#2 in Sagarin

[PUownsIU]

Purdue is up to #2 in the Sagarin computer rankings, and that's with playing almost half the games missing a starter (either Davis or Hammons). They would probably be #1 if they played with a full complement in each game.

Purdue has won two games against the top 50 teams...no one else has won more than that.

 

[WestCoastRoses]

#6 on KenPom

 

[Grover]

It is WAY too early for those computer polls to accurately rank all of the teams.

In a few days when Purdue finally is rated #1, then we'll know they are accurate.

 

[SnowLeopard]

Purdue is up to #2 in the Sagarin computer rankings, and that's with playing almost half the games missing a starter (either Davis or Hammons). They would probably be #1 if they played with a full complement in each game.

Purdue has won two games against the top 50 teams...no one else has won more than that.

Dayton, the secret scrimmage, is also in the top 50

That makes 3 teams (kinda, not really, I know)

 

[toddbrewster]

Watch out for New Mexico. The name does not mean much, but they have a pretty good team. They are well balanced and we have to play hard against them.

 

[Schnelk]

In other words...NO LET DOWN! Right Todd?

 

[artee]

In other words...NO LET DOWN! Right Todd?

Hopefully the depth of this team will not allow a letdown.

 

[toddbrewster]

Exactly! The mindset should be that the toughest team Purdue plays this year is the team they play in practice. I also hope they have match ups in practice where Haas and AJ are on the same team, even if it is 2 on 2. Our team is very good, but there is plenty of room for growth.

 

[Title_BU]

It is WAY too early for those computer polls to accurately rank all of the teams.

In a few days when Purdue finally is rated #1, then we'll know they are accurate.

Its way too early for backwards-looking metrics like RPI, but for forward-looking metrics like Pomeroy or partially forward-looking metrics like Sagarin, its not. The data is fully vetted at the end of December, but there's already enough data that its good enough.

 

[TopSecretBoiler]

Its way too early for backwards-looking metrics like RPI, but for forward-looking metrics like Pomeroy or partially forward-looking metrics like Sagarin, its not. The data is fully vetted at the end of December, but there's already enough data that its good enough.

I think your sarcasm meter is on the fritz.

 

[toddbrewster]

According to Sagarin, New Mexico will be the 3rd toughest team we faced this year.

 

[FiveWeight]

I think your sarcasm meter is on the fritz.

I don't think he was being sarcastic. I haven't heard the terms "backwards or forwards-looking metrics" used in this context, but Title_BU's comment made a lot of sense to me.

RPI is only based on a computer calculation of performance so far in the season. Early in the season, there isn't much to base RPI on and so it is inaccurate and swings greatly from week to week. As the season progresses and we have more data points, RPI becomes one of the most accurate predictors of success (though with flaws since it doesn't take injuries and similar issues into account). I like it because it does a rigorous job of comparing success to relative strength of schedule in a way that no human can in their head. Later on in the season, polls also become more accurate but aren't as calculated as RPI and therefore are always subject to hype and the perspective of the poll voters.

 

[Purdue Grad in Texas]

I don't think he was being sarcastic. I haven't heard the terms "backwards or forwards-looking metrics" used in this context, but Title_BU's comment made a lot of sense to me.

RPI is only based on a computer calculation of performance so far in the season. Early in the season, there isn't much to base RPI on and so it is inaccurate and swings greatly from week to week. As the season progresses and we have more data points, RPI becomes one of the most accurate predictors of success (though with flaws since it doesn't take injuries and similar issues into account). I like it because it does a rigorous job of comparing success to relative strength of schedule in a way that no human can in their head. Later on in the season, polls also become more accurate but aren't as calculated as RPI and therefore are always subject to hype and the perspective of the poll voters.

I'm having "flashbacks" to that "spirited" debate last year in several threads concerning RPI versus BPI...anyone recall that? My understanding was that the NCAA Tournament Selection committee was placing less emphasis on using RPI as in previous years.

 

[punaj]

BPI seems more useful to me based on this definition:

"The Basketball Power Index (BPI) is a team rating system that accounts for the final score, pace of play, site, strength of opponent and absence of key players in every Division I men's game. BPI can be used to measure both how well a team has performed (going far beyond just wins and losses) and how powerful it is likely to be going forward."

 

[Danfish98]

Historically there are several things that matter as far as RPI is concerned when the committee is picking teams. None of them are your actual RPI number. Wins against the top 50, record against the top 100, road wins against the top 100, strength of schedule, and losses against below 100. All these factors go into determining who gets in and who doesn't, but it seems very little consideration is given to that team's actual RPI number. That's how mid majors with RPIs in the 20s have been left out, while teams in the 70s occasionally make it.

 

[FiveWeight]

BPI seems more useful to me based on this definition:

"The Basketball Power Index (BPI) is a team rating system that accounts for the final score, pace of play, site, strength of opponent and absence of key players in every Division I men's game. BPI can be used to measure both how well a team has performed (going far beyond just wins and losses) and how powerful it is likely to be going forward."

I agree that sounds even more accurate than RPI. I thought there was a movement a few years ago, at least in football, to move away from computerized comparisons that take margin of victory into account. Not that they are less accurate, but they provide an incentive to run up the score on lesser teams.

 

[CRBoiler]

BPI seems more useful to me based on this definition:

"The Basketball Power Index (BPI) is a team rating system that accounts for the final score, pace of play, site, strength of opponent and absence of key players in every Division I men's game. BPI can be used to measure both how well a team has performed (going far beyond just wins and losses) and how powerful it is likely to be going forward."

I went looking on ESPN for BPI, and they are apparently not doing it yet (not enough data points?). Anyone know when they will start?