Tag: OSI

by Mike Pavlichko

What would happen if NJSIAA dropped power points altogether? With one tweak: not much

As the NJSIAA continues to tweak the playoff formula year after year, New Jersey’s high school sports governing body has said it’s committed to power points. The question is: Why?

This coming school year, in all sports, the NJSIAA eliminated the group point component of power points. Traditionally, power points have three factors:

  • Teams get 6 “quality points” for a win, 3 for a tie, none for a loss.
  • Teams (used to) get “group points” based on the size of a school they beat (1 through 5 in football)
  • Teams get residuals: 3 for every win by an opponent they beat, 1 for every win by an opponent they lost to. (Ties make it a bit trickier.)

But now, the group points are out, which means other than the numbers, there’s no real difference between OSI and power points. Both give credit for a win – in OSI you get the full value of a team you beat, half for a loss and 75% for a tie – and both have a strength of schedule component; in power points, its residuals, while OSI itself is the strength factor.

So the NJSIAA has neutered power points in football, making it somewhat toothless. In addition, they moved the multiplier from power points to the OSI formula.

They also brought the values somewhat closer to reality. With power points, a loss to a mutliplier would gain a team more points than they could get by beating a large, undefeated public school. But with the move to OSI, beating a mutliplier gets you the same point value as a beating a public school or non-multiplier, while losing gets you anywhere from 70 to 80 percent of the OSI value, as opposed to 50 percent.

So, where’s the benefit of keeping power points.

We evaluated the 2022 playoff teams a year ago. Out of 160 public schools qualifying for the playoffs (non-publics are seeded by a committee with the metrics used only as a guide) only seven different teams would have made the playoffs. Generally, the teams moved very little.

We just got around to looking at the 2023 playoffs. And a move to OSI would be even better.

Only six different teams would have gotten into the top 16 out of 160 overall. The negative side of that is that five of the six would have beat teams with worse records.

Further, two of those newly-qualified teams were 2-7. Now, we’ve never liked the NJSIAA eliminating the .500 requirement. But, a simple increase in the minimum number of wins from two games against NJSIAA member schools to three would take care of that problem. In that case, a switch solely to OSI would net only three – yes, three – different teams in the playoffs than with the state’s UPR formula, with is made of 60 percent of a team’s OSI ranking, and 40 percent of a team’s power point ranking.

This year will be interesting to watch, with the elimination of group points and the multiplier being moved from power points to OSI, and with more realistic values. That could be the tipping point that gets OSI more in line with the UPR, so close that power points may essentially be worthless.

Many coaches still believe in power points, and the downside of OSI is that strength of schedule is based more on scores (read: performance) than wins and losses. One idea would be to add some sort of residual factor, based on wins, to the OSI formula.

For example, if Team A beat Team B with an OSI of 80 and a record of 6-2, they would get 80 for the win, and add 18 residuals (3 for every win by a team you beat). Losing to that team – which would now be 7-1) would get you 47 points: that’s 40 as half of 80 for the loss, and 7 for their seven wins.

Here’s a look at each supersection from the 2023 playoffs, and how they would have looked if the playoffs were seeded strictly by OSI – no residuals added, maybe we’ll check that another time – instead of power points:

North 5: The top five teams here would stay the same, with Passaic Tech followed by Phillipsburg, Union City, Watchung Hills and Montclair. P’burg and Montclair made sectional finals, while PCTI beat Union City in the North Group 5 semifinal. Paterson Eastside (3-5) would have climbed from 17 to 15 and made the playoffs, knocking out 16th-seed Bayonne (5-4). But they would be one of three 3-6 teams that made the playoffs either way, the other two being Clifton and Bridgewater-Raritan. Piscataway (5-4) would miss the playoffs either way, although they qould have qualified strictly on power points. Overall, only two of the top 17 teams moved more than two spots, and they only moved three places.

