Started By
Message

NCAA Regional Host Candidate Comparison

Posted on 5/25/22 at 11:44 pm
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/25/22 at 11:44 pm
Nothing that I include here should be taken as my prediction of how the committee will decide hosts. It’s simply a comparison of key metrics that I think should go into the decision making process. Some of it may reflect what the committee considers. Much of it probably doesn’t.

First, I’m going to go ahead and lay aside the 6 teams at the top who are clearly hosting (though an argument can be made that Maryland’s resume is pretty thin.) The others (Tennessee, Oregon St, Virginia Tech, Miami and Stanford) are all in the top 2 of Big 4 conferences and in the top 7 of the RPI.

Next, I’m going to look at teams with winning conference records in the Big 4 conferences (SEC, ACC, Big 12, PAC-12) and in the Top 5 in the conference standings. That leaves out teams in the RPI Top 20 like Vanderbilt, Wake Forest, Georgia, UNC, and Florida. For the teams meeting those criteria, I am going to look at weekend record against Big 4 conference teams (plus any other series they lost) and number of weekend series won against the Big 4 vs weekend series lost. (I have updated these numbers to include conference tournament games completed through May 25.)

Weekend W-L vs Big 4 (plus series lost):
UCLA 24-13 (.649)
7) Texas 23-14 (.622)
8) Notre Dame 18-12 (.600)
9) TA&M 23-16 (.590)
9) UNC 23-16 (.590)
11) Okla St 20-15 (.571)
11) Okla 20-15 (.571)
13) Lvlle 19-15 (.559)
Tx Tech 16-13 (.552)
14) TCU 18-15 (.545)
Arkansas 18-15 (.545)
15) Florida 22-19 (.537)
Virginia 17-15 (.531)
16) LSU 19-17 (.528)
16) Auburn 19-17 (.528)
Oregon 21-19 (.525)

Weekend Series W-L vs Big 4 (plus other L):
UCLA 8-3
7) Okla St 7-3
8) TA&M 8-4
Tx Tech 6-3
9) Texas 7-4
Oregon 7-4
Arkansas 6-4
Virginia 6-4
10) Notre Dame 6-4
11) Oklahoma 6-4
12) UNC 7-5
12) Florida 7-5
14) Lville 5-4-1
15) LSU 6-5
16) Auburn 6-6
TCU 5-5

vs RPI Quad 1 (Excluding Midweek):
7) UCLA 11-4 (.733)
8) ND 14-7 (.667)
9) Texas 15-10 (.600)
10) TA&M 14-10 (.583)
10) Lvlle 14-10 (.583)
12) Auburn 12-10 (.545)
12) UNC 18-15 (.545)
14) Okla 11-11 (.500)
15) Okla St 12-13 (.480)
Virginia 11-12 (.478)
Arkansas 11-13(.458)
16) Florida 13-16 (.448)
17) LSU 12-15 (.444)
Tx Tech 8-10 (.444)
TCU 8-12 (.400)
Oregon 5-8 (.385)

Other Weekend vs Big 4 (+series lost):
7) UNC 5-1 (.833)
8) Okla St 8-2 (.800)
9) LSU 7-2 (.778)
Arkansas 7-2 (.778)
10) TCU 10-3 (.769)
11) Florida 9-3 (.750)
Tx Tech 8-3 (.727)
12) Okla 9-4 (.692)
13) Texas 8-4 (.667)
Virginia 6-3 (.667)
14) TA&M 9-6 (.600)
Oregon 16-11 (.593)
UCLA 13-9 (.591)
15) Lvlle 5-5 (.500)
15) Auburn 7-7 (.500)
17) Notre Dame 4-5 (.444)

This next number is a little more difficult to explain/follow. The idea is to give a consistent weight to games vs Q1 vs the games in the other category. The idea is keep teams from padding the WPct with more non-Q1 games. On average, these teams played 63% of games in the above two categories against Q1, so I gave 63% weight to WPct against Q1 and 37% weight to the WPct against the Other category.

Weighted WPct:
UCLA (.649)
7) UNC (.645)
8) Texas (.623)
9) Okla St (.591)
10) ND (.590)
11) TA&M (.589)
Arkansas (.569)
12) Okla (.567)
13) LSU (.560)
14) Lvlle (.554)
15) Florida (.553)
Virginia (.544)
Tx Tech (.543)
16) Auburn (.530)
17) TCU (.528)
Oregon (.462)

What’s very interesting about this is UCLA at the top of those lists but #49 in the RPI. That is driven by 6 midweek losses. Texas Tech is near the top but #34 RPI due to also having 6 midweek losses. TCU has 5 midweek losses but won the Big 12. How much weight you give midweek influences how you feel about those teams. Overall, you can see these teams are very tightly bunched with most having 17-19 wins and 13-15 losses. I think that has a lot to do with the odd ordering of teams in the RPI (especially when compared against conference standings) and how much fluctuation there is from week-to-week.

