Betting on the emergence of prospects can be frustrating—apart from such matters as the caliber of the guy above the prospect on the organizational chart, and the will of the parent club to avoid contractual triggers, there’s the question of whether the prospect’s approach in the minors will even translate to the majors. However, once a player reaches the majors and gets an extended audition, predicting when he’ll leave should be more straightforward.
We decided to explore this question. We wanted to find the level of performance at which fielders post their final season of consequence.
We gathered all seasons from 1990-2008 that may have been fantasy-relevant. We chose 300 AB as the minimum load for a potentially valuable season; that’s probably low, but we wanted to err on the side of too many seasons. In each season, we assigned players to the position where they played the most games. (If a player played two positions equally often, we qualified him at both.)
We then asked: What is the OPS for these players in the year before they disappear—i.e., in their final 300-AB season? We could have used a more sophisticated metric than OPS, such as wOBA or Predicted OPS, but we went with OPS for its simplicity and familiarity. (In truth, the question is not by what measure players should be deemed done but by what measure players are deemed done by major-league GM’s. For all we know, GM’s steer by BA….)
We expect that the onset of obscurity varies by the offensive demands put upon the player, so we grouped the results by position. Note that the only position that mattered was the batter’s position in his final 300-AB season; we did not track whether players were shifted from more defensively stringent positions.
Because 2009 is not in the books, we did not treat 2008 as anyone’s final season (there may be a few players who last played in 2007 but whose careers are not over, but they should not soil the analysis). We did include players who had just one 300-AB season (by definition, their last one).
Here are the results:
|End of the Road|
Note that these are aggregate levels; some batters had better numbers when they checked out and some had worse. And obviously, there are team-level considerations that we are missing, notably who (if anyone) is ready to take over. Still, the trends are as we would expect: The bar for further paychecks is high for first base and the corner outfield positions and lower for 2B, SS, CF, and C.
This is a sound starting point—when a player slips to this level, he (and you) should be scouting other opportunities. However, even better would be to say “For an OPS of X, a player’s chance of losing his job is Y.”
So let’s try that. For players who qualified at first base, we arranged all the player-seasons from highest to lowest OPS. We would expect to find many more “final destinations” at the bottom of the list than at the top, and indeed there are: Only one of the 25 top-rated seasons (4%) was a cul-de-sac, whereas 9 of the 25 bottom-rated seasons (36%) were.
Here is the graph for first base. The blue line is OPS, for non-overlapping buckets of 25 batters. The purple columns are the observed fade rate for each group; the thick black line is a trend line. (We say “fade rate,” not “extinction rate,” because the players might still putter around baseball; however, never again do they log 300 AB in a season.)
The trend line lolls around 5% for a while (even the best players are vulnerable to a career-ending injury). And then, starting around .850 OPS, fade rate rockets up, eventually surpassing 30% for the dregs of MLB first basemen. (The miracle might be that that rate is not higher; it may speak to the slow pipeline of talent within an organization.)
Armed with this chart, and knowing the OPS of a first baseman, we can now guess his chance of not attaining 300 AB next season. Here are the projected fade rates for the first basemen who are on pace for 300 AB this season:
|Bailout Neediness, 1B|
|Player||YTD OPS||Fade Rate|
There’s roughly a 50% chance that at least two of the bottom five players won’t be entertaining fans in 2010.
There is a lot of room to extend this study. The natural next variable would be age—it is possible that MLB owners are pokier with the pink slip for younger players than for older ones. We could also focus on base skills, such as contact rate and walk rate, rather than on surface stats. Still, this approach is fresh ground for figuring long-term worth, for fantasy and major-league GM’s alike.