South 5: The top 16 remain the same either way, OSI or UPR. South 5 semifinalist Cherokee would remain No. 1, while Toms River North, the overall Group 5 champ, would have fallen two places to fourth. Does that sound right? Maybe, considering they finished as high as No. 2 in the UPR mainly based on playing two multipliers during the regular season: Red Bank Catholic and Donovan Catholic. The top four teams were the same, just in a slightly different order.

North 4: This is another supersection with little change. The top 3 wouldn’t have changed their order – North 4 champion Ramapo, sectional finalist Roxbury, and North 5 semifinalist Mount Olive. The other sectional finalist, Northern Highlands, would have flip-flopped with Ridge, jumping up to fourth. Overall, ten of the 16 teams didn’t budge with the change.

South 4: The top four teams would be the same, but while the top two would remain sectional finalist Millville and South 5 semifinalist Winslow Twp., seeds three and four would flip-flop, putting South 4 champ Mainland third and dropping Brick Memorial to fourth. The other sectional semifinalist was Long Branch, which made a nice run. They would drop two places from 10th to 12th. Overall, seven of the 16 remained unchanged, and no new teams made it; the top 16 was the same either way.

North 3: Summit would have dropped to the second seed overall, but they still would have been in the top two, meaning they still could have hosted a sectional final if they got that far. (They didn’t.) Old Tappan would flip-flop with them and be No. 1 overall; they made it to a sectional final. North 3 champ West Essex would hold at No. 8, while North 3 finalist West Morris would jump from 11 to 9. Passaic Valley, the other sectional semifinalist, drops one spot to 13th. This is one section where it doesn’t work out well, with Governor Livingston – which had a nice year at 6-3 – would get bounced using OSI. They made it at No. 16, but would fall all the way to 19th with OSI instead of UPR. With six wins, though, maybe they would have benefitted from moving residuals over to OSI? The new team that got in? Mendham, which was 4-5, not an egregious switch, to be fair. Only one three-win team qualified – Sparta, with a 3-5 record – but they would have made it either way.

South 3: Here’s one where a bad team would have made it over a good one, at least on the surface, based on wins and losses. Burlington Township (2-7) was 17th in the UPR standings, but would have made it at No. 13 going straight OSI. That would knock out 6-2 Hopewell Valley. But, if the NJSIAA raised the minimum win requirement to three (which we think is fair because it takes 3 wins to clinch a sectional title) Burlington Twp. would be out. Hopewell Valley still wouldn’t get in, but at least the others would be Ocean Twp. (3-5) and Nottingham (4-4). Otherwise, only one team in the Top 16 moved more than two places. South 3 champion Delsea would fall from No. 1 to No. 2 overall, but still would have been a top seed in their section and hosted their sectional final. South 3 semifinalist Camden would have improved is six-seed to No. 4. And sectional finalists Seneca would have dropped one place, while Cedar Creek would have held steady at seven.

North 2: The top seven teams all stayed the same, three sets of two teams flip-flopped, and overall eight stayed put in the switch to OSI. Top overall seed Caldwell didn’t move, North 2 champ Westwood held in second, sectional finalist Rutherford held third, and North 3 semifinalist Bernards stayed at No. 4. Lakeland, a sectional semifinalist, improved from 9th to 8th. One new team made the field, with 17th-place Becton (4-4) moving up to bounce out 16-seed Madison (3-5). Call that one a push.

South 2: The top four teams remain the same here, but with the middle two in a different order. Sectional finalist Point Pleasant Boro remained No. 1 overall, while sectional finalist Haddonfield and South 3 champion Rumson-Fair Haven flip-flopped to 2 and 3 respectively. South 3 runner-up Willingboro held fourth. Overall five of the top seven teams would have ended up in the same spot. One downside is that Spotswood would get left out at 8-1 – they were 16 in UPR and finished 18 in OSI – while Delran (2-7) would get in. But guess what? If the NJSIAA raised the minimum win threshold from two to three, that would eliminate Delran as well as Raritan (2-6) which got in at 14 in power points and 11 in OSI. Knocking them out would put in Mastery Charter of Camden (3-5, 18th in UPR) as well as Spotswood, in the same 16th slot they got with the UPR formula.

