RPI/BPI/Ken Pom/Sagarin Combined Top 25

[Money]

Maryland gets penalized because they've been obliterated by middlin' teams on the road and have barely squeaked by some bad teams.

Ken Pom and BPI account for score difference. The polls and RPI just account for if you win.

That's all well and good until you realize that MSU (with double the losses including two to Maryland) is 11 spots ahead of them. That's pretty ridiculous. I'm guessing common sense isn't a factor in those ratings.

 

[TrollyMcTroller]

That's all well and good until you realize that MSU (with double the losses including two to Maryland) is 11 spots ahead of them. That's pretty ridiculous. I'm guessing common sense isn't a factor in those ratings.

No, but margin of victory is, to one degree or another, and since Maryland got their asses handed to them in pretty much every one of their losses, that hurts a bit. (You'll notice most of the MSU losses on the other hand were pretty competitive games)

If you want to figure out how the seeding is going to pan out, put your faith in the RPI. If you want to know how the games turn out, you're better off looking at the ratings from KenPom, Sagarin and BPI.

 

[Money]

No, but margin of victory is, to one degree or another, and since Maryland got their asses handed to them in pretty much every one of their losses, that hurts a bit. (You'll notice most of the MSU losses on the other hand were pretty competitive games)

If you want to figure out how the seeding is going to pan out, put your faith in the RPI. If you want to know how the games turn out, you're better off looking at the ratings from KenPom, Sagarin and BPI.

You do realize that this statement is completely contradicted by the fact that MSU lost both games to Maryland (getting "their ass handed to them" in the 2nd game). Right???

Thank you, however, for clearing that up though. Margin of victory is pretty worthless which explains these rankings.

 

[tducey]

Neat, thanks for sharing, going to be interesting to see the final rankings.

 

[TrollyMcTroller]

You do realize that this statement is completely contradicted by the fact that MSU lost both games to Maryland (getting "their ass handed to them" in the 2nd game). Right???

Thank you, however, for clearing that up though. Margin of victory is pretty worthless which explains these rankings.

MSU lost one game by more than 10 points. Maryland lost all but one game by more than 10 points. But I never claimed it was 100% accurate. To the best of my ability no metric takes style of play, individual matchups and other factors that can determine the outcome of a game. But as a whole the metrics listed above are way better than the RPI for predicting game outcomes.

And margin of victory isn't particularly useful with regards to evaluating a team's resume which is one of the reasons it isn't included in the RPI.

But if you're evaluating an actual team (not it's resume) metrics that include MoV are far, far more accurate at predicting game outcomes. To call it worthless just demonstrates your lack of understanding.

 

[TrollyMcTroller]

Neat, thanks for sharing, going to be interesting to see the final rankings.

I'll post a more thorough write up prior to the tournament starting.

 

[Money]

MSU lost one game by more than 10 points. Maryland lost all but one game by more than 10 points. But I never claimed it was 100% accurate. To the best of my ability no metric takes style of play, individual matchups and other factors that can determine the outcome of a game. But as a whole the metrics listed above are way better than the RPI for predicting game outcomes.

And margin of victory isn't particularly useful with regards to evaluating a team's resume which is one of the reasons it isn't included in the RPI.

But if you're evaluating an actual team (not it's resume) metrics that include MoV are far, far more accurate at predicting game outcomes. To call it worthless just demonstrates your lack of understanding.

If, by this, you mean whether or not a team is going to cover the spread...yes, I agree. If, however, you are stating that MSU should be the favorite to win against Maryland if they should meet again...we'll have to agree to disagree.

 

[TrollyMcTroller]

If, by this, you mean whether or not a team is going to cover the spread...yes, I agree. If, however, you are stating that MSU should be the favorite to win against Maryland if they should meet again...we'll have to agree to disagree.

If a game can predict well against the spread it MUST be good at predicting outright winners. If it can't do the latter, then the former is pretty much impossible. both are predicting game outcomes, but predicting against the spread is just doing it with a great level of accuracy.

