How to Fix the MLB Draft

Fangraphs writers have recently written two articles discussing how they would fix the draft. David Cameron wrote the following:

So what if there was no draft? Instead, what if we just lumped all new players -- foreign or domestic -- into a single acquisition system where each player was free to sign with the team of his choice, only with firm spending caps in place to ensure that young talent flows more freely to clubs that can't compete on major-league payroll alone? In other words, a team's talent acquisition budget would be inversely tied to its major-league payroll; the more you spend on big leaguers, the less you get to spend on prospects, and vice versa.

This proposal has a few problems. The first problem is that it will likely convince teams not to spend more money on payroll than other teams. Carson Cistulli created a graph showing how large each team's draft pool would be in 2015 based on this proposal and determined that the Dodgers would have no money in their pool and the Yankees would only have $2.2 million. It is unlikely that teams are going to be willing to spend a lot of money on players if it means they can't afford to draft any talented prospects. In addition, teams like the Yankees are currently able to add international talent with only minimal repercussions. If this method is implemented then this loophole will be closed and it will make teams really loathe to spend more money than anyone else.

The second problem is that even if teams like the Yankees were still willing to spend a lot of money on players they would likely fail in this system.  The Yankees have one of the largest payrolls in the majors but are struggling to remain competitive because they've been unable to develop quality homegrown talent. Most production is contributed by team-controlled players. If teams like the Yankees and Dodgers are prevented from drafting any of the best players then they will struggle even with large payrolls. This will cause fans of the Yankees and Dodgers to stop attending their games and therefore a significant loss in revenue, which would be disastrous for baseball.

However, it does make sense to disassociate draft position with record. Associating draft position with record encourages the Astros to spend minimal amounts on free agents in order to pocket revenue and receive top draft picks. Does baseball want to encourage the Astros to tank? Nor is it reasonable to give the Mets or Phillies better picks than teams like the Royals, Orioles, and Pirates simply because they had a worse record. Shouldn't we want MLB to reward small-market teams for being successful? If the draft is a way to ensure parity by rewarded small market teams then it should reward small market teams. The current method rewards failure instead of supporting small market clubs.

I think Cameron's idea of a draft pool where players can be selected is a good idea but using payroll is a mistake. Rather the draft pool should be based on an obscure metric called the Revenue Sharing Performance Factor.

The Revenue Sharing Performance Factor is discussed in the CBA and one can find a chart with the factor for each team on attachment 26, page 235 in the PDF and page 222 in the paper copy. Basically, this factor is meant to ensure that each team is paying a fair amount into revenue sharing based on revenue earned and market size. You can find a more detailed description starting on page 132 in the PDF and 119 in the paper copy if you're interested.

Using the metric, the pools would look like this in 2012 and 2013 presuming that teams could spend a total of $295M in each of 2012 and 2013:

The Yankees would have a pool of less than $7 million while the Royals would have a pool of nearly $10.3 million. This would allow the Royals to sign better prospects than the Yankees but would still give the Yankees ample opportunity to add talent. In addition, the Astros no longer receive a reward for tanking as their draft pool is at $9.4M each year while the Mets and Cubs would have a draft pool of about $8.5M. This wouldn't force those clubs to spend money in free agency but it would make it harder for them to succeed if they didn't.

Implementing this method would lead to other changes. It would no longer make sense to allow teams to give qualifying offers to potential free agents because small market teams would have a built-in advantage due to the fact that they can’t sign their best players when they become free agent eligible. In addition, it would make sense to completely abolish the luxury tax. This method encourages teams in large markets to spend money on free agents if they want to be competitive. If so, it isn’t fair to penalize them for spending money. It would also make sense to eliminate the supplemental revenue sharing plan. That way, teams that have a larger revenue and market size than average but aren’t likely to spend enough money to be subject to the luxury tax would see some benefits from this deal and would be more likely to support it.

