What constitutes a good draft? My study of 1998-2005

[Sports Guy]

Where did you find the WAR numbers?

[Can_of_corn]

Will be interesting to see if the changes to the draft cause a shift in where the value is coming from.  I'd think that the slotting would discourage big overslots in later rounds and lower the overall value of picks taken later in the draft.

[Frobby]

11 minutes ago, Sports Guy said:

Where did you find the WAR numbers?

It’s all in BB-ref’s draft database, found here: https://www.baseball-reference.com/draft/?draft_type=junreg&query_type=overall_pick&from_type_jc=0&from_type_hs=0&from_type_4y=0&from_type_unk=0.  You can sort by year, by team, by round, etc.   But I had to do a lot by hand, like removing players who were drafted but didn’t sign.   It was time consuming, probably took me 4-5 hours in total to put that spreadsheet together.   

[Sports Guy]

It’s interesting because whatever the total WAR was for 2001, 150ish of that came from 3 players…Mauer, Tex and David Wright.  Another 15 from Gavin Floyd.

I wonder how normal that is during the years?   I suspect that a few players really end up skewing the numbers each year.  
 

 

[Frobby]

11 minutes ago, Can_of_corn said:

Will be interesting to see if the changes to the draft cause a shift in where the value is coming from.  I'd think that the slotting would discourage big overslots in later rounds and lower the overall value of picks taken later in the draft.

There have been a lot of draft rule changes since the period I studied, plus evaluation techniques have gotten more sophisticated.   So, I definitely can’t say that the patterns from 1998-2005 would hold true today.  But in terms of what is a good draft and the frequencies of obtaining a certain amount of WAR in the draft, I’d think they’d stay pretty constant from one 8 year period to another, obviously with a good bit of variation from any single year to another.  

[Sports Guy]

32 minutes ago, Can_of_corn said:

Will be interesting to see if the changes to the draft cause a shift in where the value is coming from.  I'd think that the slotting would discourage big overslots in later rounds and lower the overall value of picks taken later in the draft.

And how much does it change with the draft essentially cut in half?  I would guess very little change since so few players come from outside the top 20ish rounds and a lot of them don’t even get signed.

[Frobby]

Just now, Sports Guy said:

It’s interesting because whatever the total WAR was for 2001, 150ish of that came from 3 players…Mauer, Tex and David Wright.  Another 15 from Gavin Floyd.

I wonder how normal that is during the years?   I suspect that a few players really end up skewing the numbers each year.  

It varies a bit.   Pujols was worth 99.5 rWAR by himself in 1999, which is why it produced more total rWAR (831) than any other year I studied.   700 is about average, 550 is a really bad year.   

At some point I’m going to try to list all the players over a certain threshold in that period.   Maybe 30 rWAR.   I’d guess the average draft has 10-15 players above that and they make up more than half the WAR in that class.

On the other hand, the year we drafted Nick Markakis (33.5 rWAR) he was the highest value player taken in the entire first round (Adam Jones was second at 32.6).   In fact, only one player drafted and signed in 2003 was more valuable than Nick — Ian Kinsler (54.6 rWAR), who was selected in the 17th round!   It was a really weak draft year, and the O’s made a fantastic choice considering how little talent was available.   Kinsler, Markakis and Jones were the only 30+ rWAR players drafted that year.  
 

[Frobby]

10 minutes ago, Sports Guy said:

And how much does it change with the draft essentially cut in half?  I would guess very little change since so few players come from outside the top 20ish rounds and a lot of them don’t even get signed.

I guess you’ll find some undrafted free agents who produce some value, but they’ll be few and far between.  

[foxfield]

18 hours ago, Aristotelian said:

I thought they were supposed to be better at player evaluation because Yaz. I'm so confused.

Well, this is pre Preller...

[Frobby]

9 minutes ago, foxfield said:

Well, this is pre Preller...

What’s Preller got to do with the Giants? Your general point is correct, however.  Brian Sabean was the Giants GM during 1998-2005, and Farhad Zahidi is in that role (but not with that title) now (and when they traded for Yaz).   

[foxfield]

5 minutes ago, Frobby said:

What’s Preller got to do with the Giants? Your general point is correct, however.  Brian Sabean was the Giants GM during 1998-2005, and Farhad Zahidi is in that role (but not with that title) now (and when they traded for Yaz).   

Sorry, my bad...totally lost my mind there....I knew the study predated the current leaders, but the conversation of SF and SD earlier this week got me crossed.  Thanks for pointing it out and sorry for the confusion.

 

[GuidoSarducci]

The data I posted for the draft years 1990-1999 was wrong.  Danged spreadsheets...

This should be correct data

	Avg rWAR
1	28.39
2	13.31
3	9.88
4	7.77
5	7.03
6	6.25
7	4.36
8	8.37
9	9.43
10	10.65
11	1.55
12	12.69
13	10.23
14	8.97
15	3.16
16	9.56
17	8.62
18	2.05
19	5.18
20	22.28
21	4.06
22	8.25
23	5.75
24	2.58
25	0.28
26	1.01
27	-0.05
28	3.34
29	1.75
30	1.14

[Frobby]

2 minutes ago, GuidoSarducci said:

The data I posted for the draft years 1990-1999 was wrong.  Danged spreadsheets...

This should be correct data

 

	Avg rWAR
1	28.39
2	13.31
3	9.88
4	7.77
5	7.03
6	6.25
7	4.36
8	8.37
9	9.43
10	10.65
11	1.55
12	12.69
13	10.23
14	8.97
15	3.16
16	9.56
17	8.62
18	2.05
19	5.18
20	22.28
21	4.06
22	8.25
23	5.75
24	2.58
25	0.28
26	1.01
27	-0.05
28	3.34
29	1.75
30	1.14

This makes more sense. It was shocking that the no. 5 pick was in negative territory before.   Kind of funny how no. 20 was so productive that decade, with Mussina, Sabathia, Torii Hunter, Ian Kennedy and Eric Milton all popping.   The first three of those are the top three no. 20 picks in the 55+ year history of the draft, and they were all drafted in the same decade.   Fluky stuff like that happens.  

[Sports Guy]

It’s such a crapshoot.

[GuidoSarducci]

9 minutes ago, Frobby said:

This makes more sense. It was shocking that the no. 5 pick was in negative territory before.   Kind of funny how no. 20 was so productive that decade, with Mussina, Sabathia, Torii Hunter, Ian Kennedy and Eric Milton all popping.   The first three of those are the top three no. 20 picks in the 55+ year history of the draft, and they were all drafted in the same decade.   Fluky stuff like that happens.  

Well after I posted that I went backed and looked again... and was wrong again. This should be right.  Unless I'm losing my mind..

1	31.23
2	14.64
3	10.87
4	8.55
5	7.73
6	6.87
7	4.80
8	9.21
9	10.37
10	11.72
11	1.71
12	13.96
13	11.25
14	9.87
15	3.48
16	10.52
17	9.48
18	2.26
19	5.70
20	24.51
21	4.47
22	9.07
23	6.33
24	2.84
25	0.31
26	1.12
27	-0.06
28	3.67
29	1.93
30	1.27