Finally, I’m going to look at the minor conference teams likely under consideration to host. It’s hard to draw a direct comparison with the major conference teams because their resumes are so different. Most of the major conference host contenders had 3-5 weekend series losses. To be a legitimate contender against far weaker schedules, these teams should have no more than 3 lost weekend series. Dallas Baptist did well in series against Maryland and USM (5-1) and went 5-5 midweek against the Big 12, but they lost a whopping 6 weekend series and finished 3rd in their conference. Georgia Southern lost 4 weekend series (including getting swept by Tennessee), lost their conference and went 2-4 midweek against Big 4 conferences. RPI #18 East Carolina went 2-2 on the weekend against major conferences and has 4 weekend series losses. That leaves USM with 3 series losses, Gonzaga with 2, Texas St with 1, and UC-Santa Barbara with only a tied series.

USM- #9 RPI, 0-3 vs DBU, 2-1 midweek vs SEC, 3 series losses

Gonzaga- #26, 4-2 vs Big 4 in weekend series, 4-4 midweek vs Big 4, 2 series losses

Texas St- #25, 2-1 vs Arizona, 2-2 midweek vs Big 4, 1 series loss

UCSB- #42, 2-2 vs Oregon, 0-4 midweek vs PAC-12, 1 series tie.

At the end of the day, I think LSU is in the running due to conference rank, but their resume is really in the lower half of the pack at the moment. I don’t think they are a lock.

ETA: Changed rankings to start at 7 and added UCLA loss.

ETA 5/26: Updated thru games of 5/26 (pending LSU-UK)

ETA 5/27 AM: Added Q1 and other categories. Updated LSU.

ETA 5/28: updated thru games of 5/27. Added UNC. Did not number teams likely out of running due to RPI.

ETA 5/29 AM: Updated thru games of 5/28 (pending UCLA). Added Florida.
This post was edited on 5/29/22 at 12:03 am
Posted by Hold That Tiger 10
Member since Oct 2013
21075 posts
Posted on 5/26/22 at 12:17 am to
So you weighted all games that people play against 4 conferences equally? Some conferences have absolute dog shite teams at the bottom of them compared to the SEC. Of course when you stack up the numbers that way, SEC schools will not come out looking as pretty record wise.

RPI isn't the best thing in the world, but the reason it's used, along with SOS, quad 1 wins etc is because it is miles better than your breakdown.

This post was edited on 5/26/22 at 12:52 am
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/26/22 at 3:28 am to
quote:

Some conferences have absolute dog shite teams at the bottom of them compared to the SEC.


It’s really hard to make an argument that getting to that level of detail is really going to be favorable to LSU given they did not face Tennessee and went 6-0 against the two teams that did not make the SEC tournament. The problem with using RPI as you suggest is that it is so dependent on conference games. It’s very difficult to determine how much of a team’s apparent strength or weakness is due to the strength or weakness of the remainder of the conference. Overall, the 4 conferences have RPIs in the same ballpark. Absent significant numbers of non-conference matchups of high quality on weekends, it’s not unreasonable to conclude that conference play evens out more or less.

11 of the 14 teams in the ACC are in the Top 35 of the RPI. BC is terrible. Duke has a poor RPI but outside of conference the only series they lost was to Baylor, who also beat LSU. They had 3 losses outside the RPI Top 100 and 2 non-conference wins over the Top 80. Missouri had 2 losses outside the Top 100 and 2 non-conference wins over the Top 80 and the same 10-20 conference record as Duke. Miss St had 4 losses outside the Top 100 and 1 Top 80 non-conference win.

In the Big 12, Kansas is awful. KSU had 6 losses outside the Top 100 and 2 non-conference wins over the Top 80. They are bad but not materially worse than Miss St.

The PAC-12 suffers from the fact that most of their non-conference opponents are geographically isolated, which makes it hard for the RPI to get a good read on their strength relative to the other side of the country. They and their opponents play mostly one another forcing their RPIs toward .500. That said last place USC had 4 non-conference losses to teams outside the Top 100, but one of those was a Hawaii team with a 27-22 record. They also beat UCSB twice. Utah had 4 non-conference losses outside the Top 100 and 1 Top 80 win like Miss St. ASU had 4 non-conference losses outside the Top 100, but 2 of those were to 36-20 UNLV. They also had 1 Top 80 win. Cal beat TCU and Florida St. WSU beat A&M and Gonzaga.

