Where's the disconnect?

Note that ISR (which has been posted in the left hand column in this thread) is a creation of Boyd Nation. As far as I know, nobody (including the NCAA selection committee) uses that statistic except Boyd himself. RPI which Boyd also publishes may exhibit less discrepancy.

If you are arguing for any one school, the trend in college is to argue the poll or the statistic that puts one in the best possible light. Thus, if someone has a better ISR than RPI, then argue the ISR. With respect to ISR, none of the baseball publications such as Baseball America, Rivals.com, sebaseball.com ever discuss that stat. RPI is the driving stat when it comes to post-season at-large recognition.

Similarly, there are four major polls with big discrepancies. Most schools list all the polls they are ranked yet feature or highlight the polls that ranks them the highest. Human nature I guess.
My son's school is ranked anywhere from 14th (in two polls) to 26th in another and 23rd in BA. If you asked me, I would tell you we were 14th ranked. Likewise, until these ISR numbers were posted in this thread, I couldn't have told you our ISR was 29 but I could have told you our RPI was #9.

My mistake. I thought I was posting RPI's.

For a newbie like me, could you please explain the difference between ISR and RPI in layman's terms? Thanks!

I just looked at Boyd's World and noted the two lists for ISR's and RPI's. There definitely are differences between the two, but they typically are not that stark. For example:

Ariz. State has a 1 ISR, 4 RPI
Nebraska has a 2 ISR, 6 RPI

Further down the list:

Pepperdine has a 15 ISR, 27 RPI
Vanderbilt has a 30 ISR, 25 RPI
Dallas Baptist has a 38 ISR, 38 RPI

One glaring difference:

UC Santa Barbara has a 25 ISR, 69 RPI

.
Here is a layman's/44's explaination...and I could be wrong...But I don't think so...

For starters, after watching Boyds World for years, IMO he is by far the most astute and neutral observer of collegeiate baseball there is, and I'm not the only one who thinks so. He used to have more great rankings on player performance a few years ago, but was so good that he was hired by MLB to provide them with player data to help identify players for the for the draft and they bought up his data and kept it for themselves.

Here are the three major rankings you see, an offcial explaiantion...and 44's view:

ISR - "The Iterative Strength Ratings, a measure of team quality for NCAA Division I baseball" This is Boy's best and personal measure of team stregnth. Now while any statistical measure is only as good as the numbers chosen, I can assure you that Boyd's numbers are better and far more accurate than the numbers/political bias/favors/dice/straws the knuckleheads at the NCCA who do the tournament invitations use.

RPI - "An accurate simulation of the Ratings Power Index, the official NCAA measure of team quality used by the tournament selection committee" THis is Boyd's best guess of what the NCAA Knuckleheads will do...and he is most often very close. (He knows the knucklheads better than they know themselves)

SOS - "Strength of schedule as presented in these listings is the average ISR of a team's opponents. For a more sophisticated measure, the standard deviation is also included." This is Boyd's best estimate as to the ranking of a teams schedule, nobody else goes to this much trouble.

SO....for UCSB for example...Boyd (ISR)thinks that they are the 25th best team in the country...what Boyd in his statistical fantasyland fully fails to appreciate understand is that they have no political/geographical/alumni/historical/buts in the stands clout...so NCAA reality (RPI)is that regardless of who they play, or how they do, or how strong their conference they cannot possibly be any good....and as a result have no business in the NCAA'S private members only postseason party.

Hope this helps.

44
.

I hope I didn't imply that I did not respect Boyd's opinion. I love numbers so lets talk about them.

quote:

For starters, after watching Boyds World for years, IMO he is by far the most astute and neutral observer of collegiate baseball there is

Surely, we do not want computers deciding the two teams who are in the National Championship game like football? Maybe there are some legitimate arguments about bubble teams in baseball but stats are subjective just like the rankings. How could that be?

Lets pretend I am going to become the next Boyd Nation and I am going to develop an algorithm that definitively predicts the correct order of the teams.

I like offense so I am going to give computer points to those teams who score the most runs. But wait you say, if that same team who scores the most runs gives up more runs than they score, then they are probably not as good of a team.

Good point. I'll modify my algorithm. I am going to give the most weight to run differential, the team that scores the most and gives up the least runs. But wait you say, some teams who have a great run differential do not win many games. Maybe they are a poor fielding/fundamentals team and are poorly coached you might say.