North 1: Remarkably, the top ten teams wouldn’t have moved an inch going with OSI instead of UPR. That includes sectional finalists Butler (No. 1 overall) and Hawthorne (4th overall). North 1 champ Mountain Lakes also would have held in 8th, while North 1 runner-up Shabazz would have dropped from 11 to 14. At the bottom, North Arlington (3-5, 17th in UPR) would have gotten in, and Glen Ridge (5-4) would have dropped from 15th to 17th and been bounced by OSI. But again, raising the win minimum to three would have eliminated Verona, which was 2-7, and made the playoffs on UPR at No. 16. OSI would bump them to 11, but again, we think two-win teams don’t belong in the playoffs. Take them out, and Glen Ridge is back in.

South 1: South 1 runner-up Woodstown would remain No. 1 overall with a switch from UPR to OSI, while sectional finalist Schalick would have jumped from 6th to 2nd. South 1 champion Glassboro would edge up from fifth to fourth, while sectional finalist Woodbury would move from seven to No. 2. This section actually had the most volatility. Only two teams would have had the same seeds, and the average move was 2.8 seeds, with two teams going up five spots and one – Penns Grove – moving up six places. (By the way, we looked purely at standings, not head-to-head tiebreakers.) The only downside here is that Paulsboro – somehow, since they’re 1-7 – would have jumped from 18 to 12, but they don’t even meet the current NJSIAA win requirement. This is where the rule works, and it would have kept them out of the playoffs, and everyone who got in on the UPR formula also would have gotten in with the OSI formula, including Dunellen, which fell to 17th on Paulsboro’s jump.

by Mike Pavlichko

St. Joseph-Metuchen, JP Stevens have the BCC’s most difficult schedules, by different metrics

You can do all the realignment you want and have the most perfect scheduling on the planet, but at the end of the day, some teams will win championships, most will not. Someone will have a higher ranking, more wins, and others won’t. That’s the nature of competitive sports when numbers come into play.

Central Jersey Sports Radio took our annual look at two different metrics, and found St. Joseph of Metuchen and JP Stevens have the most challenging schedules in the Big Central Conference this year, depending on which metric you look at.

The two we examined are OSI (Opponent Strength Index) and one we created, called “Play Up Or Down.”

OSI is what’s used by the state to determine 60 percent of the playoff formula, with power points – albeit in an altered state this year – making up the other 40%. OSI is comprised of the average Strength Index of opponents, with a win counting for the full value of a team’s OSI, and a loss garnering half those points.

Inotherwords, beating a team with an SI (Strength Index) value of 80 gets you 80 points, while losing to that team gets you a 40.

For our analysis, we took the full available OSI points and averaged them out for each team to determine a strength of schedule. St. Joe’s had the schedule with the highest average opponent Strength Index values.

But the “Play Up Or Down” is another metric we decided to look at that shows how much stronger or weaker your opponents are. And it’s a good counter to OSI.

Consider that the “best” or highest SI value team can only play down. If they have the highest SI value, everyone will be lower than them, so naturally their schedule will be “weaker” opponents. On the flip side, the team with the “lowest” SI value is always going to be playing up.

For example, Phillipsburg has the second-highest SI value in the Big Central at 80.01, but the Stateliners rank 54th among 59 Big Central teams, playing “down” to opponents who, on average, are 10.57 points lower than they are.

However, even though teams with high OSI might not always be playing “up,” they can still have a very challenging schedule. If their division is balanced, they’ll be playing similarly-ranked teams, who would conceivably also be very good.