This deal is good for small market clubs by ensuring that they have a larger draft pool than large market clubs regardless of record. It would help large market clubs by ensuring that they no longer need to pay luxury tax and wouldn’t have to share as much money via revenue sharing. It would be good for all clubs because it would control spending on both foreign and domestic prospects. The richest clubs wouldn’t need to spend millions on foreign prospects and the poorest clubs would have a fair chance to convince foreign prospects to join their organization. It would be good for the players union because prospects would be able to choose their team and the elimination of the luxury tax would encourage spending by rich teams. Unlike Cameron's plan, this has benefits for all teams as well as the players union.

It seems like a win-win-win-win situation to me.

Receivers in Name Only

You may have never heard of the baseball term RINO (Receivers in Name Only).  The likely cause of this is that I think I just made it up.  What it is trying to make clear is to communicate what is a popular perspective in several front offices throughout baseball: catchers without sufficient defensive skills when drafted are highly unlikely to acquire those skills during their career.  Maybe the term sticks, maybe it does not, but hopefully this article addresses whether or not this sentiment is true.

The way I decided to test this was pretty simple.  I took all draftees listed as a catcher from 2004 through 2008.  I decided to ignore any position switches because what we are interested in are catchers at the MLB level.  Whether or not they hit well enough to assume another position (e.g., Gaby Sanchez) does not resolve the issue of catching depth in a system.  I gave them credit for service at a level (e.g., MLB, AAA, AA, A, Hi A, A, Short A, Rookie) if they had over 100 games played at that level or accumulated at that level and a above.  For instance, if a player had 70 games caught at AA and 40 at AAA, that player would be credited with having reached AA with 110 games at or above that level.  All this was how the study measured performance, nothing more.

To determine how a catcher's ability was viewed at the time of being drafted, I consulted Baseball America's draft database scouting reports.  I decided to be conservative in what I labeled as a defensive catcher.  The following are several of the terms used that would result in a player being classified as such:

best defender in the class, pure catch and throw, solid average, at worst fringe average, best defensive catcher seen since Charles Johnson (a label that adorned Matt Wieters' scouting report, which is contrary to my memory)

Draftees labelled as poor defenders included labels like these:

needs improvement behind the plate to stick, defense remains his nemesis, passable, adequate, no scouts see him at catcher, deserves a look behind the plate, will move to third

These designations left us with 37 catchers labeled as defensive drafted in the first five rounds and 28 labeled as non defensive.  This may not be the most robust data set, but it might be able to yield some answers if the differences between these two groups are large.

How well do defensive catchers advance through the levels?

 If you simply count each level attained as a single point, sum them up for each player individually, and group them by their defense, then defensive catchers significantly (p=0.001) rack up levels played as a catcher in professional baseball.  The table below shows the total number of players in each group that are credited with time spent (100 games or more at that level and above) and the percent of that group that made it to that level.

	Defensive		Non Defensive
	n		n
Total	37		28
MLB	16	43%	5	18%
AAA	24	65%	9	32%
AA	29	78%	13	46%
Hi A	31	84%	15	54%
A	36	97%	20	71%
Short A	36	97%	21	75%
Rookie	36	97%	22	79%

As you can see, defensive catchers are likely to make it to the Majors 2.4 times more likely than non-defensive catchers.  Keep in mind that this success rate does not include defensive catchers Kyle Skipworth, Austin Romine, and Travis D'Arnaud who all stand a very reasonable chance of achieving that 100 game MLB mark in the next year or two.  That would raise the success rate to 51% and increase of probability to 2.8 times more likely to make the big show than non-defensive catchers.  In other words, if you are an organization looking to improve catching depth within your system then you should stick to the amateurs who already show proficiency in the position.  However, if you really love a bat and the player just happens to be listed as a catcher then feel free to select him knowing that it is highly likely that he has to exchange his catching mitt for a fielding one.

