Amateur drafts--median values

[7Mo]

Is there a source that evaluates the median value for a team from an amateur draft? 

[Frobby]

13 hours ago, 7Mo said:

Is there a source that evaluates the median value for a team from an amateur draft? 

Do you mean, median WAR/team from the draft?    I haven’t seen that exact calculation, but here’s an article that breaks things down a bit:

https://community.fangraphs.com/analyzing-the-draft/

Finger to the wind, I’d say a median team draft produces about 20 rWAR.   I think the median may be considerably below the mean, and so an endless string of median-level drafts would not be a good result.

I get there three different ways.   First, the article shows that the available WAR in a draft is roughly 600 on average.  600/30 = 20.

Second, I did a study of the AL East drafts a while back and found that the drafts from 2000-09 had produced 1100 WAR.   1100 for 5 teams over 10 years = 22 WAR per year.   That number can rise a bit since some of those players are still playing, though most aren’t.   

Third, we know that WAR is calibrated so that an average team will have 34 WAR in a year.   70-75% comes from drafted players, the rest from foreign players.  So figure 26 WAR/year from drafted players.   Logically, the draft should be supplying about that much talent every year for the system to be self-sustaining.  

The reason I think the mean is probably significantly higher than the median is that one star player might produce enough WAR to cover 2-3 drafts.   

FWIW, the Orioles’ 2000-2009 drafts had produced 153 WAR going into this season.    Clearly a below average total, which is not surprising.   

[Just Regular]

On 8/2/2021 at 8:30 AM, Frobby said:

I get there three different ways.   First, the article shows that the available WAR in a draft is roughly 600 on average.  600/30 = 20.

Okay, so this makes me wonder if evaluating the Domestic/Foreign split to retroactively assess the Orioles neglect over the long haul is maybe like this.   

https://www.beyondtheboxscore.com/2018/12/26/18155292/correlation-war-wins-pythagorean-expectation-second-order-wins-third-order-wins

This piece guesses 48 wins for the imaginary replacement level team of Freddy Galvis, Shawn Armstrong, Pat Valaika, Matt Harvey and their closest friends.   48 to 81 is 33 wins - I believe that would be appropriate if we are looking at closed environment of 81 * 30 wins and losses each per team per year on average.

So if Rule 4 draft players give ~20 out of 33, it leaves ~13 out of 33 for international free agents.   60/40?

The good Orioles teams of Manny Machado and Adley Rutschman lacked/will lack anything from that ~40% pie slice while trying to beat Jasson, Vlad, Wander and Xander these coming years, until poster children like Maikol Hernandez or Samuel Basallo bubble up barring any pleasant surprises.   I would sign up for Ramon Urias being Miguel Gonzalez!

But they do at least have Elias/Sig, and maybe a couple high school pitchers Decision Sciences would probably say weren't the best choices.

I imagine Mike Elias articulated all this to the sons much more suavely in his interview.    We debate about Mancini after 2022 a good deal, but unless Elias' original contract term was five or more years, next year might be his walk season.   I know he was a well-regarded GM prospect but guessing he didn't get a 10-year Gunnar Henderson special or anything like that.

[Frobby]

It’s more like 70/30 than 60/40, but that doesn’t change your point.  Ignoring the international market for so long certainly hurt the Orioles.