And yes, I believe MSU would cover (and win) against more teams than Maryland, as indicated by the metrics.

 

[Money]

If a game can predict well against the spread it MUST be good at predicting outright winners. If it can't do the latter, then the former is pretty much impossible. both are predicting game outcomes, but predicting against the spread is just doing it with a great level of accuracy.

And yes, I believe MSU would cover (and win) against more teams than Maryland, as indicated by the metrics.

Even though they haven't (by a long shot). Classic case of not seeing the forest for the trees.

I'll take Maryland's 5 more wins (and sweep of MSU) as indicated by the facts

 

[TrollyMcTroller]

Even though they haven't (by a long shot). Classic case of not seeing the forest for the trees.

I'll take Maryland's 5 more wins (and sweep of MSU) as indicated by the facts

Take whatever you want. Every metric outside of the RPI agrees that MSU is significantly better than Maryland.

Care to make a last-longer av bet on the tournament?

Tell ya what, let's take all of the teams that my metric rates better than Maryland, but are ranked behind them in the AP (or RPI if you prefer, I'll let you pick) and we'll see if more teams make it farther than Maryland or not. Sound fair?

 

[Money]

Take whatever you want. Every metric outside of the RPI agrees that MSU is significantly better than Maryland.

Care to make a last-longer av bet on the tournament?

Tell ya what, let's take all of the teams that my metric rates better than Maryland, but are ranked behind them in the AP (or RPI if you prefer, I'll let you pick) and we'll see if more teams make it farther than Maryland or not. Sound fair?

No thanks. I'm kinda done with this topic and the word "metric".

Freakin' stat geeks

 

[TrollyMcTroller]

No thanks. I'm kinda done with this topic and the word "metric".

Freakin' stat geeks

We prefer the term "gamblers." (at least according to the metrics I've looked at)

 

[jontaejones]

Take whatever you want. Every metric outside of the RPI agrees that MSU is significantly better than Maryland.

Care to make a last-longer av bet on the tournament?

Tell ya what, let's take all of the teams that my metric rates better than Maryland, but are ranked behind them in the AP (or RPI if you prefer, I'll let you pick) and we'll see if more teams make it farther than Maryland or not. Sound fair?

I'll take that bet and go with the Terps if you tell me what an AV bet is...

 

[TrollyMcTroller]

I'll take that bet and go with the Terps if you tell me what an AV bet is...

If I win I get to pick your avatar (say for a week?) and if you win, you get to pick mine.

I'll let you wait until the final rankings before you pick AP or RPI. I'll be posting my updated data that weekend so we can get the list based on that.

Any teams that exit at the same stage of the tournament (not necessarily the same day) will count as a "push" Any team that makes it farther than Maryland will be a point for me, and any team that doesn't go as far as Maryland will be a point for you. If we end up with the same number of points, then it's a push and neither of us change avatars. Sound fair?

 

[jontaejones]

If I win I get to pick your avatar (say for a week?) and if you win, you get to pick mine.

I'll let you wait until the final rankings before you pick AP or RPI. I'll be posting my updated data that weekend so we can get the list based on that.

Any teams that exit at the same stage of the tournament (not necessarily the same day) will count as a "push" Any team that makes it farther than Maryland will be a point for me, and any team that doesn't go as far as Maryland will be a point for you. If we end up with the same number of points, then it's a push and neither of us change avatars. Sound fair?

If you explained that bet properly, it means that if Maryland makes the Sweet 16, I win. If they go out the first weekend, you win. In a nutshell.

I thought we were going Terps heads up versus Sparty. I'll let you choose either. I'll take either.

 

[TrollyMcTroller]

If you explained that bet properly, it means that if Maryland makes the Sweet 16, I win. If they go out the first weekend, you win. In a nutshell.

I thought we were going Terps heads up versus Sparty. I'll let you choose either. I'll take either.

Uh no.