One question that may be asked would be whether or not non-defensive catchers are more likely to be selected in the later portions of the five round database of drafted catcher that I put together.  In fact, the average round players were drafted in was higher for the non-defensive catchers (2.5) than the defensive catchers (2.8).  For the most part, defensive and non-defensive catchers drafted in the first two rounds were in similar proportion in their groups.  The non-defensive catchers attained a higher average draft round because teams drafted few of them in the fifth round in comparison to defensive catchers.

Below is a table parsing out the round data for each group as well as the MLB attainment rates for those groups.

	Defensive		Non Defensive
Round	Success	n	Success	n
1st	64%	11	14%	7
2nd	57%	7	0%	5
3rd	33%	6	14%	7
4th	33%	6	29%	7
5th	14%	7	50%	2

A first round catcher with existing professional defensive skills is 4.5 times more likely than a non-defensive catcher.  If you add Skipworth and D'Arnaud in that then the percent goes up to 82%, 5.9 times more likely than a non-defensive first round catcher.  Romine's addition would put the second round at 71%.  Keep in mind that these are small sample sizes, but these are rather impressive differences that suggest that there is a trend here.

Conclusion

The conclusion is this: if you want a catcher, then draft an individual who has already displayed professional catching skills.  If you really like a bat and the player just happens to have played catcher in high school or college, then do not talk yourself into dreaming about him as a catcher and respect him simply for his bat.  Otherwise, an organization may just find itself a few years down the road with no internal solutions behind the plate.

Cup of jOe's: Value of Drafting Young High Schoolers

I tweeted about this the other day, but figured that these two articles published by Rany Jazayerli on Baseball Prospectus need to be highlighted. 

Starting Them Young Part I
Starting Them Young Part II

Now, he calls these pair of papers the "most significant finding of his career."  Well, this is not DIPS.  This is not exactly earth shattering and I imagine that quite a few front offices have had this information for years with a large number of scouts knowing this since the 1800s.  However, it is significant in terms of what was likely known ten years ago and it is significant in that this information is now public.  What makes it significant is that we now have a quantitative way of expressing the advantage behind drafting younger high school players.  It is also significant because it shows that MLB has not really figured it out as of 2003 otherwise we would not see such a great difference between ages.

Wait, a second...you might say...what about Billy Rowell? he was quite young when drafted.

Well, it certainly is not a fool proof system.  No one is saying that drafting a young player automatically means you are making a greater selecting.  It means that you have a greater probability of getting more value out of high schoolers drafted at a certain slot if they are younger. 

It also brings up another point, which is why use mean instead of median?  Using mean probably makes more sense because you tend to draft for stars instead of average players.  However, median makes sense if you are trying to build depth.  It may also make sense within small samples because massive outliers can overwhelm a mean.  I think this is still up for discussion, but it certainly is a discussion not many are having.  Or, it was decided a long time ago and I have been unaware.

Another article you should also read is this one by Scott McKinney.  It is a comprehensive break down of minor league prospects.

Is a draft pick worth the mean or median value?

A topic that has long be gnawing on me is whether or not it makes sense to value draft picks as many who connect a dollar sign to picks do.  To me there is an argument between using the mean (which is common) versus using median (which I have not seen).  Both are ways to measure central tendency.  Mean is often the best to use in order to either describe the population as a whole or to determine typical value if value are normally distributed.  Medians are used often when you want to determine what is a typical result when the population is distributed asymmetrically.

The following graph shows the difference between how we assign draft picks value either by mean or median.

click to make larger

In the first 30 selections from 1991-2000, there were only four instances where the median was higher than the mean.  Seventeen median values within the 30 picks were worth zero WAR or less.  In other words, a team is not likely to receive mean value for a pick.  Picks tend to be overvalued because a few individuals increase the worth of the pick within the population.

