How and why the salary calculation is changing

We’re making some changes to the way player salaries are calculated in Parity League this year. In the interest of transparency, we wanted to take this chance before the season started to make sure everyone (who cares) understands the differences and our reasoning for making the changes. Check out Parity Podcast episode #402 for more discussion!

For context, we started by asking ourselves these four questions about the current salary calculation:

• Does the salary calculation accurately reflect individual player value?
• Does the salary calculation accurately reflect the performance of a team?
• Is the salary calculation simple and intuitive enough for all players in the league to understand?
• Is the salary calculation spirited (i.e., does it celebrate the contributions of everyone in the league, does it motivate people to play better, is it equitable, etc.)?

Upon review, we felt there were outstanding balance issues with the formula that needed to be addressed this year. Specifically, we wanted to fix the recurring problem of strong, high-efficiency teams having lower salaries than weak, low-efficiency teams. Secondarily, we wanted to address the issue of high-touch handlers being disproportionately valuable relative to other types of players.

That being said, the current salary calculation has proven to be relatively successful, so we felt it was more appropriate to tweak the existing formula rather than restart from scratch. For the sake of simplicity and continuity, we also didn’t want to make too many changes at once, so we left several categories alone even though they’re probably worth discussing in the future (e.g., the value of second assists and D blocks).

With all that in mind, we are experimenting with three changes this season, listed here from least to most impactful:

• Reduce the penalty for drops from $5,000 to $4,000 and reduce the penalty for threw-drops from $2,500 to $1,000. This is the smallest change we’re making. The average player only has 5 drops or threw-drops over the course of a season, so the net salary impact of this change is very modest. Nevertheless, we think it’s important that the team value of a drop be brought in line with that of a throwaway. Previously, a good throw that was dropped downfield cost your team $7,500 overall, whereas a throw into the ground cost your team only $5,000. Now either scenario costs your team $5,000.
• Reduce the value of each catch and completion from $1,000 to $500. Although catches/completions correlate fairly strongly with individual performance (i.e., the players that touch the disc more on any given team tend to be the most valuable), the total number of touches by a team is a poor indicator of team performance. In the past, this has led to losing teams that pass the disc a lot going over cap, while winning teams that minimize touches can still end up under cap. Reducing the value of catches/completions to $500 each maintains some of the individual reward for being involved in the play while reducing the penalty to a team for playing a possession style of offense.
• Add a $1,000 bonus to each player who is on the field when their team scores a goal after starting the point on offense, or $2,000 if the scoring team started the point on defense. Players will now receive a small salary bonus every time their line scores (double the bonus for a break). We think this change better captures the contributions of players that don’t show up elsewhere on the stat sheet, especially on defense, while pushing up the net salary of high-scoring teams. We arrived at the $1,000/$2,000 values through experimentation with past data; any less and the effect on teams isn’t large enough to force meaningful player movement, any more and the scoring bonus distorts individual player values. We also experimented with a penalty for getting scored on, but the effect is not much more accurate and it would have the undesirable side effect of increasing the frequency of negative salary games.

We’ve tested the revised calculation with past years of Parity player data and the overall impact of these changes is modest (most players see an overall salary change of less than 10%), but we’re confident that taken together these changes will lead to a more balanced league overall.

At the team level, high scoring teams will now go over cap more quickly and more consistently, requiring them to trade away better players sooner. Consider the following scenario:

• Team A plays a run-and-gun style of offense that always scores in two passes
• Team B plays a possession style of offense that always scores in 10 passes.

Under the old system, team A receives $32,000 in team salary for each goal (1 goal + 1 assist + 1 second assist + 2 completions + 2 catches), while team B is hit with $48,000 per goal. Even if team A wins the game by as much as, for example, 25 goals to 20 goals, team B will have the higher team salary. Under the new system, team A receives $36,000 per goal (or $42,000 for a break) while team B receives $44,000 per goal (or $50,000 for a break). There is still an incentive to score more efficiently, but now it is much harder to win games (regardless of play style) without ending up with the higher team salary.

On an individual level, the players seeing the biggest reduction in salary are the “whales” whose salaries are inflated by touches alone. Conversely, the biggest beneficiaries of the changes are low-touch, high-impact players, especially lower-salaried women. Based on 2019 data, the average woman will see an 11% higher salary under this system compared to a 3% gain for the average man.

Although salaries are slightly higher on average, the reduction in catch/completion value makes it harder to achieve high-salary performances in individual games. We should not expect to see $100,000+ performances any more frequently.

Hopefully that makes sense to everyone and you’re as excited about these changes as we are! Thanks for reading and please don’t hesitate to post your questions or comments below.