The bet was basically to highlight the disparity between where I (KenPon, Sagarin and BPI) have Maryland ranked, vs where the polls/RPI have Maryland ranked.

So as things stand now, Maryland is ranked 10th, but I have them ranked 30th. we'd take all the teams between 10th and 30th and and see how far each goes compared to Maryland. If more make it farther than Maryland I win. If more go out sooner than Maryland you win. Based on these number if Maryland made it to the EE ten you'd be guaranteed a win because only 7 other teams could make it that far so the remaining 13 obviously went out sooner. In all likelihood the gap between them will shrink somewhat between now and the tournament so the EE might not clench for Maryland.

But I'm not just going to do a last longer with Maryland and MSU, because 1) I think they're both hot garbage this season, so I don't trust either one enough to bet on them, 2) I don't really like MSU so I don't want to have to cheer for them over some stupid bet and 3) it defeats the whole purpose of the wager I proposed which was to show the overall accuracy of the ratings, not how well the work on one specific matchup.

 

[jontaejones]

Uh no.

The bet was basically to highlight the disparity between where I (KenPon, Sagarin and BPI) have Maryland ranked, vs where the polls/RPI have Maryland ranked.

So as things stand now, Maryland is ranked 10th, but I have them ranked 30th. we'd take all the teams between 10th and 30th and and see how far each goes compared to Maryland. If more make it farther than Maryland I win. If more go out sooner than Maryland you win. Based on these number if Maryland made it to the EE ten you'd be guaranteed a win because only 7 other teams could make it that far so the remaining 13 obviously went out sooner. In all likelihood the gap between them will shrink somewhat between now and the tournament so the EE might not clench for Maryland.

But I'm not just going to do a last longer with Maryland and MSU, because 1) I think they're both hot garbage this season, so I don't trust either one enough to bet on them, 2) I don't really like MSU so I don't want to have to cheer for them over some stupid bet and 3) it defeats the whole purpose of the wager I proposed which was to show the overall accuracy of the ratings, not how well the work on one specific matchup.

Let's say 5 teams ahead of Maryland in RPI (it will probably be more this year) make the Sweet 16 and MD does too. That leaves 10 spots left in the Sweet 16. Even if all of those 10 come from the 11-30 range (which is impossible), the other 10 from that 10-30 range will have bowed out before then costing you 10 points, and you cannot win.

 

[TrollyMcTroller]

Let's say 5 teams ahead of Maryland in RPI (it will probably be more this year) make the Sweet 16 and MD does too. That leaves 10 spots left in the Sweet 16. Even if all of those 10 come from the 11-30 range, the other 10 from that 10-30 range will have bowed out before then costing you 10 points, and you cannot win.

Teams ahead of Maryland have no bearing on the bet. The only thing that matter is teams between where I have Maryland ranked and where AP/RPI have Maryland ranked. Yeah, some ahead, and some behind will likely make the sweet 16 too, but there is no way to tell that ahead of time. The only spot (without seeing a bracket, and assuming 20 teams qualify) that would guarantee a win would be EE or better. It's all subject to change vastly depending on the actual bracket.

 

[jontaejones]

Teams ahead of Maryland have no bearing on the bet. The only thing that matter is teams between where I have Maryland ranked and where AP/RPI have Maryland ranked. Yeah, some ahead, and some behind will likely make the sweet 16 too, but there is no way to tell that ahead of time. The only spot (without seeing a bracket, and assuming 20 teams qualify) that would guarantee a win would be EE or better. It's all subject to change vastly depending on the actual bracket.

I don't think you realize how much of an advantage you're giving up. Even if MD only makes the round of 32, I still have the edge.

 

[TrollyMcTroller]

I don't think you realize how much of an advantage you're giving up. Even if MD only makes the round of 32, I still have the edge.

Depends on the bracket where everyone is seeded, and how big of a gap between my ratings and the real ones.

But you have no idea how much edge you have until a bracket comes out. Anything at this point is beyond a wild ass guess. That said, it's a fucking avatar bet. I'm not really sweating a loss.