What is interesting here is how do you exactly come to a specific value for a pick?  Does a 33% chance of getting a useful MLB player mean that it is permissible to overvalue the population as a whole?  It is easy to see how many MLB teams deplore spending money in the draft because much of the money spent on individuals is wasted.  The question is whether or not the money spent on the entire portfolio is beneficial to the team.  I would say that if you are evaluating a single player, the median makes sense.  If you are evaluating the population, then the mean makes sense.  To look at it differently, it is like a lottery that benefits the buyer of the tickets.  Most likely, most of your tickets are worthless, but a couple might wind up bringing back a great deal of value.

What is the Value of a Compensation Draft Pick?

This past season Tyler Beede decided not to take a signing bonus north of 2MM with the Blue Jays and instead committed himself to college.  He was the 21st selection in the first round.  As compensation, the Jays will receive an unprotected pick in next year's draft, which is the 22nd selection in the first round.  A question quite a few asked was: what, if anything, have the Blue Jays lost in terms of value?  There are several considerations.

A pick being delayed a year.

A team often relies on waves of highly talented, cost-controlled players graduating to the Majors each year.  If these players are not available, the team has to utilize the free agent market where costs are significantly higher.  For instance, if a team loses it's 21st round pick it will have to supplement it's lineup roughly the cost of what that pick would have provided.  The 21st round pick is worth about 10MM on average over the course of his career.  If that first year needs to be covered, then it will cost the team arguably 1.7 MM.  This value is not uniform for all selections, of course.  The higher up in the draft a player is selected, the greater the expected value of a player as players with greater value are typically chosen earlier.  If it was a first round pick being lost, the expected loss of one year of value would be around 13.3 MM.  The graph below shows the relationship between cost-controlled (first six years) WAR and draft pick selection.  The drafts used for this graph are from 1991-2000.

click to enlarge

It may be argued that in the grand scheme, a one year delay on receiving value is largely inconsequential.  This is likely to be a rather accurate assessment.  The difficulty in projecting players into the future and when they may be able to help the Major League team means that teams largely are not relying on these players to develop.  In a general sense, they need these draftees to develop and contribute, but few teams will set their watches to players producing except at the front end of the draft.  Only there do you find truly remarkable talents that teams will expect to advance quickly and be productive members of the organization at the highest level.

Draft Budget and the Unprotected Nature of Draft Pick Compensation

In practice, the greater concern in practice is how having multiple draft picks in the first round affects the quality of talent being selected.  Additionally, compensation picks are unprotected, which has tended to cause teams to select players more conservatively and reach a little bit.  In 2009, the Nationals selected Drew Storen for 200k less than slot.  Storen has been incredibly successful in the Majors as a reliever.  However, one wonders how important it was for the Nationals, a team in need of impact players, to select a pitcher who throws an inning every few days.  This, however, is not a uniform strategy as this year the Diamondbacks drafted and signed top ten pitchers Trevor Bauer and Archie Bradley.  Both required significant investment and the Diamondbacks accomplished that.

So, how much more useful is one approach than the other?  In the graph below, picks are grouped in fives over the course of the first thirty picks in the draft from 1991-2000.  Those players' control year WAR is compared to players selected the following year a selection behind them or a more conservative ten places later.

click to enlarge

The graph shows that there is a negligible difference between the value of the initial selection vs a selection a year later and a pick later.  However, there appears to be a sizable difference if a team with a compensation pick in the first 20 selections decides to be conservative and draft a signable player (defined in this study as a player taken 10 slots later).  A conservative approach for a top 5 pick would result in a loss of about 4 WAR (~20 MM) while picks 6-20 result in a loss of about 2 WAR (~10 MM).  It appears to be a rather large misfire if a team does not fully utilize their picks.  A simple investment of an extra 1-3 MM results in a several fold return.  Even if an unprotected draftee recognizes his signing team is in a position where they need to sign him, it is highly unlikely the pick would ever receive more money than what the average pick would give back to the team.

Differences in Talent Between Draft Classes

However, this brings us back to the Blue Jays.  Are they worse off for not signing Tyler Beede?  According to this quick study, no they are not worse off.  In an average year, players available at pick 21 are typically equally available the following year at pick 22.  The assumption is though that this year is an average year and next year will also be an average year.  The following graph shows differences in total six year WAR for the first 30 picks in each first round from 1991-2000.

