I think a win over B helps you more and a loss to A would hurt you more.

Posted by wesmike on 8/2/2020 10:44:00 PM (view original):
Since, as oldwarrior mentioned, opponents' records is 50%, if your own record/RPI is good enough it could make the 104 RPI jump into the top 100 even if they lose to you (also dependent upon what teams in the 90 through 110 range do on that given night). That could give you a top 100 win even though when you played them they were not in the top 100. In general, it's kind of a silly question taking it out of context of the season at large.

The last sentence of this post is spot on in my opinion.

So far mully and i have gotten along during this debate, and that's cool. I just feel it's not a very important thing for discussion purposes. For the record, he was trying to explain a point. And was NOT trying to say "you're killing me Top. I'm not gonna make the NT because of you". So it's not really a big deal in the end. I was just trying to understand if I was missing a point that I was unaware of.

Ultimately if my rpi, sos, and projection numbers are all better than an opposing school, I feel I'm doing the right thing. Maybe I scheduled too hard, maybe not. But again, it's only season 3 for me. All I did was accept requests, and set up a home and home with Kentucky. That's a MUST being at Louisville! I didn't put a lot of effort into the scheduling at this stage. Hopefully going forward this won't matter, as I plan to be 1/1/1 for the rest of my career there anyways! (We can all hope, right?)

Posted by mullycj on 8/2/2020 11:30:00 PM (view original):

Posted by shoe3 on 8/2/2020 10:14:00 PM (view original):

Posted by mullycj on 8/2/2020 7:48:00 PM (view original):

Posted by oldwarrior on 8/2/2020 7:30:00 PM (view original):
if you're looking at strictly RPI the opponent record makes up 50%, and the opponent opponents record is 25%. In that case the RPI should improve more with a win over 15-2 record. Long ways to the end of the season for those records to move.

^^^ This was the answer I was waiting for.
Team A has an 88% winning %.
Team B has a 53% winning %.

RPI is STILL the biggest driver for projection report placement and I don't believe the "quality opponent" bump you get for playing Team B offsets the 35% difference in winning %.

If the question is *at a specific point midway through the season*, what win is going to move you higher up the PR, I still think the answer is B. At game 18, notching another top 100 RPI win will have a bigger impact, I think, than the amount the opponent’s record changes those SOS/RPI formulas; in other words, the quality opponent bump is direct.

Now there is lots of season left, and as I understand it, team A could end up being a top 100 team, and thus end up eventually counting as a top 100 or even top 50 win, which would change the equation. But the point at which you ask the question, i think B is doing more for you.

Yes shoe - just assuming a single point in time. Not trying to run assumptions on the rest of the season.

I ran the numbers through last night's games.
Current RPI .5797 (#60) (playing team B)

Substituting Cincinnati for Louisville on my schedule and assuming it was a win.
Hypothetical NEW RPI .5914 (#51) (playing team A)

So some of you think ONE win over a top 100 team (B) gains the equivalent of 9 spots on the projection report?
I just don't think it has that much of an impact.

I don't think one win over a top 100 team is equivalent to 9 spots on the projection list. But I also don't believe having an rpi of .5914 is "better" than having an rpi of .5797, if we're leaving out ALL the rest of the context.

There's been plenty of times I've been higher on the projection list than other teams that have a higher rpi than I do. It's all one big bowl of oats and rpi is just a piece of the puzzle.

I think it’s been said repeatedly but I am in concurrence that since you had the higher projection number, it would be more beneficial to beat Team B, RPI and SOS are more window dressing at this point. They determine who makes the postseason from the projection report nothing else.

Also, what no one else has said, which I think validates your approach, imagine if you won a few more games instead of being 9-8, your RPI and Projection Report would’ve been much higher, if you played all crappy teams and was like the 15-2 squad, then you could’ve won what two more meaningless games, I don’t think that helps much. Lot more room for error in your approach, the other way, you’d HAVE to win almost every game and still not be better off in my opinion.

Rowle - I still think a couple of you don't understand the question (especially based on your last response). I am not asking if you should schedule as Team A or Team B.

I am asking that as "Team C" would you rather have a conference full of Team As or Team Bs.

I hear what you are saying Top. We are just not aligned. Its hard to prove mathematically either way.