Good point. I'll modify my algorithm. Run differential + wins scores the highest in my system. But wait you say, some teams have a good record, but they play in the local Nun's softball conference. Somehow, there has to be consideration given to the quality of opponents.

Good point. I am going to add a ratings power index to my system. It gives points based on the number of wins the teams I beat rack up. But wait you might say, those teams that you beat are regionalized with other teams who do not play anybody.

Do you see where I am going with this. You say tomato and I say tomotto We can now continue down a path of endless arguments and tradoffs trying to determine what strength of schedule means and what points are awarded and so forth based on subjective decisions built into the algorithm. No one opinion is necessarily more persuasive than another when it comes to designing the ultimate algorithm, it is just one man's opinion at the end of the day.

The only reason I don't pay attention to ISR is because it does not affect decion-making of who gets into the tournament and how they are seeded. It may be the ultimate statistical tool. Is that right? Maybe not, but I only care whether or not my kid gets to play in the post-season.

What I have noticed is that when you compare ISR with RPI, more western schools move up in the rankings and more eastern schools move down in the rankings. If I am a fan of the western schools, I argue that is the way it ought to be and that Boyd's ISR makes all the sense in the world. Same with the rankings I noted above. The most accurate poll is the one that ranks you the highest. That should be obvious to even the casual observer

quote:

Originally posted by ClevelandDad:
What I have noticed is that when you compare ISR with RPI, more western schools move up in the rankings and more eastern schools move down in the rankings. If I am a fan of the western schools, I argue that is the way it ought to be and that Boyd's ISR makes all the sense in the world.

Just to add a comment before people throw out conspiracy stories; Boyd lives in the southeast.

quote:

Originally posted by Beezer:

quote:

Originally posted by ClevelandDad:
What I have noticed is that when you compare ISR with RPI, more western schools move up in the rankings and more eastern schools move down in the rankings. If I am a fan of the western schools, I argue that is the way it ought to be and that Boyd's ISR makes all the sense in the world.

Just to add a comment before people throw out conspiracy stories; Boyd lives in the southeast.

Beezer - I was not arguing that Boyd is biased. At the end of the day, his list is no different than any other list (e.g., rpi, polls) subjectively compiled.

quote:

What I have noticed is that when you compare ISR with RPI, more western schools move up in the rankings and more eastern schools move down in the rankings.

My son will play in the middle of the country next year - neither east nor west. Wonder where this leaves his team. Hmmmm....

Wow -- I had no idea my question was so loaded! Guess I'll just kick back and try not to think so much.

Thanks for all the insights!

quote:

Originally posted by Infield08:
My mistake. I thought I was posting RPI's.

For a newbie like me, could you please explain the difference between ISR and RPI in layman's terms? Thanks!

http://www.boydsworld.com/baseball/faq.html#1

O-44 ...

quote:

...so NCAA reality (RPI) is that regardless of who they play, or how they do, or how strong their conference they cannot possibly be any good....and as a result have no business in the NCAA'S private members only postseason party.

Comparing the ratings gives one a strong impression that the coaches' poll is strongly biased toward the east.

.

IF08...

Try this...the latest Boyd article comparing selction citeria including ISR and RPI...a bit rugged for the non stat guys...but very statstically enlightening....

http://www.boydsworld.com/breadcrumbs/committee_factors.html

44
.

quote:

Originally posted by Infield08:
My son will play in the middle of the country next year - neither east nor west. Wonder where this leaves his team. Hmmmm....

Wow -- I had no idea my question was so loaded! Guess I'll just kick back and try not to think so much.

Thanks for all the insights!

The idea of measuring teams even with statistics is to a certain degree a measure of futility when you are comparing some numbers and looking at it outside of just the #'s. Mid week vs Weekend series is just 1 reason the #'s can be misinterperted.

As far as where does that leave your Son's team, look at the history of the conference is it a 1 bid league, has it had success in the past, look at the schedule and most importantly does it win.


On a side note one thing that bothers me when comparing schools from around the country are the conferences that play 7 inning scheduled Double Headers. IMO that changes everything if you have a couple strong starters who either pitch complet games or go 6 and hand it off to you closer. Its a lot different then pitching innings. I realize it is done due to scheduling issues etc,, just seems thatits not little league a game should be 9 innings.