The MLB Prediction Market

Sunday, June 01, 2008

1. Overview of the World of Scouting

In the world of building winning sports franchise nothing is more critical than the prediction of future success of athletes. In order to build a successful franchise, organizations must evaluate how a player will perform on for them in the future. A single wrong estimation could result in millions of lost revenue for organizations.

Of these player evaluations the most volatile are those whom have never performed in your league. In this world a player there is no barometer which shows what how he can perform at the next level, rather all evaluations and predictions must be creating using only physical attributes and statistical success versus lower competition. Whereas when evaluating players to acquire via free agency or via trade there is accurate data on how the athlete has performed in a similar environment.

In today’s world there are many websites including very popular ones such as and that are solely dedicated to evaluating and predicting how amateur athletes will perform both in the collegiate world and professionally. Several TV analysts including Mel Kiper have made a significant amount of money solely on the reputation for being able to project how collegiate athletes will transition into the pros. Additionally each professional organization has a branch of scouts whose sole objective is to find the player who will eventually help lead them to a championship.

The scouting world is universally important to all professional franchises however it is Major League Baseball that has the largest structure to support finding their next superstar. Unlike the NFL and the NBA where athletes are only drafted from college institutions, the MLB must scout and develop opinions on both High School and College Players. All in all the major league baseball draft is a total of 50 rounds in comparison to 2 and 7 rounds respectively for the NBA and the NFL. Additionally the MLB is much more dependent on scouting on foreign soil than the NBA or NFL. All foreign players are not subject to the MLB draft but rather open to sign with any team who offers him a contract. Thus finding the prospect first and developing a relationship can be your ticket to a ring.

Thus Major League Baseball organizations must go through extreme lengths to evaluate talent both on American soil as well as globally. Using your high draft picks on players who never amount to anything professionally can both doom your franchise as well as those who make the decision. But finding that who can throw a shutout in game 7 of the World Series can turn you into a hero.

2. Difficulty of the MLB Draft

To give a quick sum up of just how difficult it can be to select the right prospect let us look at the 2000 draft. Given that the draft occurred 8 years ago all of these prospects would be at minimum 26 and would be entering the prime of their baseball career, thus if they should be contributing for their pro team. Every year the first round consists 30 picks, these picks obviously give you the best chance to find a superstar as there are a larger pool of prospects to choose from. Of the 30 players selected in the first round only 5 players are on an active roster and only one, Chase Utley, has made an all star team so far in their careers.

Meanwhile several players drafted in the later rounds have had incredible success in MLB. The first of whom is Brandon Webb whom was taken with the 249th overall selection and has won a Cy Young (awarded to the best pitcher in the league). Other players drafted in 2000 that have been selected to at least one all star game are Grady Sizemore (75th), Bobby Jenks (140th), Dontrelle Willis (223rd), and Jason Bay (645th). For Jason Bay that means that MLB franchises thought 644 players were better than him and in his career he’s been selected to 2 all star games.

Each year the draft offers similar stories. Players that were evaluated by organizations as expected superstars who never end up amount to anything and players taken near the end of turn out to be diamonds in the rough. It is the job of a good organization to maximize good selections and minimize mistakes.

3. Main Purpose of the MLB Draft Market

The main purpose of the market is to create an increase in knowledge surrounding a process where decisions are based solely on the predictions of a collective group. By spreading out the predictions, the market would be able to pool more opinions together and potentially give a better (more uniform) prediction of the absolute unknown. For the purpose of avoiding incredible complexity the market will focus on those players that are eligible for the MLB draft and will not include foreign prospects.

4. Assets to Be Traded

All assets will be traded in association with an individual whom is projected to be taken in the MLB draft and would either be a normal predictive contract (0-100) or a strategic forecast which the buyer picks a range based on the total number of times an event will happen.

4a. Long Running Assets

These assets mark the prediction of the future success of the player. These assets give a prediction of how successful a prospect’s professional career will be.

Years to MLB: A key factor in drafting players is the estimation of how long it will take a prospect to break into the major leagues. If a college player is already at the level where he can be productive at the major league level it makes him that much more intriguing for contending teams. Sometimes teams will take a player in the top portion of the draft because they see a major league concern that the prospect could address and this tends to over way the prediction of whether the prospect can eventually become a star. The biggest example of this phenomenon is college relief pitchers, whom tend to have a quicker rate at which they reach the MLB. This would be a strategic forecast to predict the season they make their MLB debut.

