doesn't count special teams.

This is how Patrick Queen scored an AV of 10 (ten) last season, while Derek Wolfe scored 5 (five) and Tyus Bowser scored 3 (three).

It's also why Anthony Levine has a lower career AV than Matt Elam.

[L. G. Dupre]

Uh-oh... it uses Career AV to assign value. There are definitive flaws in Career AV. But it's still likely better than the old chart.

The most striking thing is the curve is much less extreme than the JJ Chart. Pick 179 has one tenth the value of Pick 1. On the JJ Chart, it's pick 60.

I haven’t read the paper, but based on the term and 50 years as an actuary, I can picture what they did. Calculate the Career AVs for players coming out of each draft slot. Calculate the average AV for each slot. Fit a curve to those averages et voila! That would explain why the curve is much flatter. First round draft choices are sometimes JaMarcus Russell and sixth round choices are sometimes Antonio Brown. Career AV undoubtedly has flaws as a true measure of relative values, but the process does provide a measure of objectivity I’m sure is lacking in the JJ chart. It would be interesting to know about the tables used by an analytically driven team.

Is it a coincidence that it is we two Ashevillians geeking out on this stuff?


Sent from my iPad using Tapatalk Pro

[JimZipCode]

Is it a coincidence that it is we two Ashevillians geeking out on this stuff?

Hey, I geek out over using AV to drive a draft value chart, too.

AV is about the best we've got as an all-encompassing stat. However it penalizes people that excel in roles and sub packages. It [recognizes] games started, snap counts, base stats and unit success. Also it doesn't count special teams.

This is how Patrick Queen scored an AV of 10 (ten) last season, while Derek Wolfe scored 5 (five) and Tyus Bowser scored 3 (three).

It's also why Anthony Levine has a lower career AV than Matt Elam.

That's fair.

AV is definitely not the tool for distinguishing between guys who – all due respect to Co-Cap – are fringe contributors.

[L. G. Dupre]

Hey, I geek out over using AV to drive a draft value chart, too.



That's fair.

AV is definitely not the tool for distinguishing between guys who – all due respect to Co-Cap – are fringe contributors.

Didn’t mean to slight you. Just found it interesting that two guys In this neck of the woods were geeking out.

I’m new to this. Do you have any thoughts about something better than AV for building a draft value chart? All I know is that the basis of the Harvard chart seems much more credible than JJ’s butthole.


Sent from my iPad using Tapatalk Pro

[jonboy79]

Uh-oh... it uses Career AV to assign value. There are definitive flaws in Career AV. But it's still likely better than the old chart.

The most striking thing is the curve is much less extreme than the JJ Chart. Pick 179 has one tenth the value of Pick 1. On the JJ Chart, it's pick 60.

In reading his post I could easily see that it clearly over-value the 5th round pick.

The JJ chart seems much closer to accurate to me.

Just look at some of these trades at the top like SF/MIA/ and PHI made... Pick 1 should be worth 100 times pick 179, not just 10... JJ has it at 150x’s as much, that makes some sense... 10? Lunacy... would you really give up Trevor Lawrence for 10 6th round picks? Completely absurd and wipes out any hope at usefulness.


Sent from my iPhone using Tapatalk

[Ortizer]

Hey, I geek out over using AV to drive a draft value chart, too.

Guess it's time for you to head down to NC then.

Sent from my LM-G820 using Tapatalk

[Ortizer]

Didn’t mean to slight you. Just found it interesting that two guys In this neck of the woods were geeking out.

I’m new to this. Do you have any thoughts about something better than AV for building a draft value chart? All I know is that the basis of the Harvard chart seems much more credible than JJ’s butthole.


Sent from my iPad using Tapatalk Pro

Might be able to work something out with sufficient data using career length and contract earnings after the rookie contract relative to position.

Sent from my LM-G820 using Tapatalk

[WNCRavensFan]

Is it a coincidence that it is we two Ashevillians geeking out on this stuff?

Wait, you're in Asheville? We need to grab a beer sometime