Pittsburgh Pirates: MLB Needs to Re-Work Their Arbitration Process
By Noah Wright
With the upcoming CBA, there have been some talks on reworking how players entering arbitation should be valued and paid. But the current system needs reworked.
MLB’s arbitration system is very flawed. With the CBA set to expire this week and talks about the new CBA heating up arbitration has already been a talking point. The arbitration system is one that has a huge impact on the Pittsburgh Pirates and the rest of baseball.
Currently, the league is proposing that arbitration would be based on WAR, rather than the current archaic stats that are used in arbitration. But neither proposal is a good idea.
Right now the arbitration system for MLB is very subjective. While arbitration will always be subjective, there are many steps MLB can take to make it more objective. Right now, stats such as RBI, pitcher win/loss record, home run totals, saves, errors/fielding percentage, etc. are still considered when arbitration is still being done. The most glaring are RBI, pitcher record, and errors/fielding percentage.
If arbitration is to judge a player on their individual performance alone, why are so many counting stats used that can be swayed so much by the player’s surrounding team being used? RBI’s are one of the biggest examples of this. Last season Josh Bell had 88 RBI with the Washington Nationals. Would he have that same RBI total if he were still batting in the middle of the lineup for the Pittsburgh Pirates instead of behind Trea Turner and Juan Soto (the ladder having an OBP above 50% from June 16th to the end of the season, which was 98 total games played)? Meanwhile, you have Albert Pujols still averaging 100+ RBI seasons from 2017-2020 despite having a sub-.700 OPS because Mike Trout is a god and averaged a .440 OBP in those 4 seasons.
That also means that leadoff men are never going to have high RBI totals. A leadoff guy will never have a pretty looking RBI total, even if he bats .500 with men on base. Sure, some guys will rack up 80+ RBI’s despite batting at or near the top of the order, or not on a very good hitting team, but those are more outlier cases rather than the trend.
Pitcher wins/losses are about as archaic as it can get. Baseball’s best pitcher, Jacob deGrom, has 32 wins across his last 91 games started, the 35th most in baseball since 2018. The top 5 in wins since 2018? Gerrit Cole, Max Scherzer, Charlie Morton, Zack Greinke, and Marco Gonzales. Now while that’s not to say that those 5 aren’t very good pitchers, they pale in comparison to deGrom. Plus, deGrom won back-to-back Cy Young awards with only 21 wins between the 2 campaigns, one fewer than Dodger lefty Julio Urias this past season.
Then we have errors and fielding percentage. Good fielders make the more difficult plays. There is a higher chance of a guy like Nolan Arenado, Matt Chapman, or Trevor Story to make an error than Xander Bogarets or Rafael Devers. I think the four players with the most errors this year perfectly encapsulate why errors are very poor judges of defensive ability. Bo Bichette, Rafael Devers, Isaiah Kiner-Falefa, and Gleyber Torres were top four in errors. Bichette and Kiner-Falefa might not be Gold Glove level, but they’re at least average when it comes to defending their positions. Meanwhile, Torres and Devers might be the worst defenders at their respective positions.
Wil Myers may have not made an error this year in right field, but is he truly better than someone like Mookie Betts or Aaron Judge? The answer? Absolutely not. Ozzie Smith (25th best of all-time among SS) has a worse fielding percentage than Freddy Galvis (3rd best of all-time among SS). Brooks Robinson made 10+ errors in all but one season he played 100+ games at third base. Neil Walker ranks 18th in fielding percentage while Bill Mazeroski doesn’t even rank top 75. Neither is considered the best defensive second basemen in the franchise’s history according to fielding %. That would belong to Freddy Sanchez. According to fielding percentage, Jordy Mercer is the best defenisve shortstop in the Pirates’ franchise history.
There are about ten dozen other examples of how fielding percentage and errors are a bad reflection of a fielder’s ability. It also comes down to a scorekeeper’s own decision, which can be different from person to person.
So MLB needs to leave behind the old-timey stats for new stats and WAR (wins above replacement) is probably the most popular cited sabermetric in today’s game. It’s an all-encompassing stat that takes into account everything; hitting, base running, fielding pitching. It’s also positionally and league adjusted. It’s also the stat that MLB has considered implementing into arbitration talks.
But it’s also one of the more misunderstood ones. The big reason why WAR and only WAR shouldn’t be used to judge players’ salaries is that there is no defined WAR.