Click to enlarge

This past year's draft was considered one that was quite full with talent.  The majority opinion would find that this year's draft is likely to be similar, in total value, to those in 1992, 1993, 1995, 1997, and 1998.  The early opinion for next year is that it will be an average class which would be similar to 1991, 1994, 1996, 1998, or 1999.  The difference between this and next year is about 30% if these opinions are accurate.  That assumed difference in draft classes between this year and next is illustrated on the following graph which takes the values in the second figure and adjusts them according to general draft worth.

click to enlarge

In general, if the talent level in the current draft appears to be significantly greater than the following draft then it makes sense to aggressively sign those players.  However, if the following draft appears to be more talented then it might make sense to be not so giving during negotiations and to feel free to utilize a compensatory pick the following draft. 

Oh, krikes!

I was just scooped.  I am halfway through my analysis on draft pick performance, doing almost the same exact thing.  I guess that is done for me.

Anyway, I have not yet gotten to it, but read this.

Looks like I will have something cool to read tonight.

Looking at the Orioles and Rays Prep Focus

The following maps show the intensity of draft focus for the Orioles and Rays as measured by the percentage of prep players drafted from each state over the past five years.





After the jump, a comparison graft between the Rays and O's.

More thoughts on Puerto Rico and Draft.

As some of you long timers may remember, I wrote a somewhat popular post back in the day comparing the decrease in MLB talent from Puerto Rico and how it may be similar to the decrease in American blacks in baseball. Today, I decided just to illustrate the decrease in MLB production from Puerto Rico. In the graph below, I included three countries who were strong in developing talent in the 1970s and 1980s. I used birth date to approximate how the 1990 inclusion affected each country and the prospects they produce. With that in mind, I would read 1973 and prior as before the draft. 1974-1978 as a transitional time where the infrastructure is still having an affect on prospects. Post 1978 should be seen as when the infrastructure has probably been significantly degraded and no longer much of an issue.

Click on the picture to make it larger and more legible.


After the jump, the raw numbers.

Years PR Ven DR
54-58 16 5 26
59-63 20 15 31
64-68 32 18 57
69-73 29 28 81
74-78 39 47 108
79-83 19 67 105

Likelihood of not making the Majors: 1990s Draft Picks

The graphic is pretty self explanatory . . . I think.




Forewarning . . . there is nothing after the jump.

What does Puerto Rico's inclusion in the draft mean?

For a long while, I have been thinking about the draft and what it means to domestic and international talent. As can be seen on a pedestrian level, domestic ball players are greatly restricted and must subject themselves to a draft where they have no real ability to determine who they will sign with and for how much. They can certainly demand a price, but they have only two options: go back into the draft (losing a year of earning potential) or come to some compromise with the drafting team. International talent is not subjected to this. They can sign with whomever they wish and, arguably, have more control over their own future. This seems grossly unfair for domestic talent. One idea is to incorporate international talent into the draft.

Another blog hit on this . . . Fantasy Baseball Generals (hat tip to Sabernomics). They mentioned how the talent pipeline from Puerto Rico was quite rich in the eighties. Several players were produced from that time period, including the Alomar Bros., Carlos Baerga, Wil Cordero, Carlos Delgado, Jose Hernandez, Javy Lopez, Juan Gonzalez, Ivan Rodriguez, Benito Santiago, Ruben Sierra, Jose Valentin, and Bernie Williams. Since 1990? Carlos Beltran, Jorge Posada, Javy Vazquez and Jose Vidro. Thirteen major players in the 80s and four in the nearly twenty years since. What happened? In 1990, Puerto Rico was incorporated into the amateur draft.

The basic reason?

