Weighted WAR: Negative Seasons and Center Field Rankings

In the discussion of weighted WAR, we’ve covered the weighting system, schedule length, and timelining (part 1 and part 2), plus an aside to touch on positional classification. Now, we’ll finish up the discussion of WAR adjustments with one more topic: negative seasons, or more specifically, the reasons that I discard negative seasons from the rankings.

To be clear up front, I understand why negative WAR scores exist and agree that they should. It is in fact possible that a player can perform poorly enough to expect that a minor league journeyman would do better. Negative WAR values are relevant in any number of contexts – evaluating a team’s overall performance, projecting production moving forward, and so on. I just don’t think they add anything useful to weighted WAR’s evaluation of the player’s career.

Weighted WAR: Timelining part 2 and Second Base Rankings

To this point in the weighted WAR series, we’ve introduced the weighting system, the schedule length adjustment, and the method for classifying players by position. Most recently, we looked at the top shortstops through the perspective of the timeline adjustment – in particular, the adjustments for segregation and minor league control.

In the timelining discussion, I mentioned performing a sanity check by examining the total number of top-100 players at each position who come from each decade. (More specifically, I classified each player by the season in which their median plate appearance occurred and broke those down by decade.) In general, the number of representatives from each decade increased over time, but there were two major exceptions to this. First, the 1940s trailed behind the ‘20s and ‘30s; this is easily explained by World War 2 cutting short the careers of many players from that decade.

The second anomaly was much more unexpected: the 1880s were significantly ahead of the 1890s. After considering this discrepancy, I decided to implement a third timeline adjustment, this time for league expansion.

Weighted WAR: Timelining (part 1) and Shortstop Rankings

So far in the weighted WAR series, we’ve introduced the weighting system and the schedule length adjustment, and most recently talked about positional classification of players. Now, we’ll shift back to the adjustments I’m making to WAR in this system, which means it’s time to discuss the elephant-sized landmine in the room: comparing players across eras, often referred to as “timelining.”

Weighted WAR: Positional Classification and Left Field Rankings

So far in the discussion of weighted WAR, we’ve explained the weighting system itself, as well as the first adjustment to the base WAR totals, schedule length. We’ve also examined the system’s ranking of the best third basemen ever. But as we continue exploring both the system itself and the rankings it produces, there’s a question that needs to be answered: What does it mean to rank, say, the best left fielders of all time? More specifically, are we ranking players based on their value in left field alone (which I’ll call the “positional purist” approach), or are we evaluating their entire careers and using some criteria to classify that career into one position (the “whole career” approach)?

Weighted WAR: Schedule Length Adjustment and Third Base Rankings

Resuming the discussion of the weighted WAR system, we start with (hopefully) the least controversial modification of the WAR totals from individual seasons: schedule length.

Schedule length is an obvious problem when comparing players across seasons. Fortunately, using estimates of wins provided by a player (e.g. WAR) allows for a fairly systematic approach here. If a player posts a 3-WAR season, it will have a much bigger impact on an average team over a 60-game schedule (improving winning percentage by .050) than over a 125-game schedule (.024) or a 160-game schedule (.019). But a straight-line adjustment doesn’t quite work, as the standings will generally have a wider spread after 60 games than after 160. What you’d want to do is estimate how much impact an individual win has on the standings at various schedule lengths.

Introducing Weighted WAR

The last several decades have seen a veritable explosion of baseball statistical analysis, both by fans and by actual MLB teams. One of the primary products of this effort has been Wins Above Replacement (WAR), the general estimate of the number of wins a particular player provided in comparison to a freely available substitute. The merits of WAR have been debated ad nauseam since its introduction, but its appeal is obvious; it combines all of a player’s contributions (or at least, all that we have measurements for) into a single number that allows direct comparisons between players on different teams, at different positions, or from different time periods. For all its faults, WAR at least serves as a reasonable estimate and a good place to begin discussion.

It is not, however, the end of the discussion. I have been experimenting with a series of adjustments to baseline WAR totals which I feel lead to a more reasonable overall rating of baseball players throughout history, and will be presenting both those adjustments and some of the results over a series of upcoming posts. The adjustment I started with is a method of combining peak and career value into one number.

Melog Rankings: Post-Indian Wells 2017

So, this is later than I thought it would be. Through a combination of happenstance and procrastination, I failed to run fully updated rankings before Indian Wells started - and I would have put them up after the tournament began, except that the draw promised extensive madness and then delivered in spades, and it seemed like a good idea to let that play out.

Everyone knows the big story: Roger Federer sat out the last six months of 2016, then came back to start this year and has won the two biggest events played to date. Which is marvelous theater, of course, but it's accompanied by so many other things that I'm reluctant to discuss them in too much detail before we get to the numbers. So let's do that.