Overall, I’m seeing no evidence for your assertion that the other conferences are filled with horrible teams at the bottom. BC and Kansas may be materially worse than anyone else. Apart from that, I’m just seeing teams around the same caliber as Missouri and Miss St. If you want to discount those teams, then you might as well throw LSU out of the hosting conversation.
Posted by BayouPride
Member since Sep 2006
527 posts
Posted on 5/26/22 at 6:29 am to
These types of well thought out posts are a welcome change around these parts. Thank you
Posted by Hold That Tiger 10
Member since Oct 2013
21075 posts
Posted on 5/26/22 at 7:13 am to
quote:

The problem with using RPI as you suggest is that it is so dependent on conference games. It’s very difficult to determine how much of a team’s apparent strength or weakness is due to the strength or weakness of the remainder of the conference.


The reason RPI, quad victories, SOS exist is so that they can try to determine the best teams by looking at different metrics. It's pretty impossible to do whenever you have 300 teams and the vast majority of them don't play each other.

It's imperfect, but it's better than just throwing a huge net over every team in major conferences and calling that equal.
This post was edited on 5/26/22 at 7:14 am
Posted by Mobiletiggah
Mobile Alabama
Member since Mar 2021
2688 posts
Posted on 5/26/22 at 7:47 am to
Ok not a lock…agreed. This week will really weigh heavily as to whether LSU moves up high enough to host. But…one game at a time. I think LSU has put themselves in a position to be a host for a regional and possibly a super. It is right in front of them. 3 wins could make it happen….Kentucky first.
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/26/22 at 8:46 am to
quote:

It's imperfect, but it's better than just throwing a huge net over every team in major conferences and calling that equal.


It’s really not better. I agree that some metric that tries to get at strength at a deeper level would be better, but that’s not the RPI. All you need to do is look at the SEC rankings to see the critical flaws of that metric. The 5 teams in the Top 20 are 4 SEC East teams and the one West team with a winning conference record who happened to play Tennessee. Those teams get to count Tennessee’s 49-7 record as part of the 50% weight of opponents’ record. The East teams further get to count Tennessee’s record multiple times in the 25% weight of opponents’ opponents’ record. This is despite Florida being 3-3, Georgia being 2-4, and Vandy being 4-8 against the Top 4 in the West. The bottom 3 in the East were 8-7 against the bottom 3 in the West and 5-16 against the Top 4, so not significantly different in strength to the bottom of the West.

So, if you want to rely on a metric that considers the SEC East materially stronger than the SEC West, that is your choice. In that case, you should conclude that LSU with its relatively poor Q1 record and relatively low RPI should not be in consideration to host.
Posted by Meauxjeaux
98836 posts including my alters
Member since Jun 2005
39975 posts
Posted on 5/26/22 at 8:55 am to
quote:

Weekend W-L vs Big 4 (plus series lost): 1) UCLA 21-12 (.636) 2) Texas 21-13 (.618) 3) Tx Tech 16-11 (.593)


Can you edit and start your numbering at 7?

7) UCLA
8) Texas
9) Tx Tech


Etc… otherwise it was confusing
Posted by Hold That Tiger 10
Member since Oct 2013
21075 posts
Posted on 5/26/22 at 8:57 am to
quote:

All you need to do is look at the SEC rankings to see the critical flaws of that metric. The 5 teams in the Top 20 are 4 SEC East teams and the one West team with a winning conference record who happened to play Tennessee


That's why I used the word "imperfect."

RPI isn't the end-all-be-all to determine the national seeds. It's one tool. Vandy, UGA etc can be ranked high because they played Tennessee but they have zero chance of hosting. There are flaws in the RPI, just like there are flaws in your blanket records.
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/26/22 at 9:11 am to
quote:

Can you edit and start your numbering at 7?


Good suggestion. This has been edited.
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/26/22 at 9:17 am to
quote:

There are flaws in the RPI, just like there are flaws in your blanket records.


