|
Tell me if you've heard this one before: a newspaper reporter, a sportscaster, and an accountant all walk into a ...
Okay fine, you probably haven't heard that one before. But, this is a story of how those three professions intersected and helped create one giant idea that could help us look differently at athletics in Idaho.
It's customary of former athletes, like basketball players for example, to gather at a local gym or church building to play some pickup ball in an attempt to stay in shape (round is a shape, right?) and still see if we have some of the skills from our playing days. While we were 'warming up' (which mainly consisted of some minor stretching and Stephen Curry range 3-point shots), my good friend Dayton Robison and I got into a discussion about something we had both read online, which stated that the Washington Interscholastic Activities Association has released a set of 'Ratings Percentage Index' rankings, or RPI, ahead of their district tournaments. Dayton and I went back and forth about the pros and cons of Washington's RPI format and then spoke hypothetically about what would happen if this were implemented in Idaho, including what certain criteria we would want in the formula if we were the ones in charge. Eventually, we hit the court and started getting some exercise.
Prior to the 2016-17 school year, the Washington Interscholastic Activities Association announced that they would be changing the way in which they qualified teams for their state basketball tournament.
Many felt that the system at the time, which basically consisted of winning one of your district's alotted seeds and then being paired at random with someone else who had qualified, wasn't working anymore. Too often, a matchup of highly ranked teams would occur well before the state championship game or state semifinals would occur.
This led the WIAA to move to an 'RPI' based system, which stands for Ratings Percentage Index, and would be used to seed the boys and girls basketball tournaments. To break it down in a way most folks could understand, a team's RPI consisted of 25 percent of their own winning percentage, 50 percent of their opponents winning percentage, and 25 percent of their opponents' opponents winning percentage.
What would then happen is that teams would try to qualify for the state tournament, much in the same manner as they do here in Idaho. A certain number of berths were awarded to each district and the teams would play a district tournament to try and secure one of those berths. Then, once all district tournaments from around the state were completed, the WIAA used the regular season RPI rankings to seed the state tournament. For example, in Washington they play a 16-team tournament and the top ranked team in the RPI that had earned a berth would play the lowest ranked team in the RPI that had earned a berth. This would go on in the same manner for the other teams, matching up No. 2 vs. No. 15, No. 3 vs. No. 14 and so on and so forth. Think of that bracket looking a bit like one of the regionals for NCAA March Madness.
Fast forward now to February 8, 2017.
Responses from coaches to Nelke varied. Some thought it would be a great idea, others weren't so sure.
“I think it’d be great,” Lake City boys basketball coach and athletic director Jim Winger told Nelke back in February. “Numerous times in the older days, we were No. 1 and Pocatello was either undefeated, or one loss to Highland at regionals, and we drew Pocatello in the first round. If there was an RPI system it would solve that, because there always seems like, especially in Boise, some weird things that go on at district tournaments.”
Post Falls boys basketball coach Mike McLean saw both sides of the argument.
“As far as Idaho goes, I think our state tournament setup is OK,” McLean said to Nelke. “The preset tournament works pretty well until a league champion gets beat in a regional tournament. I do think the state can make it better, and I am not against seeding the tournament after the teams qualify.”
“One problem I have with the way it is now is for the teams that end up getting second, third and fourth place is so luck of the draw. The way it is now, I feel that there is no distinction between second, third and fourth place. All three teams go 2-1 and their placing is almost completely dependent on the side of the bracket a team gets. With the current system it is very unlikely that the best eight teams are in the state tournament every year. Last year and this year District 5 has two of the best eight teams in the state and one won’t be at the tournament. We get two teams this year (in District 1) and probably shouldn’t. But back in 2010 Priest River won state, Kellogg got second and we didn’t get to go even though we beat Priest River once during the season and beat Kellogg three times that year."
There are many sides to the issue with many people in the same geographical area taking different sides to those issues. As Winger put it in his remarks to Nelke, "You draw it out, and you play who you play — you’ve got to beat everybody to win a state title anyway. (But) these (RPI) systems, at least, reward you for what you’ve done for four months, instead of in one week at a tournament.”
Skip ahead a few days now to February 11, 2017.
I was in Arimo to call the 'if necessary' game of the 3A District 5 girls basketball tournament. It had come down to No. 1 seeded Snake River playing No. 2 seeded Marsh Valley. Both teams had been ranked in the top 5 all season long in the state media poll, IdahoSports.com poll, and coaches poll. During the regular season, the Panthers beat the Eagles by 16 and 13 points in their two contests. In the first round of the tournament, Snake River took down Marsh Valley for a third consecutive time, beating the Eagles by 10 points. However, in the game that would have clinched Snake River the district title, they lost 51-44 due to some missed free throws and errors in the final moments. That sent the tournament into the 'if necessary' game which would be hosted by Marsh Valley.