For P’burg, that’s still a challenging schedule, meaning their opponents have an average 69.45 SI value, which is solid. It’s just that mathematically, since they’re second best in the BCC in SI value, there’s simply no one “better” than them to play.

JP Stevens has the fourth lowest SI value in the league, and the average SI of the teams on their schedule is higher compared to their own ranking than any other team in the Big Central: 18.03 points. That means, at least at the start of the season, on average, their opponents are 18 points better than them.

So how did it shake out overall in the Big Central?

Here’s a look at every Big Central team’s OSI based on the best schedule information currently available. Teams are grouped by division, and we include whether they are playing Up or Down.

As BCC President Scott Miller told us last week on Media Day, but which also has been evident through its new divisional alignments, the league has been striving for competitive balance lately, eschewing a total reliance on Group Size – which the NJSIAA has followed in eliminating Group Points from Power Points this year – and focusing more on performance on the field.

Here’s a look at each Big Central Division on the whole:

There are a few numbers to look at here.

  • “Average SI” is the average strength index value of the teams in that division. According to the official starting SI numbers for 2024 released by Gridiron New Jersey, the American Silver (Phillipsburg, Ridge, Bridgewater-Raritan, Hunterdon Central and Hillsborough) has the strongest grouping of teams – on average – with a 71.57 average SI. The American Gold (with Union, St. Joseph, Westfield and Elizabeth) is next at a 67.72 average SI.
  • “Range” is the margin between the lowest and highest SI value teams in a division. So while the American Silver may be ranked higher than the Gold, the Gold would be considered more competitive, as only 7.11 SI points separate the top team (Westfield, 72.30) from Union (65.19). It essentially says any game between these teams would ideally be decided by a shade more than a touchdown: 7.11 points. The Silver has a higher SI, but the range is double, 14.11 points, from the highest-ranked team (Phillipsburg, 80.01) to the lowest (Hunterdon Central 65.90).
  • “Average OSI” is the average of the OSI of all the teams in a division.
  • “Average +/-” is the “Up or Down” for a whole division. The United Silver will “play up” the most this year, by 9.04 SI points, while the Patriot Silver “plays down” the most this year, by 4.89 points. But remember, that’s not a lot. Nine points is a shade more than a touchdown. Five points is less than a touchdown. Those are still close, competitive games.

It’s also important to note that this evaluation includes all games – except against out-of-state competition, like the two games St. Thomas Aquinas will play this year in Maryland, which don’t count toward OSI or power points, and non-public seeding is decided by a committee anyway.

That means Big Central crossovers and non-conference games against other New Jersey opponents are included. And the BCC often uses crossovers to help “adjust” a team’s schedule to achieve competitive games.

So how do the league’s 59 teams stack up?

Below is a team-by-team look at every Big Central team’s potential OSI – the maximum they could achieve if they went undefeated – ranked from highest to lowest:

The following chart is the second metric, “Play Up Or Down,” with all 59 Big Central teams ranked in order from Up to Down:

And, of course, here’s the most important caveat of all. Strength Index values change all season long, as games are played. So whoever has the strongest schedule before the season starts may not have played the strongest schedule at the end, and vice versa.

As we do every year, Central Jersey Sports Radio will evaluate what we believe are the minimum number of wins needed to clinch a playoff berth. While there is no “set minimum” by the NJSIAA – which got rid of the .500 record requirement in the early 2000s – we look at how much each game is worth on a team’s schedule, on the average, then look at the lowest OSI-ranked teams to make the playoffs.

We might figure, for example, that Edison would need to win five games to make the playoffs. If they won four and missed, we projected correctly. If they won seven, and didn’t qualify, we missed the mark.

But if they win five or more and get in, or only win four or fewer and lose, then we got it right.

In three prior seasons of projections, Central Jersey Sports Radio’s projections have had a whopping 95% success rate. That includes going 54 for 57 last year, 55 of 57 in 2022, and 55 of 58 in 2021.

In those three years, our all-time record stands at 164 out of 172 correct in a three-year span.

