Overall Draft Picks and Cumulative WAR: Is it really better to draft early?

by Aranathor on Jan 11, 2011 3:44 PM EST

I was inspired to write this fanpost after Kris Benson (former 1st overall pick extraordinaire) announced his retirement. Despite being drafted so early his career was less than stellar for one reason or another. This is a study (using Baseball References handy Draft section) to determine how likely it is that a player will succeed in the Majors based on his draft position.

I know this is hardly the greatest study in the world, but its January, what else is there to do?

Firstly i will look at the cumulative WAR of every pick from 1st overall to 33rd overall (the Mets had the 33rd pick in 2008 and it was still in the 1st round), ever in the June Amateur Draft.

Click for bigger pic. cause SBN f'n sucks.

Now I will look at picks of particular interest, starting with the first overall pick;

Drafting 1st overall is undeniably awesome, of the 46 players drafted 1st overall, 41 made it to the Major Leagues in some form or another. Of the five that didn’t, two are still prospects (Tim Beckham and Bryce Harper), one could still make it (Matt Bush, 1st pick in 2004 is currently in the Rays organisation and transiting to pitching) and two never made it. This is an unbelievably high success rate, having the 1st overall pick practically guarantees having a player who will contribute at the highest level.

Notable Players drafted 1st overall: Alex Rodriguez, Chipper Jones, Ken Griffey Jr., Joe Mauer

There is an unusual dearth of success in players drafted 5th overall, usually players who fail to live up to expectations because of injury or ineffectiveness. Still plenty of useful players here, but unlike picks 1-4, very few stars, especially stars with a long shelf life.

Notable Players drafted 5th overall: Dwight Gooden, Vernon Wells, Mark Teixeira

The talent explosion in pick six is all down to Barry Bonds and his 171.8 WAR, HFS. Remove that and you get 295.4 WAR which is more in line with expectations.

Notable players drafted 6th: Barry Bonds, Derek Jeter, Andy Van Slyke

Pick 10 has a lot of underrated players which makes for a very high WAR total. The top 10 players, by WAR, taken as the 10th overall pick have only two major awards between them; both of which are Lincecum’s Cy Young Awards.

Notable Players Drafted 10th: Mark McGwire, Robin Ventura, Ted Simmons, Tim Lincecum

I think the less said about the 11th pick the better, 2nd lowest cumulative WAR of all the picks in this study and just a bad time to draft anybody. Gotta feel sorry for the Astros who will have the 11th pick in 2011, that has to be a double jinx right?

Notable Players drafted 11th: No-one.

The 30th pick is crazy for one reason: Mike Schmidt and his 108 WAR. There is something mind blowing about players who amass more WAR in their careers than some teams manage in a season. Also; all seven left handed pitchers drafted 30th overall played in the majors at some point and together had 109.5 WAR. This is also the last pick, inside the top 45 or so, to have 200+ cumulative WAR.

Notable players drafted 30th: Mike Schmidt, David Wells

This proves that it is very difficult to guess where you'll get value in the draft it really is just one big crapshoot. For every Joe Mauer there is a Danny Goodwin (the only player to drafted 1st overall twice) and for every Derek Jeter there is a Johnnie LeMaster.

Tweet Comment 19 comments | 5 recs |

Do you like this story?

Comments

Bigger picture of graph

i for one welcome our new fabulous overlords - gdm

by Aranathor on Jan 11, 2011 3:59 PM EST reply actions

Wow...

that 11th overall pick was just absolutely the death pick for almost 40 years.

Greg Luzinski, Shane Mack, Walt Weiss, and Shawn Estes are the cream of the crop.

this may have finally stopped though.

11th overall picks since 2004:

Neil Walker, Andrew McCutcheon, Max Scherzer, Phillippe Aumont, Justin Smoak, Tyler Matzek, Deck McGuire

by The Ghost of Todd Burns on Jan 11, 2011 4:07 PM EST reply actions

I think you should look at WAR per PA

Cumulative WAR will be biased towards higher picks as they will receive more playing time than lower picks.

Good work though.

Skip Schumaker is a scapegoat

by vivaelpujols on Jan 11, 2011 4:12 PM EST reply actions

Well, they receive more playing time because they're good, not because they're higher picks.

I don’t see the need for this tweak. However, one that may be useful is to throw out the highest and lowest WAR contributor (taking the WAR contribution of someone who never played in the bigs as zero). That gets rid of the Bonds/Schmidt anomalies while keeping it an apples-to-apples comparison.

For grins, the 402nd picks have been worth about 94 WAR over the history of the draft, making that slot look slightly more valuable than the 11th slot, with the gap likely to widen for the next few years. Remove the highest (a certain #402 pick in the 1999 draft) and lowest, and the slot looks … more as expected.

StanTheManFan
Contributes any way he can.
He's normally a nuclear physicist
Except when writing for this list.

by StanTheManFan on Jan 11, 2011 5:22 PM EST up reply actions

That is a good idea

Expect an edit some time in the future.

i for one welcome our new fabulous overlords - gdm

by Aranathor on Jan 11, 2011 7:28 PM EST up reply actions

Quiz

If you have a 1st overall draft pick and a 30th overall draft pick, and they both put up the same numbers, who do you think will get more playing time? I would think it’s obviously the number one pick.