That final game of the district tournament started out close and tense. The fans knew what was at stake. So did the players and the coaching staffs. After a see-saw affair throughout much of the game, Marsh Valley pulled away in the end and took home the district title in a 59-45 win, their third straight conference championship.
During the championship presentation, Marsh Valley head coach Kyle McQuivey told the Snake River team "it's a sad thing that we've both been four and five in the rankings in the state and that one of us has to stay out." His words resonated with many in the crowd afterwards, many of whom talked about it with me following the conclusion of the game, including many Marsh Valley fans.
Later that night, while making a quick pit stop at the local Taco Time, some fans from Snake River said to me "that article about the Washington RPI... what will it take for that to happen in Idaho?" Other comments were given, all along the lines of "is the system that we have in place working for us right now?" Of course, it's much easier to talk about it when your team get the short end of the stick. But the article that Nelke wrote and I had retweeted on my Twitter account was resonating with people not only in north Idaho, but now other parts of the state.
Jump ahead five days to February 16, 2017.
I'm now in Middleton calling games for the 3A Girls Real Dairy Shootout on IdahoSports.com with Lance Taylor. That night, as I was preparing for the four games I would call the next day, Dayton sent me a message with a couple of excel spreadsheets attached to it.
"Here's a preliminary look at an RPI model in Idaho," he said. "Used the same model that Washington did."
He had went through the entire boys basketball season in Idaho and had put together a basic RPI spreadsheet with every result from around the state accounted for. Dayton, who is a 2005 graduate of Madison High School and an accountant by trade, made the complex seem simple. It also helped that he knew Microsoft Excel inside and out.
"Those are some pretty interesting numbers," I quipped, knowing that I had to keep prepping for the next day while still really wanting to take a deeper dive into the spreadsheet. Fortunately for me, Dayton knew I was slammed (he was entering his busy season with taxes too) and we both agreed to take a look at the numbers closer once we both had a bit of time.
Fast forward now to February 26, 2017.
It's the day after the boys basketball state play-in games. I get another message from Dayton.
"I've got everything run through the end of the district tournaments," he said. "Want to take a look at the spreadsheet?"
Uhhh no, I don't want to. OF COURSE I DID! I wanted to see how the RPI numbers stacked up against the actual matchups we would see in the boys basketball state tournaments. The teams would be seeded much the same way they are in Washington. The results were certainly surprising.
I thought that for sure there would be maybe half of the first round matchups correct. It wasn't even close. Only three of the 24 first round matchups were the same as our rankings. Of the three that matched up, two of them were in Class 5A (Madison vs. Centennial and Rocky Mountain vs. Bonneville) while the other was in Class 2A (Firth vs. New Plymouth).
When it comes to the seeding of the tournament in the RPI format, it's to seed those eight teams that earned a berth based on what they accomplished during the regular season. It adds more meaning to the regular season as well. When looking at the top two ranked teams in each classification that qualified for state, only two classifications (2A and 1AD1) had their top two teams play each other in the state title game. The others either were upset in earlier rounds (as was the case with Carey, Weiser, Shelley, and Post Falls) or they were paired on the same side of the bracket (as was the case with Preston and Bishop Kelly).
Now, that doesn't mean that these rankings were way off base or that the state's current method of bracketing is wrong or anything like that - they are just two different systems of bracketing teams for tournament play. While Dayton's final rankings didn't pit any teams from the same conference in the first round, that doesn't mean it can't happen - and it's something that the Idaho High School Activities Association Board of Directors is aware of.
“I actually like the way that we seed the tournament now because it guarantees that the first- and second-place schools from each district are on opposite sides of the bracket and could meet in the finals,” said IHSAA executive director Ty Jones in Nelke's article. “Our board likes it that way too. If we changed the way it is seeded those schools could end up playing in the first round. I would not be in favor of a system where, as an example, Coeur d’Alene and Lake City meet in the first round of the tournament after having just met in the district finals a week or so earlier.”
Certainly, in an RPI-based format a rule could be put in which prevents teams from the same conference matching up in the first round at the state tournament. How easy it would be to draw up the guidelines of such rule depends on who you talk to.
It's also not uncommon to make changes to the general guidelines of the system itself. After their first year using the RPI formula, the WIAA opted to change the guidelines to include results of district tournament contest and out-of-state opponents winning percentages. The chairman of the WIAA's RPI committee, Greg Whitmore of Lind-Ritzville High School, told The Herald of Everett Washington that "After year one, we were really pleased with how it went. It wasn’t perfect and we’ll never get it to be perfect, but it was really pretty good. So we (tweaked) a couple things, and the changes we’ve made for next year, we’re pretty confident, will make it even more accurate and better.”