But with the addition of Puerto Rico to the annual amateur draft, a team no longer had incentive to invest money in developing relationships in Puerto Rico because a player they spent money on could be drafted by any of the other teams in MLB. So money that might have gone to Puerto Rico now went elsewhere. Like Venezuela, which has sent Bobby Abreu, Edgardo Alfonzo, Miguel Cabrera, Carlos Guillen, Ramon Hernandez, Richard Hidalgo, Victor Martinez, Melvin Mora, Magglio Ordonez and Johan Santana, among others, to the majors since 1990. The Astros have been very active in Venezuela, signing Abreu, Guillen, Hidalgo and Santana from the above list.

There was no reason to spend any money on Puerto Rico. There was no reason to reach out and invest in 13 or 14 year olds. Talent evaluators and the academies no longer served a purpose because anyone could draft the players. The draft was devastating to Puerto Rican talent. It makes you pause for a moment and fully try to comprehend what it would mean to other Hispanic nations with similar per capita income. You would see talent dry up from those regions. The great Hispanic influx is, most likely, the result of MLB teams setting up a talent market. They get to Hispanic kids earlier and help them with development. This is not what American born talent receives.

This not only addresses a rather important and significant issue for the Carribbean, Latin American, and South American talent sources . . . but also the poorer regions of our country. More specifically, I am referring to the well-worn discussion about the decreasing participation of African Americans at the Major League level. Several teams do not have a single black player. This was brought up last year as the Astros donned Jackie Robinson's 42 to commemorate his breaking through the color line. Critics commented that it done because the team had no blacks on the squad. It was considered disgraceful by some. Though, maybe not as disgraceful as having Marlon Wayans are the key note speaker for the Dodgers celebration. Really, Marlon Wayans. Anyway, the question has boiled down to way are blacks not as well represented today as they were 30 or 40 years ago?



Well . . . the simple answer is that we developed better ways of scouting and developing talent. In the United States, this development is offered to those who can afford it. Elsewhere, this development is funded by individual MLB teams, so money is not important to the talent themselves. Based on the 2000 census, white households average nearly $50,000 as their per capita income. Blacks average about $30,000. This difference may explain why white ball players are able to afford, as youths, to play baseball year-round, attend clinics, participate in expensive highly competitive amateur leagues.

Income disparity also affects early youth development. Baseball is a much more complex game and difficult to practice. You need multiple players (you cannot play 2 on 2) and you need expensive equipment (glove, bat, and several balls as opposed to a single ball). Youth baseball does not incorporate shared equipment beyond bats and balls. I would also argue that baseball requires more oversight and adult supervision simply in terms of organization and umpiring games. This makes it difficult for youths to begin developing skills important for baseball. Add this in with the difficulty in participating in expensive, highly competitive amateur leagues as teenagers . . . and there isn't much positive feedback.

The final point is college. Income disparity also affects this. The quality and number of scholarships awarded to football dwarfs what baseball programs offer. Baseball scholarships are far more competitive than football scholarships. It is also an uphill climb if you did not develop skills as a youth and only have underfunded high school programs as your only significant source of instruction and maybe a few free instructional clinics, if you are invited. It is a losing proposition. I doubt many kids and families are consciously making this decision, but after a few decades . . . you'll eventually see the guys who made good and got a college education were not the average baseball players at your school, but the average football kids that were picked up. That unspoken positive feedback has to reach the community at some point. I think this is what happened.

So, the article that inspired this entry says that what needs to happen is for the draft to be eliminated. So what would happen? I imagine that international talent will get hurt badly. Not as badly as putting a draft in place, but the infrastructure that could be utilized in the US might be better to set up academies here. I imagine that the elimination of the draft would result in more American born black youths being involved in baseball because teams would scout out the 13 and 14 year olds and provide training and support. We will be exploiting talent in the US to a similar extent to our exploitation of talent internationally where there are far fewer regulations. It definitely is a tricky situation.

I may be wrong about many of my thoughts above. I try to keep an open mind, but I am fully a product of white bread suburbia. I think Stotle worked on programs with inner city kids, so hopefully he can enlighten me and correct any misguided assumptions I have.