Players get playing time on a combination of performance and potential. Higher draft picks are generally deemed to have more potential than lower ones.

Skip Schumaker is a scapegoat

by vivaelpujols on Jan 12, 2011 12:08 AM EST up reply actions

Meaningless exercise.

No two players are identical. Even if, by some fluke, two players put up exactly the same numbers for a year, there will be things that differentiate between them — things more tangible and visible than what slot they were drafted in.

Show me one example of a situation where two guys had entirely comparable numbers (including age — that’s important), for teams with entirely comparable needs (because I claim that need is an important ingredient in who plays how much with what level of accomplishment), one drafted #1 (heck, I’ll be generous — 1, 2 or 3) and one drafted late in the same year’s first round. If you can find such a pair, a discussion becomes possible. I’ll be surprised if you find one.

StanTheManFan
Contributes any way he can.
He's normally a nuclear physicist
Except when writing for this list.

by StanTheManFan on Jan 12, 2011 12:15 AM EST up reply actions

It's a fucking hypothetical

I don’t want to go hunting for players. Refute my logic:

Players get playing time on a combination of performance and potential. Higher draft picks are generally deemed to have more potential than lower ones.

Skip Schumaker is a scapegoat

by vivaelpujols on Jan 12, 2011 2:14 AM EST up reply actions

What's to refute?

By the time players reach the major leagues, it’s not what draft slot they occupied that managers watch; it’s how they performed getting there. (“Performed,” in context, meaning not just what they did on a minor-league field on game day, but also what they did in front of coaches, scouts, etc., that the fans don’t see.) Is this not obvious?

StanTheManFan
Contributes any way he can.
He's normally a nuclear physicist
Except when writing for this list.

by StanTheManFan on Jan 12, 2011 8:56 AM EST up reply actions

And I would think prospect status is heavily correlated to draft pick..

Either way you look at it, players who are drafted earlier will get more opportunities.

Skip Schumaker is a scapegoat

by vivaelpujols on Jan 12, 2011 12:50 PM EST up reply actions

i think this would be more true in sports that don't have a minor league system....

Football and basketball.

I don’t think this holds water for baseball.

by The Ghost of Todd Burns on Jan 12, 2011 9:06 AM EST up reply actions

This data suggests

That except for the 1st pick, the order of selection in the first round is meaningless. But if you stop and consider it, that makes sense. The random variables of injury, psychological issues, failure to grow physically and other intangibles level the playing field.

Guess that’s why clubs hire legions of scouts and sabermetricians, eh?

by JWO on Jan 11, 2011 6:42 PM EST reply actions

BP looked at this a couple of years ago

and concluded that the first few picks really did have a higher probability of making it — first “few,” not just the first, but not very many either. I forget who wrote the article, but it was in the context of looking at the cash value of a #1, #2, #3, etc., pick. The finding was that the value fell off rapidly after the first pick, but there were still gradations for a while after that.

StanTheManFan
Contributes any way he can.
He's normally a nuclear physicist
Except when writing for this list.

by StanTheManFan on Jan 12, 2011 12:18 AM EST up reply actions

Looks like if you are drafted in the Top 10, you are a better bet than the rest of the 1st round

From then on, it’s fair game and doesn’t really matter where you draft. At the very least, the Top 5 is obviously head and shoulders above the rest of the draft.

DONNIE FUCKING JONES FOR PRO BOWL!

by stlcardsfan4 on Jan 11, 2011 6:53 PM EST reply actions

I wouldn't put too much stock into what the Nth pick means

at least the precise value of the Nth pick, especially when one A-Rod or Greg Maddux can make a single slot in the draft look awesome. 46 is still barely starting to approach statistically significant numbers. Especially when the overall shape of that graph looks a whole lot like an exponential curve or a power law or something like that. I’d be interested to see how it looks when you go out into the 5th round picks. (and that would be even more reason for caution. Mike Piazza would make it look like the 1390th pick is WAY better than the 1000th pick).

They say that it's never too late, but you don't get any younger...

by Valatan on Jan 11, 2011 10:27 PM EST reply actions

Yeah...

… but shouldn’t we expect an exponential decay here? The “Nth” pick might not be statistically significant, but the overall distribution is probably pretty robust. Esp. after you remove outliers as StanTheManFan suggests above.

by kindred on Jan 12, 2011 9:36 AM EST up reply actions

Yeah, that's actually my point

Espeically if the falloff continues into the higher picks (and I’d expect it would).

They say that it's never too late, but you don't get any younger...

by Valatan on Jan 12, 2011 9:54 AM EST up reply actions

Where's the acknowledgement section??

Sign Carl Pavano!!!

by guayzimi on Jan 11, 2011 11:51 PM EST reply actions

It probably hasn't been. . .

a consistent or long-standing enough phenomenon to make a difference statistically (at least not a significant one), but signability issues have made the ordering of picks less about talent/potential than they used to be.

Still, this is a fun graph. Good January post.

by SouthsideCardsFan on Jan 12, 2011 11:55 AM EST reply actions