There are two ways of looking at WAR. There’s bWAR, which is Baseball Reference’s calculation, and fWAR, FanGraphs’ calculation. So what’s the difference? For pitchers, bWAR is based on runs-allowed-per-9. Pretty simple, how many runs, both earned and unearned)\, the player allowed in total on average. fWAR on the other hand is based on FIP. FIP, or fielding independent pitching, takes how good the pitcher was at limiting the three-true outcomes (walks+HBPs, homers, and strikeouts). Take, for example, Wade Miley. Miley this past season had a 6.0 bWAR as he had a strong 3.53 RA9. However because his FIP was a solid, but unimpressive 3.97, he had an fWAR of just 2.9, essentially half of his bWAR.
For batters, WAR is calculated with batting runs, fielding runs, baserunning runs, adjusted for position, the league environment, and runs per win. But both FanGraphs and Baseball-Reference have two different ways of calculating it. Both weigh certain things differently. For example, in 2018, Bryce Harper had a 1.8 bWAR, but 3.5 fWAR. This is mainly because bWAR weighs defense differently than fWAR. Despite posting a strong .249/.393/.496 line and having a 135 wRC+, Harper was an awful defender with -13 DRS. If you heavily favor defense, then bWAR might be what is cited. If you think defense is important, but not to the same level as offense, then fWAR is the way to go. But the point remains that it can come down to personal preference.
Now that’s not to say that WAR isn’t a good statistic. It’s a good benchmark test when looking at players. But there can be so much difference in fWAR and bWAR that it’s hard to make a good argument to use the stat to determine salaries. Sure, it might be better than RBI, errors, wins/losses, etc., but it still doesn’t solve the problem of valuing players’ performances with one, concise number. When it comes to how much a player makes, it can make things foggy.
But it’s not like we don’t have the statistics that can accurately tell us how good a player was. wOBA is a more advanced version of OPS. It essentially gives weight to every outcome a batter can have (walk, HBP, single, double, triple, home run), but on the scale of OBP. wRC+ is probably the best stat that we have to evaluate how good a player was when it comes to hitting. Based on wOBA, it normalizes the player’s numbers on a scale of 100 (100 is average, 101 is considered 1% above average, 99 is considered 1% below average, etc) while also taking into account the league run environment, as well as the home playing environment. It knows how to adjust for a player with a hitter-friendly/pitcher-friendly ballpark. It will reflect a guy with noticeable home/away splits. It knows how to adjust for when the league has a juiced ball problem. wOBA and wRC+ shouldn’t be the only stats taken into account. OPS is still a solid measuring stick, as well as any rate stats.
For pitchers, the common rate stats like ERA, FIP, and WHIP should always be considered, but there are also stats like wRC+ which normalizes a pitcher’s performance based on the run environment and their home field like ERA+, ERA-, and FIP-. For ERA+, it is similar to wRC+ where 101 is 1% better than average and 99 is 1% worse than average. ERA- and FIP- are pretty much the complete opposite where 101 is 1% worse than average and 99 is 1% better than average.
Most advanced defensive statistics aren’t as accurate as offensive statistics, but they’ve gotten better as technology has advanced. Outs above average are as accurate as we get because it uses Statcast. OAA takes into account about everything you can think of: how far a fielder had to move to get to a ball, how much time he had to get there, exactly where the fielder was positioned (which takes into account for shifts), how far he is to the base the runner is going to, even how fast the runner is on average (on force plays and putouts).
When it comes down to it, MLB needs to fix the way players are evaluated through the arbitration system to make it much less subjective. WAR can be subjective in terms of what you value and what site you prefer. On the other hand, stats that can be affected by the rest of the team’s performance, like pitcher wins/losses and RBI’s, shouldn’t be taken into account either as it isn’t fair that a player can be penalized. They didn’t reach 80 RBI because they batted near the top of the order, or were on a team that didn’t get on base ahead of them.
So what should be used instead? I’d suggest for batters, the rate stats such as the triple-slash (AVG/OBP/SLG), as well as wOBA, and wRC+. That way, you keep around some tradition with the triple-slash, as well as encapsulating how that player did given the overall environment. For pitchers, stats like ERA, FIP, and WHIP should be taken into account, but the stats that normalize a player’s performance such as ERA+/-, and FIP- are also very important. To measure defense, since outs above average takes so much into account as well as accurately being able to track it through Statcast, it makes it much easier to know who is a good defender and who isn’t.