That can be looked at as an advantage to the RPI system, in that changes can be made to be more accurate. It's something that the Colorado High School Activities Association had to do this past school year, when controversy embroiled them after their rankings omitted an eight-win Doherty High School team from the 5A playoffs in favor of a pair of three-win teams. Loveland High School was left out of the 4A playoffs despite posting a 9-1 record. The CHSAA voted unanimously to change their formula, according to the Colorado Springs Gazette.
Let's be honest for a moment: if there were a perfect system, all 50 states would be using it right now and you wouldn't be reading this article. But having a system that allows for tweaks and changes is certainly something that could be examined further.
And that's where today comes into play. After more discussions and tweaks and changes, Dayton has set up his database for the 2017 football season... dubbed the 'Robison RPI' or 'RRPI' in short.
"One of the biggest complaints, I don't know if you can call it a complaint, that I hear about the current playoff system in Idaho is that it isn't seeded," Dayton said. "Often times you can have the top 2 teams in the state facing off in the semi-finals, because the brackets are drawn up before the season even starts. With a 'seeded' playoff bracket you can avoid that scenario."
He added that he didn't feel much needed to change with the current format of district tournaments and the number of berths they were awarded by the state, but that the RPI could enhance that process through re-seeding.
"I don't believe that you need to use the RPI to determine which teams qualify for the playoffs like some states do, but it can even up the brackets once teams qualify for the playoffs. In my mind, by keeping the current qualification process in place you allow teams to know what the set criteria is to qualify for the playoffs. You finish in the top two (or whatever the number is) of your league you're in. Compare that to not knowing if you will qualify or not - I'd hate to see a team go 8-1 or 9-1 and get left out of the playoffs because they played in a weak league, like what happened in Colorado last year."
This still allows for the drama of the district tournaments - meaning that even the top-seeded teams aren't safe. They still have to play their way into the state tournament by either winning the tournament or grabbing one of the other berths available to their conference. A team wouldn't be able to get into the tournament via an 'at-large' berth that's based solely on their RPI ranking.
So what is the formula of the Robison RPI?
"Essentially the base for the formula is the Washington Preps RPI formula, which looks at a team's winning percentage, their opponent's winning percentage, and their opponents' opponents' winning percentage. These are weighted 25%, 50%, and 25% respectively," he said. "This essentially is looking at the strength of schedule that a team plays. So two teams with the same 6-3 record will have different RPIs depending on their schedule strength."
Dayton's formula also factors in the records of out-of-state opponents, something that the WIAA formula did not until changes were made to it in April.
"I'll have the records of the 60-plus out-of-state teams that Idaho teams will play this year factored in," Dayton said. "The out-of-state teams' opponents record will be weighted at 0.500 in the formula."
The Robison RPI will also have additional kickers for cross-classification matchups and road wins.
"Cross-classification games will be weighted by 15 percent," he said. "That means that if a 4A schools beats a 5A school, that win will count for 1.15 wins. Likewise if a 5A school beats a 4A school, the win will be discounted from 1.00 to 0.85 wins." He added that losses in cross-classification contests would be weighted no differently than a standard loss against a team from the same sized classification.
The long and short of it is that scheduling matters when it comes to an RPI-based formula. If you want to have a good strength-of-schedule, you also have to schedule good teams and not 'cupcakes'. It also means that you are also rooting for your opponents to win on every single night except the ones when you play them. A better record for your opponent (and your opponents opponents) will factor in to an eventual higher ranking.
As for road wins, Dayton added that there would be more weight given to those games overall.
"Winning on the road is never easy in any sport. So if a team wins on the road they will be given an additional 5 percent weight. If that road win is against a team with a .6500 record or better they will be given an addition 2.5 percent weight. So if a team beats a really good team on the road they could receive an additional 7.5 percent weight to their win." He added that it's important to note that there is no penalty or discount for teams losing on the road.
As for games that get out of hand and end up being blowouts, Dayton said that would not play a factor in his RPI formula.
"As of right now there is no added kicker for blowing out an opponent (or getting blown out). So a Mountain View 30-point win against Rocky Mountain is the same as a Borah one point win against Rocky Mountain. It may be possible to add that in down the road, but as of right now that isn't a factor."
Just like with any data set, the more data available means the more accurate a certain projection or, in this case, ranking will be.
"The rankings may take 2-3 weeks to reflect what we see on the field. There won't be a lot of data the first weeks to measure teams accurately (versus what we see on the field). As with any rankings system the more data you have to look at, the more accurate it will be," he said.
These RPI rankings are sure to pique the interest of high school fans throughout the Gem State. We will keep tabs on these rankings throughout the season and see how they line up against the actual tournament brackets, the coaches poll, and the state media poll.
It's an experiment born of an accountant and a sportscaster, all because we read a reporter's newspaper article... and, let's be honest, one of us knew how to use a spreadsheet too.
|