I’m not suggesting this is perfect, but it’s better than what you get looking at Q1 and RPI. It’s a starting point for discussion. It would be helpful if you could point to specific examples where you think this approach is overrating or underrating particular teams. I’m sure there are examples that could be used either to tweak the method or at least adjust the way it is interpreted. That would be more helpful in advancing our understanding than simply dismissing it out of hand and concluding that we are stuck with the obviously flawed RPI rankings we have this year.
This post was edited on 5/26/22 at 9:18 am
Posted by Hold That Tiger 10
Member since Oct 2013
21075 posts
Posted on 5/26/22 at 10:33 am to
quote:

I’m not suggesting this is perfect, but it’s better than what you get looking at Q1 and RPI


Disagree completely.

quote:

It would be helpful if you could point to specific examples where you think this approach is overrating or underrating particular teams


As a whole it is not equal. At least RPI, SOS, quad wins are trying to take teams who don't play each other, and in some cases the same common opponents even, and trying to determine which is better.

I think the SEC and ACC are hands down the two best conferences. If you take Ole Miss, Miss St, Mizzou (SEC bottom dwellers this year) and place them in the PAC, or Big 12 then they are more towards the top half of those leagues.

That's the main issue with your blanket records. The vast majority of those games come from conference opponents. The SEC/ACC beat on each other more, thus giving them a worse record. The top of your list is littered with PAC/Big teams, yet the SEC and ACC will dominate in the number of national seeded teams.
Posted by RightWingTiger
Louisiana
Member since Oct 2003
5303 posts
Posted on 5/26/22 at 11:43 am to
quote:

I think LSU has put themselves in a position to be a host for a regional and possibly a super
Short of just blowing out UK, Tenn twice & then say Bama/A&M in the Final I just don’t think LSU has (or will have) done enuf to get a Top 8-Super Reg Host.

I love LSU Baseball & always have BUT this 2022 Team (MY Opinion Only) is just not a Top 8 Baseball Team this year. They’re capable of getting hot & making a mini-run to a Super but this team just isn’t deep enough to make it to Omaha.

Hope Like HELL Im Wrong by the way!!

Posted by Meauxjeaux
98836 posts including my alters
Member since Jun 2005
39975 posts
Posted on 5/26/22 at 12:56 pm to
quote:

Can you edit and start your numbering at 7?


Good suggestion. This has been edited.


Thanks! Helps see quickly where we are in the pecking order.

Nice work and info.
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/26/22 at 3:08 pm to
quote:

I think the SEC and ACC are hands down the two best conferences. If you take Ole Miss, Miss St, Mizzou (SEC bottom dwellers this year) and place them in the PAC, or Big 12 then they are more towards the top half of those leagues.


I actually generally agree with you. The main problem with a Q1 record for LSU is that Ole Miss is Q1 and the other two are not. Using that metric you pull in the 0-3, but lose the 6-0. You also pull in the two losses to La Tech. I‘m open to the idea that some conference wins should be excluded from the comparison. The question is where you draw the line. No matter where you do, it’s going to create some distortion. What consistent cutoff gives LSU credit for a quality series sweep of Missouri but doesn’t include the bottom teams in other conferences? Missouri doesn’t have much else on their resume but 10-20 in the SEC. How do you make the case for them above anyone other than Kansas and BC? Yes, their RPI is higher than some of the teams in other conferences, but that’s more an artifact of the overall inflation of the ratings of the SEC East than a reflection of their own inherent value. It’s cherry picking to take that at face value without doing the same for Vandy, Georgia and Florida.
This post was edited on 5/26/22 at 3:10 pm
Posted by gaetti15
AK
Member since Apr 2013
13365 posts
Posted on 5/26/22 at 4:21 pm to
quote:

I actually generally agree with you. The main problem with a Q1 record for LSU is that Ole Miss is Q1 and the other two are not. Using that metric you pull in the 0-3, but lose the 6-0. You also pull in the two losses to La Tech. I‘m open to the idea that some conference wins should be excluded from the comparison. The question is where you draw the line. No matter where you do, it’s going to create some distortion. What consistent cutoff gives LSU credit for a quality series sweep of Missouri but doesn’t include the bottom teams in other conferences? Missouri doesn’t have much else on their resume but 10-20 in the SEC. How do you make the case for them above anyone other than Kansas and BC? Yes, their RPI is higher than some of the teams in other conferences, but that’s more an artifact of the overall inflation of the ratings of the SEC East than a reflection of their own inherent value. It’s cherry picking to take that at face value without doing the same for Vandy, Georgia and Florida.



let's just get rid of the top team and bottom 3 teams in each conference, then recalculate RPI
Posted by Hold That Tiger 10
Member since Oct 2013
21075 posts
Posted on 5/26/22 at 4:47 pm to
quote:

The main problem with a Q1 record for LSU is that Ole Miss is Q1 and the other two are not


I just threw those three teams out as examples.

quote:

I‘m open to the idea that some conference wins should be excluded from the comparison. The question is where you draw the line


But now you are explaining exactly what splitting into quads does.

quote:

It’s cherry picking to take that at face value without doing the same for Vandy, Georgia and Florida.


