3-3 against the RPI top 25, undefeated against everybody else (but no real signature wins in that bunch either), where should you be? 

I'm in this position mid-season in WIS, and I still have time to move up or down, so this isn't me complaining about my seed but rather trying to figure out how the projection (which I think is usually very good) matches onto what we think it should. And if it doesn't, where the discrepancy lies. 

More details. . . 

Signature wins: 
*10-point home win over RPI #8
*2-point home win over RPI #10
*OT road win over RPI #10

Losses:
*9-point road loss to RPI #24
*1-point home loss to RPI #23
*2-point home loss to RPI #7

As far as I can tell, the best real-life comparison is San Diego State, who went 3-3 against top 25 teams and didn't beat anyone else in the top eighty. They did have a loss outside the top 100, but other than that the resumes look pretty similar. They're a four. Does four sound about right in this situation? 

26-3 after winning the CT?

This team is similar and they are  21 in the projection report.

http://www.whatifsports.com/hd/TeamProfile/Schedule.aspx?tid=12146

This is mid-season. I said that, but I guess it would've been important to say how many games midseason. 16 games deep. 

That Ferris State team is a good comparison though. Not quite as good on the quality of best wins (only two against the top 25, which I think is one of the major cutoffs used by the PR) but has more wins in the 25-50 range. 

My Colorado team (the one being used as the example) is currently #28 in the projection report, which is quite a bit lower than I would've expected. The reason is my own fault. I accidentally left half my non-conference scheduling to the whims of the Sim, and they decided to fill out the non-con schedule to include eight home games and only two road games. Even worse, three of the opponents added by the Sim have records of 3-13 and worse and 250+ RPIs. 

I knew in advance that having a Sim-scheduled non-con that included some really bad teams would completely tank my RPI. And it did. But I expected that performing well against the top half of my schedule (and nearly 40% of my games have been against top 25 teams, against which I have a .500 record) would make up for it. It hasn't, at least not to the degree I expected. 

Let's take some further samples. So our first one: 

3-3 vs RPI top 25
0-0 vs 26-50
0-0 vs 51-100
10-0 vs 101+
PR #28

And then. . . 

1-2 vs RPI top 25
1-0 vs 26-50
4-1 vs 51-100
7-0 vs 101+
PR #18

0-1 vs RPI top 25
1-0 vs 26-50
3-0 vs 51-100
10-1 vs 100+ 
PR #26

0-3 vs RPI top 25
2-2 vs RPI 26-50
1-0 vs RPI 51-100
8-0 vs 101+ 
PR #23

I think what we're learning here is that either home/road plays a big difference (I didn't check the home/road splits in those other samples) or the projection report still has a fairly heavy RPI component, albeit not as big as back in the day when it was ridiculously RPI-driven. 

Unless someone else can see another explanation in those numbers. 

And when I say "learning," of course, I might just mean "confirming," depending on what you already knew about the PR. I honestly hadn't paid this much attention. 

(and after getting my *** kicked by the seventh top 25 team I played, I'd like to reiterate that this is a discussion of how the PR treats various resumes and NOT me complaining that my team was underrated). 

tarvolon, it's actually not a home/road thing or an RPI thing. It's like SMU in real life this year -- if you play a bunch of really awful teams, the committee is going to penalize you.

(NOTE: I think SMU should've made the tourney.)

girt, that sounds exactly like an RPI thing to me. Or else a penalty that exactly duplicates penalties in the RPI. 

I've always thought it was a little stupid that there is a penalty for playing bad teams as opposed to average ones--I care a lot more about whether you can play with the big boys, not whether your non-con gimme games were against teams in the mid-100s or in the mid-300s--but I've viewed that as a criticism against the RPI, not against any independent criterion the committee uses.