MLB Longevity: Prospects are inherently more valuable to a franchise the longer that they play in the league. This would be a strategic forecast of the total number of games the prospect plays during his career in the MLB.

MLB Position: These would be predictions on what position the player winds up playing the majority of his games at in the MLB.

MLB All Star Appearances: A key factor in the success of a Major League Baseball team is to have players perform at an all star level. This long running asset would last the entirety of their career and would also be a strategic forecast of the total number of all star appearances.

MLB Hall of Fame: The absolute best case scenario for a draft pick is that they eventually develop into a player good enough to be given MLB’s highest honor. This would be a normal (0-100) contract and would pay out 100 if they made the Hall of Fame and zero if they did not.

4b. Short Running Assets

All of these assets will be complete either by the draft day itself or the following years draft when the previous years draft pick is no longer signable.

Projected Draft Position: A key prediction when setting up a draft strategy is knowing when other players will be selected by other teams. If a team has two players that they are equally interested in than often the tie breaker is which player they believe is more likely to not be selected by a different team prior to their next selection. This would be a strategic forecast based on what range of draft picks the player will go in.

Projected Salary: In MLB drafts there is no set scale as to how much a drafted player will be paid. For this reason each player has a given idea with how much they plan to ask for and how much they will settle for. Because these players may never have an impact on your professional team, having an accurate expectation for the athletes first contract salary is essential in selecting that player. This would be a strategic forecast based on the dollar amount the athlete signs for.

Likelihood to Sign: Because the MLB drafts High School players there is always the inherent risk that the player drafted will instead opt to go to college. Thus prior to drafting a high school player it is very important to have a prediction of whether or not they will sign with you, else you end up losing that seasons draft pick. In other cases players ask for more money than the franchise is willing to offer and opt to play for an independent team and enter the draft the next season (i.e. JD Drew). In this case the utilization of Projected Salary and Likelihood to sign would be used in correlation by teams. This would be a normal prediction (0-100) and would pay out 1 if they were to sign and zero if they opted not to sign.

5. Potential Users

I believe this market would both work externally or internally. Because MLB teams have so much invested in finding talent and drafting valuable players they have employed scouts around the globe. Thus they would have enough players with information regarding prospects where the market could potentially be successful. This could become a serious competition between different employees of the company and bonuses could potentially be awarded for whomever was most successful in their predictions. Additionally an organization could look to utilize this internally to get a competitive advantage over their competitors as it would be another tool that would could be utilize to aggregate all of the opinions and information from the organization.

If created externally it could serve as an information store for the entire league to utilize and for potential baseball fans to compete with their friends. By opening it up to the public many individuals, not just team scouts, would be able to predict how well the given prospect will perform at the next level. Granted some of these opinions will not be expert opinions but overall it would give

6. Main Rules

The Prediction Market can be run either with fake or real money. If the market was created as a public market than it may be beneficial for it to be run off real money. The reason behind this is that MLB scouts, the best source of information, would likely need an incentive to participate. If the market was run with real money than it may give scouts who are confident in their ability to predict the future performance of players to partake in the market.

7. Potential Limitations

The potential limitations are the depth of players whom could be entered into the prediction model. Because the draft is 50 rounds long with typically around 1,500 players selected it is obvious that not every individual will have any information, especially first hand knowledge on a tradable asset. Thus there is the potential for a large amount of noise trading if only a select few have valuable information on a tradable asset.

Additionally it would probably much more difficult for someone to offer a negative opinion about an athlete whom is not yet 22 than it would be for a positive one. Thus it might be expected that the predictions be slightly over optimistic. This may especially be the case when only few individuals have knowledge on the asset as these individuals might think that they found the hidden gem that no one else has thus buying up the asset and driving up the assets cost.

Finally some of the assets rely on the player (MLB Position) at least playing a single game in the MLB, a strategy would need to be taken into account on how to payout the assets if the player never in fact played a single game in MLB.

8. Conclusion

The Major League Draft is based solely on a handful of predictions. An organization must first be able to evaluate accurately the future success of athletes who may only be 18 years old. Before they actually make these selections they need to predict how much these players are going to ask for and what round they may go in so that can better prepare themselves to make solid decisions. Finally, the MLB Draft might be the most important two days each year for a franchise as their talent is dependent on its success. For these reasons the MLB draft is the perfect prediction market.

Posted by Simon at 9:29 AM   Digg! submit to reddit BallHype: hype it up!


Post a Comment

Advertise Here!