Strength of Schedule

Welcome to Discuss Fastpitch

Your FREE Account is waiting to the Best Softball Community on the Web.

Aug 10, 2016
687
63
Georgia
Was looking through the rankings on MaxPreps and they have the SOS for each team. While DD's team has one of the highest as one of the top teams in the nation happens to be in our region (yay), I saw some negative SOSs. I tried to read to see exactly how these are calculated but could not find a simple explanation for it. How do you even get a negative number?

I know it has to do with the records of the teams you play but it seems so complicated.
 
Jun 8, 2016
16,118
113
Was looking through the rankings on MaxPreps and they have the SOS for each team. While DD's team has one of the highest as one of the top teams in the nation happens to be in our region (yay), I saw some negative SOSs. I tried to read to see exactly how these are calculated but could not find a simple explanation for it. How do you even get a negative number?

I know it has to do with the records of the teams you play but it seems so complicated.
It is probably a displaying thing. Sometimes SOS is displayed as a distance from the average so if you have a below average SOS it will be negative.
 
Jun 6, 2016
2,724
113
Chicago
Was looking through the rankings on MaxPreps and they have the SOS for each team. While DD's team has one of the highest as one of the top teams in the nation happens to be in our region (yay), I saw some negative SOSs. I tried to read to see exactly how these are calculated but could not find a simple explanation for it. How do you even get a negative number?

I know it has to do with the records of the teams you play but it seems so complicated.

A negative SOS just means an easier schedule.

But don't put much into the MaxPreps rankings. The algorithm is terrible.

Right now in girls volleyball, our school is ranked behind a nearby school. Not only do we have a much better record than them, but we also crushed them h2h (it was 25-11, 25-6; our JV team could beat their varsity team). They weirdly have a positive SOS, but I don't think there's a school on their schedule that's better than some of the schools we've beaten (and ours is a negative SOS, by quite a bit!). The schools they play only play in their low level public league conference.

So basically, they've lost to a bunch of bad teams that have a good record only because they've beaten worse teams. Even though they have a .250 winning percentage, they're ranked higher than a team with a .588 winning percentage that plays better competition in a stronger conference.
 
Mar 4, 2015
526
93
New England
I know it has to do with the records of the teams you play but it seems so complicated.

I don't know what MaxPreps' algorithm is exactly, but Pattar's answer is likely the correct one, IMO.

Also, MaxPreps' rankings (including SOS) are probably not based on W-L records, although I don't know that. Some ranking systems do use W-L record, such as the antiquated RPI Index that NCAA seeding committees unfortunately still sometimes use. That's a poor way to do rankings because it is not sensitive enough to strength of schedule. The best ranking systems out there (such as Sagarin, Massey) don't consider W-L records at all. It's just a comparison of all the scores so that if you're 2-10 but beat a 10-2 team you'll be ranked higher as long as there are no other scores (ie, evidence) that contradict or cast doubt on the validity of that head-to-head meeting.
 
Jun 6, 2016
2,724
113
Chicago
I don't know what MaxPreps' algorithm is exactly, but Pattar's answer is likely the correct one, IMO.

Also, MaxPreps' rankings (including SOS) are probably not based on W-L records, although I don't know that. Some ranking systems do use W-L record, such as the antiquated RPI Index that NCAA seeding committees unfortunately still sometimes use. That's a poor way to do rankings because it is not sensitive enough to strength of schedule. The best ranking systems out there (such as Sagarin, Massey) don't consider W-L records at all. It's just a comparison of all the scores so that if you're 2-10 but beat a 10-2 team you'll be ranked higher as long as there are no other scores (ie, evidence) that contradict or cast doubt on the validity of that head-to-head meeting.

Here's what MaxPreps says about it:
1665070500185.png

Of course, how they determine SOS is suspect.
 
Mar 4, 2015
526
93
New England
Here's what MaxPreps says about it:
View attachment 26665

Of course, how they determine SOS is suspect.

Thanks, good find. It's kinda vague, and probably intentionally. Already knew it didn't take into account human opinion, else it wouldn't be a computer ranking. And W-L record should be irrelevant. That doesn't mean winning is irrelevant. Winning is the most important thing. If Team A beats Team B, that is evidence that Team A is better than Team B. But if Team A is 10-2 and Team B is 4-8, there is no evidence on that fact alone that one team is better than another.

One other thing on MaxPreps - Their rankings accuracy depends on all scores being reported. Some of the crazy rankings might occur because of missing or inaccurate reporting. MaxPreps rankings likely are much more realistic in some states than others.
 

Latest posts

Members online

Forum statistics

Threads
42,857
Messages
680,286
Members
21,527
Latest member
Ying
Top