Thursday, February 18, 2010

Change of Location

Readers:

My brackets will now be hosted at Busting Bracketologists.

I will probably be doing most of my posting from now on over there, with hopefully a greater degree of regularity. We trying to do a combined thing: both Seige and I will post our individual brackets on Monday or Tuesday, and then Tuesday or Wednesday we will come up with a consensus bracket. We will post the log of the discussions that we had regarding that week's bracket. We plan on being the best bracketologists this year.
Read more!

Ramifications? You want to know the Ramifications? Fine!

Here's what happened tonight:

Purdue beat Ohio State in Columbus. Last night, my friend and I debated for 2.5 hours coming up with a consensus bracket (which can be found here), and some time was spent debating Purdue vs. Syracuse on the 1-line. One of Purdue's knocks was that they wouldn't end up with any big road wins. Not anymore...

Saint Louis beat Rhode Island. With that, I think its safe to say farewell to the Rams, whose weak out-of-conference schedule has done them in once again. (I had them out with the assumption that they would win out their conference schedule.

Duquesne beat Charlotte. This seems like it would hurt Charlotte's bubble chances a lot, but remember - bad losses don't count that much. The Texas Tech-Charlotte discussion will be an interesting one.

Utah beat UNLV, and suddenly UNLV doesn't look like a sure tournament team. They are now the same conference record as SDSU, and are banking solely on their 4 top-50 wins and their win at New Mexico. I still think they are in, but they can't drop anymore games.

Wofford wins at Davidson, all but ensuring themselves a bye in the Southern Conference tournament.

Louisville beat Notre Dame, Notre Dame's third straight heart-breaking loss. No real ramifications, other than the near-heart attacks they are giving me.
Read more!

Monday, February 15, 2010

Darnit UConn, you're supposed to be bad!

Well, obviously UConn is going to show up here next week... So, what caused various movements in my seed list over the 9 days? (That is, what didn't I account for?)

Illinois (IN) beat Michigan St, beat Wisconsin at Kohl… what more could you ask for?
Louisville’s (IN) upset victory at Syracuse (Down 1) gave them the big win they needed to push up into the tournament, and the fact that it was on the road is the icing on top. That loss knocked Syracuse off the 1-line.
William & Mary (IN) was a change of opinion… I mean, 5-3 against the RPI Top 50, with wins at Wake Forest and Maryland… that’s pretty legit, in my opinion.
Richmond (Up 4) destroyed Temple and then went on the road and beat Rhode Island (OUT). The first is a good win, while the second is a decent win on the road. It also served as the final blow to Rhode Island’s fragile resume – 24-5 isn’t good enough when your best win is against Oklahoma St, and you’ve had several opportunities for better.
Tennessee (Up 3) impressed me in their loss at Kentucky, enough so that I decided that they will beat Kentucky when they come to Knoxville. Beat one one seed, you might be a fluke. Beat two one seeds, you’re the real deal. Obviously, if Tennessee loses to Kentucky, expect them to drop back down.
Purdue’s (Up 3) rise up the charts is due to two things. First, obviously they went out and beat Michigan St (Down 2) this past Saturday in East Lansing. Big wins are good, and this one means that I now have them tied for first in the conference, rather than in 3rd place. More importantly, however, they now have two road/neutral wins against top teams – and the committee like that. Michigan State, however, drops to only 3-4 against the RPI top 50, and as such cannot be considered to have an elite resume.
Northeastern (OUT) didn’t do anything to hurt themself. However, I decided that winning a conference that has no team above the 10 line is not in and of itself a good argument for inclusion, even if that conference seems to be on track for multiple bids. Expect this to be a big debate in the future.
Dayton (OUT) lost to SLU, which means I am now projecting them to finish 6th in the Atlantic 10. The A-10 is not quite strong enough to get 10 in. (Obviously, a change of heart from last week)
BYU (Down 4) is also a matter of changing opinions – I now think that their weak win total, and absence of strong wins will lead them to be seen as outside of the top tiers of teams.
Charlotte (Down 3), Temple (Down 2), and Xavier (Down 1) are a reaction to my overrating of the A-10 last week.
Texas (Down 2) lost to Oklahoma, then lost to Kansas. While I was expecting the former (since it was a classic trap game), I was not expecting the latter. This had a negative effect on the factors that I use to make my decisions.
Cornell (Down 1) and Northern Iowa (Down 1) were the result of weekend upsets, at UPenn and Bradley respectively.
Read more!

Sunday, February 14, 2010

Well, at least I don't have to consider my team for these anymore

EDIT: (1:56 AM, February 16 - the seed list has been editted, because I just realized that I had 35 at-large teams. What a bonehead move). The above, of course, is in reference to St. John's beating the Gody-less Irish. Two weekends sandwiching a long week, and at long last a new seed list. Sometime tomorrow I'll post detailing the moves I made, but I'll make an acknowledgement - I've been severely underrating Purdue:

1: Kansas, Villanova, Kentucky, Purdue
2: Syracuse, Wisconsin, West Virginia, Duke
3: Georgetown, Pitt, Kansas St., Ohio St
4: Vanderbilt, Texas, Tennessee, Michigan St
5: Baylor, Wake Forest, Gonzaga, UNM
6: Texas A&M, Temple, BYU, Maryland
7: Marquette, Richmond, Georgia Tech, Missouri
8: Florida St, Clemson, Butler, Xavier
9: Florida, UNLV, UNI, Oklahoma St.
10: William & Mary, Illinois, USF, Ole Miss
11: Louisville, Old Dominion, Charlotte, Connecticut
12: Virginia Tech, Washington, Cornell, UTEP
13: Siena, Murray St., Nevada


First Four Out: Miami (FL), Dayton, Rhode Island, Oklahoma
New In: William and Mary, Illinois, Louisville, UConn
New Out: Dayton, Rhode Island, Northeastern, Utah St.