At the moment, we’re waiting for the official NJSIAA classifications to be released, as they determine what supersections teams will compete in for playoff seeding, and the minimum OSI numbers will vary by group and geographic supersection.

by Mike Pavlichko

NJSIAA publishes 2024 football tournament regulations; playoff formula tweaked again, and “consolation” games are out

The NJSIAA posted its 2024 football tournament regulations on its website Wednesday, and they reflect two big changes made to the playoff system back earlier this year.

The first has to do with multipliers. In a change that’s in line with what Central Jersey Sports Radio’s Mike Pavlichko has proposed since the multiplier was first introduced almost a decade ago, multipliers will vary from team-to-team in each category, and now are based on Strength Index value rather than a fixed amount. That means the numbers are based more on an actual team’s performance, rather than all teams in one category being equal, regardless of their record.

The second change has to do with power points. For decades it has had three components: quality points (6 for a win, 3 for a tie, none for a loss), group points (based on an opponent’s group classification, 5 for a win over a Group 5 school, 4 for a Group 4, etc.) and residuals (3 points for every win by an opponent you beat, 1 for every win by an opponent you lost to).

But now, group points are out of the equation, the reasoning being that group size has taken on less importance over the years as it related to the quality of teams. And the change is being made across all sports

The NJSIAA says it still values power points, but in reality, group points were the only real difference between power points and OSI.

Both reward teams for winning. A victory gets you six quality points in the power point formula and the full value of an opponent’s Strength Index value. A loss garners no quality points and half an opponent’s SI.

Both reward teams for strength of schedule. Power point residuals earn a team 3 points for every win by a team they beat (up to their seventh game), and one for every win by a team they lost to, while strength of schedule is inherent in the OSI system, as SI values determine a team’s strength based on the scores of games, more than wins or losses.

In fact, the SI system is likely more accurate, since not all wins are created equal. If Teams A and B both have 5 wins, and you beat both teams, you get 5 points each, even if Team A beat all the top teams on its schedule and Team B beat all the weakest teams.

The NJSIAA’s football regulations for 2024 can be downloaded from their site by clicking here.

Bye-bye consolation games…

Though traditional consolation games also have been tweaked in recent years, with the NJSIAA calling it an “invitational tournament” last year, the state athletic association decided earlier this year to do away with it entirely.

Over the years, there had been controversies with matchups, and many teams had to travel long distances to play otherwise meaningless games. So, this year, the NJSIAA has dropped consolation games entirely, and will leave it up to schools that missed the playoffs to schedule their own games.

Those games must fall within the NJSIAA’s scheduling framework, with a maximum 14 games allowed through December 1. Games also are not allowed to be played on Fridays, which is when the vast majority of state tournament games are played, in order to make playoff games a priority for getting the best officials.

Other changes…

The NJSIAA also modified its running clock rule, which previously went into effect automatically once the score differential reached 33 points or more during the second half in a game. The running clock would go back to normal once the margin got closer than 25 points.

But the NJSIAA says there have been concerns about unsportsmanlike behavior during the running clock period, and the Executive Committee approved a rule to keep the running clock in place no matter how close the game gets. So once the margin is at least 33 points in the second half, the running clock stays in place for good.

Another change adopted in February allows football players to participate in more than one scheduled game within 24 hours. A two-year pilot program changes the limit to no more than five quarters in the same time period. That would allow a student-athlete to play, for example, two quarters of a JV game Thursday, and three in a varsity game Friday.

In its proposal to the Executive Committee, the NJSIAA wrote: “This recommendation has been reviewed and endorsed by the NJSIAA Sports Medical Advisory Committee (SMAC). However, NJSIAA SMAC recommends that this rule be approved under a two-year pilot program to solicit feedback and data before supporting full adoption.”

