Wins
Teams can have an advantage when their schedule has more games. Currently, we’re showing two Rankings to amend this issue: an unadulterated ranking and a normalized ranking where all teams are normalized to having played 20 games. e.g. Culver has played 20 games and have 1,000 Points. Brunswick has played 10 games and has 600 Points. It appears that Culver’s Point lead is misleading because they’ve played twice as many games as Brunswick. So we normalize Brunswick’s number of games from 10 to 20 and their Points from 600 to 1,200. So a current, unadulterated ranking has Culver ahead of Brunswick with 1,000 and 600 Points respectively while the normalized Rankings has Brunswick ahead of Culver with 1,200 and 1,000 Points respectively. But to return to the first sentence, Teams can have an advantage.
Let’s say Culver and Brunswick are both 10-0 and both played 10 teams with 1,000 Points each. It makes sense that Culver and Brunswick would be tied with, for example, 1,000 Points.
But what if Culver played ten more games against 1,000-Point teams than Brunswick? Now Culver is 20-0 with 2,000 Points while Brunswick is 10-0 with 1,000 Points. If we normalized their Points, Culver and Brunswick would be tied. But, logically, Culver deserves to be ahead of Brunswick. While Culver did have the advantage of playing ten more games than Brunswick, putting the two teams side-by-side, Culver had a better season. Even if Brunswick is restricted by Connecticut Prep School Law where they can only play 10 games per season, Culver still had a better season and should not be penalized by this theoretical CT Law. In this case, normalizing their number of games does not make sense.
What if, instead, Brunswick played ten 1,000 Point teams to be 10-0 with 1,000 Points while Culver played twenty 500 Point teams to be 20-0 with 1,000 Points. Brunswick beating ten 1,000 Point teams clearly demonstrates that they had a better season than Culver beating twenty 500 Point teams. Yet, by the unadulterated Ranking, they are tied with 1,000 Points. In this case, we’d want to normalize their Points so both teams would be measured as if they played the same number of games. With the normalized Ranking, Brunswick would have 2,000 points while Culver would have 1,000. And this normalized Ranking would appear to be a more accurate assessment of the relative performances of Brunswick’s season vs. Culver’s season.
Normalizing or not normalizing the number of games that each team has played that season is substantially less about creating an even playing field due to the number of games played by each team and substantially more about teams, whether intentionally or not, padding their schedule.
Some teams effortlessly place teams with a lot of points on their schedule. Other teams struggle to to have even one game against a team with a lot of points. None the less, when evaluating a team that is 10-0 against 1,000-Point teams and another that is 20-0 against 500-Point teams, we need to be cognizant that the former team’s season is substantially more impressive than the latter’s season and their relative Points should mirror this. Even when a team plays ten 1,000-Point teams and ten 500-Point teams while another team plays fifteen 1,000-Point teams and five 500-Point teams, we need to be cognizant that the latter team’s schedule is more impressive. This is one of the reasons why we created the Win Bonus. The team that beats a 500-Point team will receive a small bonus for their win while the team that beats a 1,000 Point team will receive a large bonus. And, as a consequence, the relative lack of impressiveness of the former team’s win over a 500-Point team is mirrored within the standings while the impressiveness of the latter team’s 1,000 Point win is also reflected. And, if teams never lost their games and, for all intents and purposes, losses were inconsequential in trying to evaluate a team’s performance, our system would be complete. But this is not the case. Thus far, we’ve only discussed undefeated teams and wins. What about the vast majority of teams that finish the season with at least one loss.