I'm not arguing one case because it helps LSU. I'm just talking overall in general, year in and year out.

The bottom line is that there is no perfect system, that will be without flaws. There are too many teams that play college baseball, so that makes for a serious lack of common opponents. What you are suggesting isn't terrible, but it's just one small piece of the puzzle, not the go-to for determining anything. Same way RPI, SOS and everything else is a piece of the puzzle. You take that, put it all together, and try to come up with the 16 best teams. It has flaws, but there just isn't any ways to make it all make perfect sense.


I enjoy the discussion btw. It's not often you get a back and forth on this board where there isn't one side that totally void of any logic, reasoning. So, I may not agree with you, but at least it's coming from an educated reasoning.
This post was edited on 5/26/22 at 4:50 pm
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/26/22 at 6:36 pm to
quote:

But now you are explaining exactly what splitting into quads does.


Understood. The difference is that I’m trying to find a dividing line that more accurately reflects the distribution of teams this season vs an arbitrary static breakpoint. Why is 50 the magic number? I also think regardless of the dividing line, some consideration needs to be given to performance against teams below that cutoff. Oregon, for example, should not be viewed more favorably now that UCLA has dropped to #53 and out of Q1 taking away 3 Q1 losses for Oregon.
Posted by MikeTheTiger71
Member since Dec 2021
2861 posts
Posted on 5/27/22 at 12:29 am to
quote:

I enjoy the discussion btw. It's not often you get a back and forth on this board where there isn't one side that totally void of any logic, reasoning. So, I may not agree with you, but at least it's coming from an educated reasoning.


I have added in a couple of other splits of the data based on your feedback. (Copied below.) I think I found a way to balance performance against Q1 and other games. That’s really where LSU has excelled. They’ve held their own against Q1 and dominated everyone else. A lot of the other contenders have performed well against one group but not the other.

vs RPI Quad 1 (Excluding Midweek):
7) UCLA 10-4 (.714)
8) ND 13-6 (.684)
9) Texas 14-9 (.609)
10) TA&M 13-9 (.591)
11) Lvlle 14-10 (.583)
12) Auburn 12-10 (.545)
13) Okla 11-11 (.500)
13) Virginia 11-11 (.500)
15) Arkansas 11-12 (.478)
16) LSU 12-14 (.462)
17) Okla St 10-12 (.455)
18) Tx Tech 8-10 (.444)
19) TCU 8-11 (.421)
20) Oregon 5-8 (.385)

Other Weekend vs Big 4 (+series lost):
7) LSU 6-1 (.857)
8) Okla St 8-2 (.800)
8) Tx Tech 8-2 (.800)
10) Arkansas 7-2 (.778)
11) TCU 10-3 (.769)
12) Texas 8-4 (.667)
12) Okla 8-4 (.667)
12) Virginia 6-3 (.667)
15) TA&M 9-6 (.600)
16) Oregon 16-11 (.593)
17) UCLA 12-9 (.571)
18) Lvlle 5-5 (.500)
18) Auburn 7-7 (.500)
20) Notre Dame 4-5 (.444)

This next number is a little more difficult to explain/follow. The idea is to give a consistent weight to games vs Q1 vs the games in the other category. The idea is keep teams from padding the WPct with more non-Q1 games. On average, these teams played 63% of games in the above two categories against Q1, so I gave 63% weight to WPct against Q1 and 37% weight to the WPct against the Other category.

Weighted WPct:
7) UCLA (.661)
8) Texas (.630)
9) LSU (.608)
10) ND (.595)
11) TA&M (.594)
12) Arkansas (.589)
13) Okla St (.582)
14) Tx Tech (.576)
15) Okla (.562)
15) Virginia (.562)
17) Lvlle (.552)
18) TCU (.550)
19) Auburn (.529)
20) Oregon (.462)
This post was edited on 5/27/22 at 12:31 am
Posted by gaetti15
AK
Member since Apr 2013
13365 posts
Posted on 5/27/22 at 1:36 am to
none of that matters. LSU RPI about to jump to like 10 from the what's been known this year as the Tennessee bumper. no matter if we get blown out
first pageprev pagePage 1 of 3Next pagelast page

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram