Basketball: OUA standings with SRS, not RPI

Analogously, Ontario University Athletics men's basketball has been like an Olympic event where it is a priori that one nation will always get the gold, two others will vie  for silver and bronze and everyone else is just hopeful of a personal best.

Carleton crushed it; Ryerson and Ottawa are 2-A and 2-B. It is not the best jumping-off point for showing the limitations of the RPI-offs and advocating for another computation, Simple Ranking System, which layers a team's margin of victory over its strength of its schedule.

(The best primer on SRS was furnished more than a decade ago on pro-football-reference.com.)
... every team's rating is their average point margin, adjusted up or down depending on the strength of their opponents. Thus an average team would have a rating of zero. Suppose a team plays a schedule that is, overall, exactly average. Then the sum of the terms in parentheses would be zero and the team's rating would be its average point margin. If a team played a tougher-than-average schedule, the sum of the terms in parentheses would be positive and so a team's rating would be bigger than its average point margin.  
It would be easy to find the Colts' rating if we knew all their opponents' ratings. But we can't figure those out until we've figured out their opponents' ratings, and we can't figure those out until. . ., you get the idea. Everyone's rating essentially depends on everyone else's rating.
This is pertinent to university hoops for a couple of off-the-top-of-one's-head reasons. The SRS uses point differential rather just the W's and L's. That makes sense in a conference with so much disparity in student populations and resources dedicated to basketball. Plus it can create incentive for a team to "compete like the game is tied," which is probably better for personal and physical development. Putting that another way: SRS addresses the imbalance mentioned off the top; if you play Carleton or another power toe-to-toe for three quarters; there's actually a tangible reward and not just a moral victory.

On Saturday night, Carleton defeated Toronto 100-36. Since RPI does not use margin of victory and uses strength of schedule, adding Carleton to their schedule lifted the Varsity Blues (11-8) into fifth place in the final standings. The only box to check off for U of T on Saturday was to taking the court, essentially.

On the other side of the nation's capital, Ryerson defeated Ottawa to finish ahead them in victories, but it had no effect on the standings. The Rams winning on the Gee-Gees' floor is symbolically huge since they have had issues winning in the nation's capital over the years. Those are glitches to fix.

Here is the playoff field for the men's standings, by RPI:
  1. Carleton .638
  2. Ottawa .600
  3. Ryerson .594
  4. Brock .563
  5. Toronto .519
  6. Nipissing .515
  7. Laurentian .512
  8. Windsor .481
  9. McMaster .479
  10. Laurier .470
  11. Queen's .469
  12. Lakehead .452
In contrast, Simple Ranking System -- admittedly one without a cap on margin of victory to cut down on the impact of blowouts -- evaluates the performance a bit differently.
  1. Carleton 32.5
  2. Ryerson 19.3
  3. Ottawa 15.8
  4. Brock 4.9
  5. Laurentian 0.7
  6. McMaster -0.7
  7. Toronto -1.4
  8. Nipissing -2.5
  9.  Windsor -4.6
  10. Queen's -5.4
  11. Lakehead -5.4
  12. Western -5.9
You see what happened here, besides Julie taking credit for the Big Salad?
  • Ryerson and Ottawa trade places in second and third. The reality is that it might not matter since their play-in game for nationals, the OUA semifinal, is on a neutral floor. Theoretically, though, there's an influence on their first playoff opponent which could make a big difference.
  • Laurentian (seventh seed) and Toronto (fifth seed) trade places. One cherry-picked compare-and-contrast is that while Toronto's score against Carleton resembled the blood-pressure count of a very unhealthy person. The Voyageurs were at least competitive the first time they played the Ravens, losing 88-71 at home on Nov. 26 and even 'winning' a quarter. The margin went from 17 points to 47 in the rematch, which was clearly all the doing of Joe Rocca. All of it, all Rocca. Look at his stat lines. Please? 
  • McMaster goes from ninth place to sixth, which would earn a home playoff game. 
  • Laurier, which had a 9-11 record in the conference but lost by 20 or more points five times, finished with a minus-8.2 SRS, in 15th place. 
The women's league, which is more balanced at the top end, would seem to have got the upper crust more or less right, apart from the fact that 18-1 Queen's is the No. 1 seed ahead of also 18-1 Carleton, which won the teams' head-to-head matchup. The top nine teams by win percentage are also the top nine in RPI.

  1. Queen's .612
  2. Carleton .601
  3. McMaster .592
  4. Windsor .568
  5. Ryerson .550
  6. Ottawa .540
  7. Laurier .532
  8. Brock .482
  9. Lakehead .481
  10. Toronto .477
  11. Western .472
  12. York .450

How do they rank by SRS?

  1. McMaster 19.3
  2. Windsor 16.0
  3. Queen's 15.7
  4. Carleton 14.6
  5. Ryerson 9.7
  6. Laurier 4.9
  7. Ottawa 4.5
  8. Lakehead 2.8
  9. Brock -2.9
  10. Toronto -3.8
  11. Western -4.6
  12. Waterloo -7.7

Wow. So three and four swap with one and two, respectively.

  • McMaster, which had the best raw point differential, finishes first. Their blowout road loss against Carleton and last-second loss to Queen's only slightly detracted from the overall body of work, so the coaches are right after all to rank them No. 2. 
  • Yes, Virginia, Queen's is freaky clutch against tough competition. Their raw point differential (17.0 points) was lower than that of Carleton (17.2), but they had higher-quality opponents. 
  • Lakehead goes from a down-and-back flight to play Brock to getting the game at home and saving a ton of cash.
  • Waterloo makes the playoffs. Whether they would stand a chance against Ryerson is immaterial; they still got deprived status that could help with recruiting.
For all the impugning of RPI, it gets teams into the right range, but doesn't sort them neatly.

To borrow a Bill James analogy, it is akin to how it would be impossible to tell the difference between a batter with a coveted .300 average and one with a decent .275 average -- a difference of one hit every 40 at-bats -- if there were no statistics kept.

The larger point is that as a sport conference, the task is to send the best representative(s) to the national tournament. There is no perfect system and the best team is not always going to win in a "best of one" format. Adding machinations that could trigger a conference's two best teams being seeded first and fourth (talking about Canada West) defeats that purpose, though.

A two-division format also works against sending the best teams. That's a step back; it probably hurt the exposure of the OUA when Ryerson got roadblocked by Ottawa in 2013 and '14, which led to there being a final four without the host school.

Switching to SRS would be more of a way to go.
Next PostNewer Post Previous PostOlder Post Home

0 comments:

Post a Comment