The idea is to allow greater flexibility to programs that may have low participation to field multiple levels of teams, as opposed to, say, having to drop a freshman team.

by Mike Pavlichko

Who has the best OSI in the state? Hint: It’s a Big Central team

As if anyone who has been following the Big Central Conference all season needs any more proof of how good Phillipsburg is, here’s more statistical proof: the Stateliners have the top OSI among all public schools in the state, and the highest in their league.

OSI is the Opponent Strength Index, and is an average of the Strength Index values of the opponents a team has played. A win gets a team the full value of its opponent, while a loss gets half.

Phillipsburg’s OSI is 72.95, tops among all 288 public schools in the state that are eligible for the playoffs in 2022. (Schools in the Super Football Conference “Ivies” and their NJIC counterparts – with “relief” schedules who are predetermined not to take part in the playoffs – are not included.)

But what does that number mean? How is that different from North Brunswick, which is also 5-0, but has an OSI of 62.44? It means, according to the Strength Index, Phillipsburg has played a more difficult schedule than North Brunswick – teams that, on average, are about 10 points better than those the Raiders have played. Although, both are good enough to be tops in their respective supersection playoff standings, with three games to go before Cutoff Weekend.

The highest OSI of all teams in New Jersey belongs to Don Bosco Prep, with an OSI of 89.5. The Ironmen are the No. 1 team in the state at the moment, according to both the NJ.com Top 20 and the USA Today Network Rankings (which includes MyCentralJersey.com).

They’re followed by Seton hall Prep (76.05), St. Peter’s Prep (75.65) and Bergen Catholic (74.53).

Among public schools, North Hunterdon (5-1) ranks 12th in the state (63.01), with North Brunswick right behind them in 13th. Hillsborough (4-1) is No. 15 at 61.8, and Westfield is 29th, with a 57.77 OSI.

Those being the top four teams in the Big Central in OSI, St. Thomas Aquinas checks in 5th at 61.71, while St. Joseph-Metuchen has a 50.35 OSI, good for 16th overall.

The OSI – as well as power points and the NJ UPR – are calculated by Gridiron New Jersey for the NJSIAA.

Below are the rankings of all the Big Central teams, according to OSI. Non-Publics are in Blue.

Below is a PDF file of all New Jersey schools ranked by OSI. Non-Publics are in Blue.

NJ HS FOOTBALL TEAMS RANKED BY OSI – Week 5Download
by Mike Pavlichko

Who’s got the toughest road in the Big Central? Updated rankings based on revised slate

by Mike Pavlichko

Even though the NJSIAA state playoffs in 2020 won’t look anything like normal – if they’re even held at all – it’s still an interesting exercise to look at some of the numbers they use, like Strength Index, and what we are calling the “Preseason OSI.”

Meanwhile, the revised schedules created by the Big Central Football Conference have in some cases created big changes in the strength of schedule for some teams.

The key stats are:

  • Preseason OSI: This is the average of the Strength Index of all a team’s opponents. During the season, OSI is calculated by using the full value for a win and half for a loss. In the preseason, we use full values for all opponents to keep the numbers evenly proportioned.
  • Play Up or Down: This is how much higher or lower a team’s OSI is compared to their own Strength Index. If a team’s OSI is higher than its own SI, that means they are playing stronger opponents, or “playing up.” In a normal playoff year, playing up by a lot could get you in the postseason even with a sub-.500 record. Playing down too much might keep you out, like happened to South River in a 6-3 campaign last year. This stats says scheduling is important.
  • Change in Preseason OSI: We thought it would be good to look at OSI values we calculated back in February when the regular season was 9 games long as opposed to 6 (and an even shorter 5 for those who had Carteret or Piscataway on their schedules, unless they find replacements).

We’ll give you some of the highlights here. For a full list of rankings of Preseason OSI, Play Up or Down, and Change in Preseason OSI, scroll to the bottom of each section and download the posted PDF files.

Continue reading “Who’s got the toughest road in the Big Central? Updated rankings based on revised slate” →