BB G3 Final: Louisiana 2, Marshall 0

[GoneGolfin]

Are you stating that the RPI formula used for college baseball and softball are different? If so, that’s a head scratcher for sure.

Certainly they are different. All sports that used the RPI have had their nuances. Baseball and Softball were closer at one time (but certainly not the same). Now they are significantly different.

Brian

[GoneGolfin]

50% is opponents records and they’re all affected by us beating or sweeping them.


Marshall is the only conference team we’ve played that wouldn’t be above .500 if we didn’t play them.


Edit. ASU would be 16-18 if you take out our games. 17-20 currently.

Just to be clear ... our performance against our opponents does not affect their OWP contribution to our RPI. As such, Marshall contributes .38710 (12-19 record ... not 12-22) (with a game weighting of 3) to our OWP. Also, you sum/weight the winning percentages in determining OWP ... not a summation/weighting of the w-l records themselves. As an example, an opponent with a 2-4 record would have the same contribution as a 20-40 opponent (same # of games played).

Finally, not to make any math-averse heads explode ... but the end result in the RPI calculation is that winning is just over 25% of the formula ... which also means that SOS (and thus OWP and OOWP) is a little less than 75%. This is due to the fact that our own winning percentage contributes to our OOWP.

Brian

[Jacob81]

Thanks for clearing that up. What percentages do think the 3 factors should be?? 33.3% for each? 1/3

[GoneGolfin]

Thanks for clearing that up. What percentages do think the 3 factors should be?? 33.3% for each? 1/3

I do not think there should be 3 factors. Burn the whole thing to the ground.

Brian

[Cajunsmike]

I do not think there should be 3 factors. Burn the whole thing to the ground.

Brian

Brian has long advocated for dumping the RPI. That likely won't happen soon. The 25/50/25 split is obviously out dated. Cajuns winning the league and having an RPI in the 50's is possible and evidence of the flaws in the current system. Therefore, if you have to have 3 factors, I propose the split be 40% your perfomance, 35% your opponents performance, and 25 % the performance of the opponents of your opponents. That way the thing you control the most has the most impact. It boggles my mind that the only thing you have control over (your perfomance) is rated equal to something you have virtually no control over (opponents of opponents) in the current system. I would like to see someone recalculate the RPI rankings based upon my proposal. I know I am unlikely to see that result anytime soon.

[Jacob81]

I do not think there should be 3 factors. Burn the whole thing to the ground.

Brian

And use what? The eye test?
We know how much commitees(humans) love mid majors.

[GoneGolfin]

And use what? The eye test?
We know how much commitees(humans) love mid majors.

No, not the eye test. I am not going to get into alternatives here ... as I have discussed this ad nauseum over the years (both possibilities I have suggested as well as analysis of other systems). I was simply highlighting that you do not put lipstick on the pig. The RPI system is ridiculously trivial and not statistically sound. It is not something you try to fix as it does not have a proper foundation.

Brian

[Jacob81]

No, not the eye test. I am not going to get into alternatives here ... as I have discussed this ad nauseum over the years (both possibilities I have suggested as well as analysis of other systems). I was simply highlighting that you do not put lipstick on the pig. The RPI system is ridiculously trivial and not statistically sound. It is not something you try to fix as it does not have a proper foundation.

Brian

Gotcha. Well if the ncaa thinks the current system is the best way to get the most big schools in the postseason then they’ll be in no hurry to change it. I will say there have been plenty of mid major hosts over the years.