Author Topic: Most difficult schedules  (Read 8302 times)

Offline frank_ezelle

  • All-American
  • ******
  • Posts: 2377
  • Karma: +159/-50
    • View Profile
    • Millsaps Sports Photo Albums
Re: Most difficult schedules
« Reply #15 on: April 22, 2009, 07:06:22 am »
OshDude, count me among the NCAA skeptics.  Like you, I don't know how Boyd comes up with his numbers and we do know that his data isn't 100% accurate because he is depending on the reporting of others (for example, the win-loss record for Millsaps is close but not correct).  However, I like a system that factors in more than simply the winning percentages of the teams.

The problem I have with the OWP and the OOWP, is that it can easily get skewed in areas with a lot of D3 programs.  The strong teams feast on the average and the average feast on the weak, and on paper it looks like the top dogs are doing great.  In areas with less teams, the cherry picking isn't so easy since you have to play just about every team within driving distance to fill out a schedule. 

I have no better plan and I think the NCAA has tried to address the problem by adding the additional layer of OOWP in recent years, but it's still not a perfect system.  I think it's a system that some teams use to their advantage and I can't blame them for that--if Millsaps was in a position to play the system to increase their NCAA Tournament chances then I would expect them to do the same.
« Last Edit: April 22, 2009, 07:09:56 am by frank_ezelle »
Millsaps Athletics:
Millsaps Photo Website:

Offline Bronko7

  • Second-stringer
  • **
  • Posts: 106
  • Karma: +8/-3
    • View Profile
Re: Most difficult schedules
« Reply #16 on: April 22, 2009, 09:13:25 am »
The OWP & OOWP is a killer to some teams. They are stuck in conferences that are just brutal. Those teams have to go out and try to schedule out of conference top in-region teams to offset the weakness of the conference schedule. It is even more evident to the pool B's than anyone. Bethany Lutheran is a great example this season, with good wins over RPI & Fischer (both non in-region games) the UMAC conference just drags the numbers down. St. Scholastica has the same issue, but they try to get the better MIAC schools on the schedule.

Offline Ralph Turner

  • Hall of Fame
  • All-American
  • ********
  • Posts: 28781
  • Karma: +1835/-402
  • Hall of Famer
    • View Profile
Re: Most difficult schedules
« Reply #17 on: April 22, 2009, 09:17:01 am »
The OWP and OOWP work well in heavily concentrated areas like the northeast, but do not work for teams in isolated areas such as the teams in the West Region.


  • Guest
Re: Most difficult schedules
« Reply #18 on: April 22, 2009, 07:34:50 pm »

Hopefully this will explain Boyd's SOS ...

SoS -- The strength of schedule, measured as the average ISR of the team's opponents.

Strength of Schedule
Publication Date: July 11, 2000

This Year's Numbers

I've just released my Strength of Schedule numbers for the 2000 season, and along with that I've redone the numbers for 1999 and 1998. There's no reason to reproduce those numbers here, other than the occasional quote, but the rest of the column is about them and their repercussions, so it'd probably be a good idea to go take a look at them, at least briefly, before reading further. Go on, I'll wait.

The first thing you'll notice, probably, is that major conference teams dominate the top of the list -- there are four Pac-10 teams in the top eleven, including the two toughest schedules, and five SEC teams in the top ten. Opinions differ on how much credit those teams should get for that, since the biggest portion of that comes from conference games where the teams had no choice but to play (that's not an absolute, as the top four all played top-flight competition in their non-conference schedule as well, but it's certainly true for someone like Arkansas, for example). I personally attach no moral value to strength of schedule, preferring to just use it as part of my means for analyzing how good a team has done, so I don't care why they played good competition, I just care that they did. It is true that I'm impressed by teams in mid-level conferences who go out of their way to schedule really tough competition, such as Cal State Fullerton, who effectively played the toughest non-conference schedule in the nation.

The top ten in non-conference strength of schedule were Alcorn State, Cal State Fullerton, Houston, Southern California, Louisiana-Lafayette, Stanford, Long Beach State, Loyola Marymount, Southwest Texas State, and Oklahoma, for those of you who don't feel like digging through the list. Alcorn State is an extreme statistical oddity, as they only played one game against a non-SWAC Division I team (Mississippi State). While I'm not willing to come up with some rule to exclude them, feel free to ignore them if you wish.

