This post could not be converted. To view the original post's thread, click here.

hmmm.

a couple of things... severalof your earlier opponents are continuing to lose every night... that aint helping

also, while your last three opponents may be top25 in the wispoll, only one is anywhere near top25 in rpi. one of them is 6-4 and the other is 7-3 and not against very tough schedules. so, in reality, those are not top25 teams.

however, i am a little bit surprised that your rpi is not better. i think it is likely that your rpi will generally improve as the season goes along.

i hope you are right ol' d.....i thought i might have stayed in the 130-15 range were i was when i started the three games. i think it might be better to beat an averge team than to lose to good or solid teams....

The other issue is that you played them at home. A home loss has more negative impact than a road loss.

This post could not be converted. To view the original post's thread, click here.

here are some rough numbers showing your RPI after playing a game against three different levels of opponents.

your current RPI is .5000 (25% your record, 50% opponent record, 25% opponent/opponent record)
assuming you've played 20 games, your opponents have a combined .500 record and your opponent opponents also have a .500 record
home wins count .6, road win 1.4

playing a team that is 10-10 with a .5000 sos
home LOSS = 4918
home WIN = 5037
road LOSS = 4964
road WIN = 5082

playing a team that is 13-7 with a .5400 sos
home LOSS = 4960
home WIN = 5078
road LOSS = 5005
road WIN = 5123

playing a team that is 17-3 with a .6000 sos
home LOSS = 5014
home WIN = 5132
road LOSS = 5059
road WIN = 5177

this shows that a loss to a top opponent can increase your RPI. But it's generally not going to increase as much as a win against an average team.

Quote: Originally posted by mtngoats on 3/30/2010<SCRIPT language="javascript><!--bluecoat_old_onload" = window.onload;window.onload = function() { if (bluecoat_old_onload != null) { try { bluecoat_old_onload(); } catch (e) { } }bluecoat_fixpopups();};//--></SCRIPT>would it be safe to say a win against an rpi of 150 is better than a loss to a top 75 rpi????  if both played at home or on the road????

yes, safe assumption.

overall, most of the issue is that rpi is meaningless when it first comes out. it is not a decent approximation until a solid 13 games or so. there are a lot of reasons this is true but i would never look at a 40 rpi drop between games 7 and 10 and conclude the last 3 games you played really hurt your rpi.

Quote: Originally Posted By gillispie on 3/30/2010

Quote: Originally posted by mtngoats on 3/30/2010 <SCRIPT language="javascript><!--bluecoat_old_onload" = window.onload;window.onload = function() { if (bluecoat_old_onload != null) { try { bluecoat_old_onload(); } catch (e) { } }bluecoat_fixpopups();};//--></SCRIPT>would it be safe to say a win against an rpi of 150 is better than a loss to a top 75 rpi???? if both played at home or on the road????

yes, safe assumption.

overall, most of the issue is that rpi is meaningless when it first comes out. it is not a decent approximation until a solid 13 games or so. there are a lot of reasons this is true but i would never look at a 40 rpi drop between games 7 and 10 and conclude the last 3 games you played really hurt your rpi

Irrelevant assumption, I think is more accurate.

It has nothing to do with the other team's RPI, it has to do with their W/L record and their opponents' W/L record. Often the 75 will be better, but not always.

isack, your point is right about "it has to do with their W/L record and their opponents' W/L record". but when you say it has nothing to do with their RPI, that is pretty misleading. that is 75% of their RPI formula. opponents opponents win % is probably the most moderate of the 3 components (smallest std. deviation, i would guess). so rpi is a decent approximation. i don't think irrelevant assumption is very accurate at all.

Quote: Originally Posted By gillispie on 3/30/2010isack, your point is right about "it has to do with their W/L record and their opponents' W/L record". but when you say it has nothing to do with their RPI, that is pretty misleading. that is 75% of their RPI formula. opponents opponents win % is probably the most moderate of the 3 components (smallest std. deviation, i would guess). so rpi is a decent approximation. i don't think irrelevant assumption is very accurate at all

I see what you're saying, but in reality, it's irrelevant insofar as its affect on RPI.

It's just as easy to look at your current opponent's W/L record and SOS to see how much it will help or hurt. At least in my opinion. I agree that it's generally going to be the case, but I just can't stand the "I beat such and such team with this RPI and my RPI didn't go up enough" complaints. I think it's important for people to understand that opponents' RPI has zero effect on your RPI.

all good and apprciated info