Posted by mullycj on 8/3/2020 10:23:00 AM (view original):
Rowle - I still think a couple of you don't understand the question (especially based on your last response). I am not asking if you should schedule as Team A or Team B.

I am asking that as "Team C" would you rather have a conference full of Team As or Team Bs.

I hear what you are saying Top. We are just not aligned. Its hard to prove mathematically either way.

I actually do understand the point you're getting at more clearly. But to be completely honest with you, and I'm not being a butthead..... I just feel it's irrelevant. If my conference opponents all went 0-10 in non con, but had SOS 1 thru 11 respectively, that's fine by me. Personally, I don't want the teams that go 10-0 with rpi 200. Because as it's stated, you get no love for a win against them. In the big picture. Maybe rpi sure. But again it doesn't go by just rpi.

I don't think my conference has much to do at all with what I'm doing. And I don't wanna "ruin" anything for my conference guys. But I don't care how they fair to be honest. I'll root for em, but if they stink they stink. (Not THIS conference. But any conference I play). To worry about what others are doing in HD is just something I've never even thought about before. Until this topic was brought to my attention.

I will promise you tho mully, I'll be good enough to hold my own! And I won't be holding anyone back that around me. Just give me more than 2.5 seasons to prove that.

Posted by wesmike on 8/2/2020 10:44:00 PM (view original):
Since, as oldwarrior mentioned, opponents' records is 50%, if your own record/RPI is good enough it could make the 104 RPI jump into the top 100 even if they lose to you (also dependent upon what teams in the 90 through 110 range do on that given night). That could give you a top 100 win even though when you played them they were not in the top 100. In general, it's kind of a silly question taking it out of context of the season at large.

opponent's record does not include those games against you, so that line of inquiry is irrelevant imo. you do count in opponents opponents record though

Posted by topdogggbm on 8/3/2020 11:03:00 AM (view original):

Posted by mullycj on 8/3/2020 10:23:00 AM (view original):
Rowle - I still think a couple of you don't understand the question (especially based on your last response). I am not asking if you should schedule as Team A or Team B.

I am asking that as "Team C" would you rather have a conference full of Team As or Team Bs.

I hear what you are saying Top. We are just not aligned. Its hard to prove mathematically either way.

I actually do understand the point you're getting at more clearly. But to be completely honest with you, and I'm not being a butthead..... I just feel it's irrelevant. If my conference opponents all went 0-10 in non con, but had SOS 1 thru 11 respectively, that's fine by me. Personally, I don't want the teams that go 10-0 with rpi 200. Because as it's stated, you get no love for a win against them. In the big picture. Maybe rpi sure. But again it doesn't go by just rpi.

I don't think my conference has much to do at all with what I'm doing. And I don't wanna "ruin" anything for my conference guys. But I don't care how they fair to be honest. I'll root for em, but if they stink they stink. (Not THIS conference. But any conference I play). To worry about what others are doing in HD is just something I've never even thought about before. Until this topic was brought to my attention.

I will promise you tho mully, I'll be good enough to hold my own! And I won't be holding anyone back that around me. Just give me more than 2.5 seasons to prove that.

i do want the teams who go 10-0 with 200 rpi in my conference. the impact of non conf play on you is huge. by the NT, each team has played an average of 28 games or so, 18/28 being in conf. thats 64% of your opponent and opponents opponents records that comes from your conf mates. in general, the in-conf play is a wash (divisions make that not really accurate, but probably close enough) because every in-conf win is offset by an in-conf loss somewhere else in the conf. this is a big part of why most RPIs fall in a relative small segment of the 0-1 RPI range - too many guaranteed win/loss pairs on your opponents records and opponents opponents records.

so, of the 75% of your RPI which is SOS, about 2/3rds comes from conf play, and most of it is meaningless BS - wins and losses offset in conf. the only controllable part is non conf record within your conf. having a great non conf record is huge, even if non conf rpi/sos sucks. its not that hard to schedule 10-0 records against decent teams from crap conferences either, resulting in pretty solid rpi/sos. i'm pretty sure those teams are doing a lot more for the conf that the 5-5 guy who has a 10 rpi.

in the end - i come down on mully's side - i'd rather beat team A. sort of assuming you have a solid schedule, and are adding this game to it - if you don't have hardly any top 100 rpi wins, i could see an argument for needing to beat team B more. i don't think either team is being a 'bad conf mate' or anything though. both seem like decent but not great wins.