Losses
Currently, when a team loses, they give a “10% Ante” of their current Points to the winning team. And, if every team played schedules with the same, if not extremely similar levels of competitiveness, this “10% Ante” system would be enough. (10% Ante chooses to emphasize rewarding wins more than punishing losses. So, while we have a Win Bonus, we do not have a Loss Bonus.) But, like in the previous examples, there are teams that primarily play numerous 200-Point teams, though in small increments, every game, these teams accumulate points, and, more importantly to what we’re discussing, have little, if any, chance of experiencing a loss. On the other hand, other teams are playing schedules where the majority of games are against 1,000-Point Plus teams and their easiest game is against an 800-Point team. And, while the team with a more challenging schedule, will fly to the top of the standings if they go undefeated, most teams lose and, the more challenging the schedule, the more losses they will incur. And, while the Points of these teams with multiple losses may be accurate relative to other teams with equally challenging schedules, their Points are not accurate relative to the teams who will all but certainly never incur a loss. The team with a less-than-challenging schedule starts the season 0-0 with 1,000 Points and finishes the season 20-0 with 1,800 Points. In theory, the #7 team in the nation starts the season 0-0 with 1,000 Points, losses to the teams ranked #1 through #6, beats the teams ranked #8 through #11, and finishes the season 4-6 with 915 Points. Obviously these teams’ points do not mirror their actual value and rating.
Starting Point
Currently, every team starts with 1,000 and, before Game One, this makes sense. Every team should start from the same starting line. But as the teams build their resumes, this Starting Point needs to be adjusted. The Starting Point for each team should be the Number of Points that best mirrors the team as dictated by the results the team has earned during the relevant season. Over the course of the season, each team will rise above and sink below this Starting Point. But this Starting Point needs to be unique to each team. Otherwise, teams with weak schedules will have undefeated seasons against 200-level teams and, even though the relevant team is actually a 300-Point team, their Points within the 10% Ante Ranking will be 1,400 Points (1,000 + 400). If a team is undefeated but only plays 200-Point teams, we know this team is at least a 300-Point team. They may be a 500-, 600-, or even a 2,000-Point team. But based on this team’s current season, all we know is that they are at least a 300-Point team. And their Starting Point should reflect this. Similarly, teams with strong schedules will play numerous 2,000-Point teams and lose as well as numerous 1,900-Point teams and win and, due to their losses (to strong teams), their change in Points from Game One to Game Fifteen will be negative. With a Starting Point of 1,000-Points, this team within the 10% Ante Rankings will be a 900-Point team when, in reality, based on their wins and losses, they are a 1,900-Point team.
As a consequence, 10% Ante is changing three things:
1- Loss Bonus – Currently, when a team loses, they give 10% of their Points (the “10% Ante”) to the winning team. Going forward, the losing team will continue to give their 10% Ante to the winning team plus… There will be a Loss Bonus (which is paid by 10% Ante, not the losing team). This Loss Bonus will increase or decrease based on the relative ratings of the two teams. Using an exaggerated example, if a 1,000-Point team loses to a 200-Point team, they will give the 200-Point team closer to a 20% Ante. Alternatively, if the 200-Point team loses to the 1,000-Point team, the 200-Point team will give the 1,000-Point team closer to a 1% Ante.
2- Win Bonus – Currently, when a team wins, they win 10% of the losing teams points as well as a Win Bonus that is based on the losing team’s Points. Going forward, the winning team will continue to earn the 10% Ante from the losing team and the winner will also earn a Win Bonus (which is paid by 10% Ante, not the losing team) and… This Win Bonus will increase or decrease based on the relative ratings of the two teams. Using an exaggerated example, if a 1,000-Point team beats a 200-Point team, the 1,000-Point team will earn 10% of the losing team’s Points as well as a Bonus that will be closer to 1% of the losing team’s Points. Alternatively, if the 200-Point team beats the 1,000-Point team, the 200-Point team will earn 10% of the losing team’s Points as well as a Bonus that will be closer to 10% of the losing team’s Points.
3- Starting Points – Going forward, each team’s Starting Point will be the average of the Points of each team’s best three wins. It’s more difficult and telling to beat a team than to lose to a team. If a team can regularly beat 312-Point teams, this team is probably at least a 313-Point team. If a team regularly beats 1,942-Point teams, the team is probably at least a 1,943-Point team. We are using this Three Best Wins for each team’s Starting Point for this reason.
