- FanGraphs Baseball - http://www.fangraphs.com/blogs -

Do Prospect Busts Ever Reach Their Potential?

The top prospect label is a form of baseball tenure. No matter if the player is deemed a bust, what once vaulted him to the top of prospect lists will continue to earn him big league opportunities.

Players aren’t considered to be the best of the best by prospect mavens unless their skills project extremely well at the major league level. When a player of this ilk fails, analysts can usually point to a specific flaw preventing his success. If a former top prospect is struggling to stick in the big leagues, many organizations will take a flier on the player to try their hand at solving his problems. If successful, a team can reap the rewards of a cheap and effective player. Otherwise, the investment was minimal, and the team can cut ties without any real risk.

This is why players like Lastings Milledge, Brandon Wood, and Jeff Francoeur will continue to get signed. In covering all three of these players over the last month, I got to thinking about how frequently these prospects-turned-busts turned things around at the major league level. Does it happen?

It should be known upfront that the idea of taking chances on these players will remain sound even if the rate at which they turn things around is low. Any strategy is reasonable when little or no risk is involved, even if it is ineffective in the end.

To that end, the goals here involve investigating the success rate of perceived failed prospects making significant improvements at the major league level and identifying players who fit the mold. These players would conceivably give suitors of the aforementioned triumvirate hope that their low risk investments will reward them greatly.

I first pooled together all batters who debuted at, or before, the age of 23 years old, and after 1950, under the mindset that a player will not usually debut at such a young age if he wasn’t ready to produce in the majors. Players considered ready at that young age were more likely than not top-tier prospects. From there, I gathered together the WAR tallies for these players from their age 23-26 seasons. The resulting players were further filtered in the following manners:

a) at least 400 total plate appearances across the four seasons
b) had to play in the majors for at least parts of all four seasons
c) not once in the four-year span did they surpass 1.6 wins above replacement

The last criteria can hereby be known as “The Brandon Phillips Filter”, as the Reds second baseman was given ample opportunities to succeed in spite of his relative failures early on, never producing more than 1.6 WAR with the Cleveland Indians. After applying these restrictions, a grand total of 340 players emerged. Some noteworthy names on the list: Jose Guillen, Vinny Castilla, Scott Speizio, Wily Mo Pena, Russell Branyan, Laynce Nix, David Ortiz, Jeremy Giambi. The players in the initial result set averaged the following:

23 yrs old: 171 PA, 0.04 WAR
24 yrs old: 240 PA, 0.14 WAR
25 yrs old: 255 PA, 0.18 WAR
26 yrs old: 253 PA, 0.14 WAR

All told, the group came close to collectively defining the replacement level in the formative years of its members. From there, I added in player statistics over the fifth, sixth, and seventh seasons of the spans, which would correspond to ages 27-29, a range in which players tend to peak. No plate appearance floor was explicitly applied, but the players did have to play in the majors at some point during all three seasons.

The sample dropped from 340 players to just 207 players, which makes a great deal of sense: given the low WAR averages from the ages of 23-26 years old, it stands to reason that many of the players were given up on, and not afforded further opportunities to turn things around. For the 207 players who survived all of my filters, here are their pertinent numbers from ages 23-29:

23-26 yrs: 240 PA, 0.17 WAR
27 yrs old: 289 PA, 0.82 WAR
28 yrs old: 294 PA, 0.81 WAR
29 yrs old: 273 PA, 0.68 WAR

The players who stuck around deserved to stick around. The average WAR marks are still rather low, but a smaller subset of players definitely turned things around and reached, or exceeded, their potential. A total of 21 players surpassed the Brandon Phillips filter by producing more than 1.6 WAR in each of their age 27, 28, and 29 yrs old seasons.* Here are the pertinent averages for these players:

23-26 yrs: 211 PA, 0.79 WAR
27 yrs old: 486 PA, 2.85 WAR
28 yrs old: 544 PA, 3.20 WAR
29 yrs old: 558 PA, 3.29 WAR

This group of players still averaged out to a below average player from the age 23-26 period, but improved drastically after being given further opportunities to succeed. But a group of 21 out of an initial resultset of 340 suggests a success rate of approximately 6.2 percent. It isn’t impossible for players like Wood, Milledge, or Frenchy to figure things out, but the odds are stacked against them.

As was previously mentioned, though, the low success rate in no way renders the low-risk/high-reward strategy ineffective. For every David Ortiz there may be 17 Ben Petrick‘s, but that means there is still one Ortiz. The production of the player that succeeds, over a few seasons, could potentially exceed what the team didn’t benefit from in the other 17 failed attempts.

Seeking these low-risk players is an apt strategy when constructing a roster, so long as the success of the team does not hinge upon such players resolving their issues.

*The full list of these 21 players: Jose Guillen, Vinny Castilla, Damion Easley, Eric Karros, David Ortiz, Matty Alou, Don Demeter, Rick Dempsey, Jim Gantner, Larry Herndon, Jerry Lumpe, Mike MacFarlane, Dave May, Rance Mulliniks, Tim Naehring, Ben Ogilvie, Leon Roberts, Roy Sievers, Duke Sims, Charley Smith, Mickey Tettleton.