Posted by rowle1js on 8/3/2020 7:47:00 AM (view original):
I think it’s been said repeatedly but I am in concurrence that since you had the higher projection number, it would be more beneficial to beat Team B, RPI and SOS are more window dressing at this point. They determine who makes the postseason from the projection report nothing else.

Also, what no one else has said, which I think validates your approach, imagine if you won a few more games instead of being 9-8, your RPI and Projection Report would’ve been much higher, if you played all crappy teams and was like the 15-2 squad, then you could’ve won what two more meaningless games, I don’t think that helps much. Lot more room for error in your approach, the other way, you’d HAVE to win almost every game and still not be better off in my opinion.

if you want to call rpi and sos window dressing, you have to acknowledge that a substantial (majority?) portion of the projection report operates off of extremely similar underlying constructs as rpi and sos.

rpi and sos are not window dressing like assists, which has some correlation to things like passing and such but reflects absolutely nothing from the underlying game.

rpi and sos are more like individual rebounds. while individual rebounds are largely a product of the assigning of team rebounds to individual, they aren't totally disconnected from what is happening under the hood. you can't draw a direct line between a player's rebounding talents and their individual rebounding stats, but the two have an awful lot in common, and over a long enough sample size, are pretty heavily correlated. are individual rebounds a little bit window dressing-y? sure, but its also one of the best measures we get for relating to how the sim engine evaluates rebounding. same goes for rpi/sos. are they a little window dressing-y? sure, its also one of the best measures we get for understanding how the game engine evaluates team success.

Posted by gillispie1 on 8/3/2020 11:22:00 AM (view original):

Posted by wesmike on 8/2/2020 10:44:00 PM (view original):
Since, as oldwarrior mentioned, opponents' records is 50%, if your own record/RPI is good enough it could make the 104 RPI jump into the top 100 even if they lose to you (also dependent upon what teams in the 90 through 110 range do on that given night). That could give you a top 100 win even though when you played them they were not in the top 100. In general, it's kind of a silly question taking it out of context of the season at large.

opponent's record does not include those games against you, so that line of inquiry is irrelevant imo. you do count in opponents opponents record though

I'm not 100% sure what you're saying, but I think you may have missed what I was saying. Your record absolutely does count in their opponents record (that middle 50%). So depending on the point in the season and relative to what other teams around them do at the same time,their RPI could still jump from 104 into the top 100 even if they lose to you.

Non-numerical (overly simplistic) example:
Your team: A 15-2 #104 RPI team loses to a 13-4 #30 RPI team
Other teams: Teams #99 - #103 in RPI lose to 13-4 #150ish RPI teams
It is easily conceivable that your team (#104) jumps into the top 100 because your SOS improves and the other team's SOS worsens.

Posted by gillispie1 on 8/3/2020 11:33:00 AM (view original):

Posted by topdogggbm on 8/3/2020 11:03:00 AM (view original):

Posted by mullycj on 8/3/2020 10:23:00 AM (view original):
Rowle - I still think a couple of you don't understand the question (especially based on your last response). I am not asking if you should schedule as Team A or Team B.

I am asking that as "Team C" would you rather have a conference full of Team As or Team Bs.

I hear what you are saying Top. We are just not aligned. Its hard to prove mathematically either way.

I actually do understand the point you're getting at more clearly. But to be completely honest with you, and I'm not being a butthead..... I just feel it's irrelevant. If my conference opponents all went 0-10 in non con, but had SOS 1 thru 11 respectively, that's fine by me. Personally, I don't want the teams that go 10-0 with rpi 200. Because as it's stated, you get no love for a win against them. In the big picture. Maybe rpi sure. But again it doesn't go by just rpi.

I don't think my conference has much to do at all with what I'm doing. And I don't wanna "ruin" anything for my conference guys. But I don't care how they fair to be honest. I'll root for em, but if they stink they stink. (Not THIS conference. But any conference I play). To worry about what others are doing in HD is just something I've never even thought about before. Until this topic was brought to my attention.

I will promise you tho mully, I'll be good enough to hold my own! And I won't be holding anyone back that around me. Just give me more than 2.5 seasons to prove that.

i do want the teams who go 10-0 with 200 rpi in my conference. the impact of non conf play on you is huge. by the NT, each team has played an average of 28 games or so, 18/28 being in conf. thats 64% of your opponent and opponents opponents records that comes from your conf mates. in general, the in-conf play is a wash (divisions make that not really accurate, but probably close enough) because every in-conf win is offset by an in-conf loss somewhere else in the conf. this is a big part of why most RPIs fall in a relative small segment of the 0-1 RPI range - too many guaranteed win/loss pairs on your opponents records and opponents opponents records.

so, of the 75% of your RPI which is SOS, about 2/3rds comes from conf play, and most of it is meaningless BS - wins and losses offset in conf. the only controllable part is non conf record within your conf. having a great non conf record is huge, even if non conf rpi/sos sucks. its not that hard to schedule 10-0 records against decent teams from crap conferences either, resulting in pretty solid rpi/sos. i'm pretty sure those teams are doing a lot more for the conf that the 5-5 guy who has a 10 rpi.

in the end - i come down on mully's side - i'd rather beat team A. sort of assuming you have a solid schedule, and are adding this game to it - if you don't have hardly any top 100 rpi wins, i could see an argument for needing to beat team B more. i don't think either team is being a 'bad conf mate' or anything though. both seem like decent but not great wins.

I always tell my conference rivals they need to run their programs the way the see fit. Half the time - at least - I forget to schedule the rest of my slots after accepting most challenges, so especially if I didn’t get a lot of challenges in a given season, I am sometimes that team with the shiny record and shaky PR midway through the season. It can be noticeable in a conference with more than a couple sims. By the end of the season, if the team is a legitimately good team, it will usually shake out. But then by the end of the season, a single top 100 win has less direct impact than it did half-way through, and that’s what I’m getting at.

Posted by wesmike on 8/3/2020 1:36:00 PM (view original):

Posted by gillispie1 on 8/3/2020 11:22:00 AM (view original):

Posted by wesmike on 8/2/2020 10:44:00 PM (view original):
Since, as oldwarrior mentioned, opponents' records is 50%, if your own record/RPI is good enough it could make the 104 RPI jump into the top 100 even if they lose to you (also dependent upon what teams in the 90 through 110 range do on that given night). That could give you a top 100 win even though when you played them they were not in the top 100. In general, it's kind of a silly question taking it out of context of the season at large.

opponent's record does not include those games against you, so that line of inquiry is irrelevant imo. you do count in opponents opponents record though

I'm not 100% sure what you're saying, but I think you may have missed what I was saying. Your record absolutely does count in their opponents record (that middle 50%). So depending on the point in the season and relative to what other teams around them do at the same time,their RPI could still jump from 104 into the top 100 even if they lose to you.

Non-numerical (overly simplistic) example:
Your team: A 15-2 #104 RPI team loses to a 13-4 #30 RPI team
Other teams: Teams #99 - #103 in RPI lose to 13-4 #150ish RPI teams
It is easily conceivable that your team (#104) jumps into the top 100 because your SOS improves and the other team's SOS worsens.

yeah you are right, i took what you said differently (wrongly) the first time. what you are suggesting is definitely possible.

i get that the context will change over the season and hence the desire from several folks to keep it open ended, but it also makes it pretty hard to address the question. i prefer to rephrase it as 'assume you already beat both of these teams, which of those wins is more valuable to you'. keeps it a lot tidier.

I am going to fall on the side of topdog here. While I think RPI is a somewhat valid method for determining team strength, I do not believe the projection report, and more importantly NT seeding decisions, always track closely with RPI. The next time tournament seeds are announced, compare the range of RPIs of the Last Four In to those of the Last Four Out. One or two of the teams on both of those lists will make you scratch your head.

We've all seen this on numerous occasions, and many of us have complained about the same. The fact is that the methodology for determining the projection report and subsequent seedings is opaque enough that we still have arguments about it many years after it first debuted.