Obviously, some of these guys have geographic advantages, while at least one of them benefitted from postseason games that boosted their overall schedule quite a bit (although in fairness to the Cajuns their regular-season non-conference schedule was not at all bad).

Standard Deviations

The usual method for determining strength of schedule, the one that I use here, is to simply average some numeric measure of each opponent's strength such as their ISR or their winning percentage. This works fine in most cases and certainly gives a good starting point for analysis. There are, however, fringe cases where it's a bit lacking, and, since that may have an effect on the accuracy of the ISR's, I'm trying to get a feel for how to measure that. Looking at the standard deviation of the opponents' ISR's strikes me as a good first measure.

The problem is that the average doesn't give sufficient information to completely measure a team's schedule -- the distribution also matters. Let's look at a simplified version of two schedules for illumination:

Schedule A:   
Florida State
High Point
New York Tech

Schedule B:
South Alabama

Assuming that we're talking about the 2000 versions of these teams, these two schedules have almost identical average ISR's, but they're obviously not even remotely the same in difficulty for most teams. The interesting thing is that how tough the schedule is depends to a large extent on how good you are.

Given three games against each team on the schedule, an average team would most likely win six of the twelve games against either schedule. A bottom fifty team, though, would likely win three games against Schedule A and only one or two against Schedule B. Similarly, a top ten team, the ones everyone is most interested in, would likely go 11-1 against Schedule B, while they would probably only go 9-3 against Schedule A.

How does this effect the real world? Well, there are a few teams, Miami and Wichita State being the most obvious, who have unusual distributions to their schedules. They attempt to schedule top teams, but also have quite a bit of really bad filler in there as well. I think that the average ISR method, which is implicitly used in the construction of the ISR's themselves, probably slightly underrates these teams. Evidence for that view might be Miami's 1999 national championship, although that's an awfully isolated data point. If you look, though, they have the highest standard deviation of any top twenty team for the 1999 season, which may be a point in their favor.

I don't have a great idea for how to handle this at this point, and the effect is not a large one anyway, but I'm open to suggestions. Some sort of probabilistic analysis is probably the best idea I have so far.

I apologize for the lateness of this column -- not a good thing to do when you're trying to build readership, I know. I'll always try to have the column out on Tuesday morning, but I was on vacation last week (no intended baseball content, but I saw a French baseball club in Charles de Gaulle airport and then ended up sitting next to the mother of a former Georgia player on the flight home from London; my life just keeps getting stranger) and didn't get my usual head start on writing. I hope no one is inconvenienced by the delay.

What are the ISR's?

The ISR's are the results of an algorithm designed to measure the quality of a team's season to date by combining their winning percentage with the difficulty of their schedule. The algorithm computes all teams simultaneously and attempts to take advantage of inter-regional games more accurately than other rating systems.

How are the ISR's computed?

The basic idea is an iterative one. Begin with all teams set to an even rating -- 100 in this case. Then, for each game played, give each team the value of their opponent's rating plus or minus a factor for winning or losing the game -- 25 in this case. Total all of a team's results, divide by the number of games played, and that's the end of a cycle. Then use those numbers as the start of the next cycle until you get the same results for each team for two consecutive cycles.

Why are the ISR's needed?

While it's still a great game, college baseball suffers from the lack of an accurate rating system for measuring team quality. The traditional polls suffer from voters running on auto-pilot, and the RPI's used by the selection committee have some serious problems with the method used to determine strength of schedule. Because of the small amount of inter-regional play in the sport, some regions tend to be under-represented in the NCAA tournament, and mid-rank large conference teams tend to be unfairly excluded. Although trying to get the selection committee to acknowledge this may be a hopeless case, the ISR's are an attempt to find a better rating system.

Who is Boyd Nation, and why should anyone pay attention to this stuff?

Boyd is a lifelong college baseball fan who has a master's degree in computer science with a focus on algorithm development. The ISR's are intended to improve enjoyment of college baseball by producing better-informed fans; some of us enjoy the games more when we have a feel for how likely certain results are. If that's you, enjoy.

« Last Edit: April 22, 2009, 07:37:50 pm by Milby »


  • Guest
Re: Most difficult schedules
« Reply #19 on: April 22, 2009, 07:39:09 pm »
Bottom-line ... does anyone have an idea on what SYSTEM would be the BEST for DIII baseball ?