Personally, in conference play I would rather face the teams with the more difficult schedules, as I think they are more representative of the teams I'll eventually be seeing in the NT. I learn very little about my own team by beating a 15-3 team with a 200 SOS.

Posted by gillispie1 on 8/3/2020 11:33:00 AM (view original):

Posted by topdogggbm on 8/3/2020 11:03:00 AM (view original):

Posted by mullycj on 8/3/2020 10:23:00 AM (view original):
Rowle - I still think a couple of you don't understand the question (especially based on your last response). I am not asking if you should schedule as Team A or Team B.

I am asking that as "Team C" would you rather have a conference full of Team As or Team Bs.

I hear what you are saying Top. We are just not aligned. Its hard to prove mathematically either way.

I actually do understand the point you're getting at more clearly. But to be completely honest with you, and I'm not being a butthead..... I just feel it's irrelevant. If my conference opponents all went 0-10 in non con, but had SOS 1 thru 11 respectively, that's fine by me. Personally, I don't want the teams that go 10-0 with rpi 200. Because as it's stated, you get no love for a win against them. In the big picture. Maybe rpi sure. But again it doesn't go by just rpi.

I don't think my conference has much to do at all with what I'm doing. And I don't wanna "ruin" anything for my conference guys. But I don't care how they fair to be honest. I'll root for em, but if they stink they stink. (Not THIS conference. But any conference I play). To worry about what others are doing in HD is just something I've never even thought about before. Until this topic was brought to my attention.

I will promise you tho mully, I'll be good enough to hold my own! And I won't be holding anyone back that around me. Just give me more than 2.5 seasons to prove that.

i do want the teams who go 10-0 with 200 rpi in my conference. the impact of non conf play on you is huge. by the NT, each team has played an average of 28 games or so, 18/28 being in conf. thats 64% of your opponent and opponents opponents records that comes from your conf mates. in general, the in-conf play is a wash (divisions make that not really accurate, but probably close enough) because every in-conf win is offset by an in-conf loss somewhere else in the conf. this is a big part of why most RPIs fall in a relative small segment of the 0-1 RPI range - too many guaranteed win/loss pairs on your opponents records and opponents opponents records.

so, of the 75% of your RPI which is SOS, about 2/3rds comes from conf play, and most of it is meaningless BS - wins and losses offset in conf. the only controllable part is non conf record within your conf. having a great non conf record is huge, even if non conf rpi/sos sucks. its not that hard to schedule 10-0 records against decent teams from crap conferences either, resulting in pretty solid rpi/sos. i'm pretty sure those teams are doing a lot more for the conf that the 5-5 guy who has a 10 rpi.

in the end - i come down on mully's side - i'd rather beat team A. sort of assuming you have a solid schedule, and are adding this game to it - if you don't have hardly any top 100 rpi wins, i could see an argument for needing to beat team B more. i don't think either team is being a 'bad conf mate' or anything though. both seem like decent but not great wins.

I have two things to add...... gil, please go look at the Smith Big East. I swear you will change your mind. My Team B is 4-7 against the top 100. Team A is 1-4 against the top 100. You said "....assuming (Team A) had a solid schedule". How could they have a solid schedule if it's ranked 298?! If you'd rather have the 100/300 team in conference because they're a punching bag and you want some easy wins, so be it. But there's no way beating that team helps you as much as beating 65/20 type of team. I feel like this thread is trying to rewrite what is important in HD. And it honestly makes no sense to me at all. It reminds me of the EARLY rise of real life Gonzaga. They were going 25-1 year after year and getting no love. Coming into the NT with a terrible SOS (the 300 team in this case) and getting an 8 seed and getting booted early after all the hype. While a team like Syracuse would come in as an 11 seed with lots of losses, and make a F4. (Not necessarily the same season. I just pulled two examples out of the air).

And the second thing, this is not direct to Gil, this is directed towards the community..... at WORST, this topic has proven to be so divided, that if there is no consensus opinion, it is a topic that doesn't mean squat. I'm not really concerned with what others in my conference are doing. I'm debating "against" the type teams with 100 rpi and 300 sos, and even that doesn't make me feel one way or another. If they choose to play that way that's fine. And I'll choose to play how I play

This is non factor in my opinion