Big Movers: BYU (Down 4), Purdue (Up 3), Tennessee (Up 3), Richmond (Up 3), Charlotte (Down 3), Michigan State (Down 2), Texas (Down 2), Temple (Down 2), Mizzou (Down 2)

Bids by Conference:
Big East (9): Villanova, Syracuse, West Virginia, Georgetown, Pitt, Marquette, USF, Louisville, UConn
ACC (7): Duke, Wake Forest, Maryland, Georgia Tech, Florida St., Clemson, Virginia Tech
Big 12 (7): Kansas, Kansas St, Texas, Baylor, Texas A&M, Missouri, Oklahoma St.
Big 10 (5): Purdue, Wisconsin, Ohio St, Michigan St, Illinois
A-10 (4): Temple, Richmond, Xavier, Charlotte
MWC (3): New Mexico, BYU, UNLV
CAA (2): William and Mary, Old Dominion

Regional Top 4:
Syracuse
1. Villanova
2. Wisconsin
3. Georgetown
4. Texas

St. Louis
1. Kansas
2. Syracuse
3. Ohio St.
4. Vanderbilt

Houston
1. Kentucky
2. Duke
3. Pittsburgh
4. Michigan St.

Utah
1. Purdue
2. West Virginia
3. Kansas St.
4. Tennessee
Read more!

Saturday, February 6, 2010

What? Another Seed List this week? Okay!

So, thanks to reader comments, I went through my last seed list and thought critically about why teams were in their places, reevaluated the ability of teams to actually win certain games, and made some changes. Also, I thought about the autobids, and where they ought to be placed. So, for the first time this year - a full seed list.

Note - the vertical lines indicate where I believe there is a drop-off in quality of team. Also, in the autobids, if the conference is listed, then I believe that as long as one of the favorites wins the conference, they will be on that seed line. In that case, the conference is followed by the team that I believe will win the conference.

1: Kansas, Syracuse, Kentucky Villanova
2: Duke, Texas, Michigan St, BYU
3: G’town, West Virginia, Wisconsin, Kansas St.
4: Temple, Pitt, Ohio St, Purdue
5: Vanderbilt, New Mexico, Baylor, Mizzou
6: Wake Forest, Gonzaga, Maryland, Texas A&M
7: Tennessee, Xavier, Marquette, Rhode Island
8: Northern Iowa Georgia Tech, Florida St., Clemson
9: Florida, Butler, Charlotte, UNLV
10. Richmond, USF, Dayton Ole Miss
11: Oklahoma St, ODU, Virginia Tech, Cornell
12: Northeastern, Utah St Pac 10 (Washington), UAB
13: Siena Murray St, Nevada, Kent St
14: Big Sky (Weber St) Summit (Oakland), Sam Houston St. , SoCon (College of Charleston)
15: Pacific, Morgan St, Coastal Carolina Sun Belt (Western Kentucky)
16: A-Sun (Jacksonville) AE (Vermont), Robert Morris, Lehigh SWAC (UAPB)

Major Changes:
*Georgetown drops because of their home loss to USF, which also leads to South Florida making the bracket at the expense of Minnesota.
*Missouri drops because of their home loss to Texas A&M.
*Realized I was giving New Mexico and UNLV too much credit, dropped them a line. BYU is a great win; SDSU is not.
*Maryland beating Florida St. had a big impact as well, pushing Maryland into the next segment of teams.
*Reevaluated Butler, Virginia Tech, Cornell, and surrounding A-10 teams Charlotte, Richmond, and Dayton.
*I now believe Nevada will win the WAC on their home floor, so they are in the bracket, but I also gave Utah St a boost (since I sold last year's similar version short), bumping out St. Mary's.
*Vanderbilt gets bumped up, because I realized I had counted the Vandy-Kentucky game as a loss for both teams. Oops.
Read more!

Friday, February 5, 2010

How do you seed the autobids?

Each year I get to the autobids, and am mystified. Last year, out of the 21 bottom autobids (those that would have either been in the last 4 in, or out, had they not won their conference tourney), I missed by 1 line on 10, and by 2 lines on 4. If I had randomly assigned the teams, I could have done just as well. How do I rank these teams, who have woefully little to show in their resumes?

To answer this question, I went to my trusted ally - statistical methodology. I entered data for the past 2 seasons, for those autobids who wouldn't have made the tourney on their own merit. (I did include 07-08 Davidson, and 08-09 Utah State, Temple, and VCU, because I'm not convinced that the committee would have included them had they lost). After running a linear regression with the seed as the dependent variable, here's the Model for Selection Committee Autobid Seeding:

SEED = 12.199 - .036 (Wins - Losses) - .076 (Conference Wins - Losses) -.707 (Elite Wins) + .029 (Non-Elite Top 50 Wins) - .063 (RPI 50-100 Wins) + .011 (Losses outside the Top 100) + .007 (RPI Rank) - .043 (Finish Within Conference) + .112 (Ranking of Conference)

I'd like to take a second to note here that out of all those variables, only the Constant and Conference Rank are significant. That is, for all other variables, there is greater than a 5% chance that that variable actually has the opposite effect on seeding, or does not affect the seeding at all. This tells me that the current amount of data is too little - over the next couple of weeks I will work on adding a few more seasons of data.

I also tried using RPI 100-200 Wins, Road/Neutral Wins - Losses, and Last 12 record, but they were all showing up as "The better the team did in this category, the lower seed they received", which tells me that those categories have little to no predictive value.

However, I did spend a lot of time on this, and I'd like to have something to show for it, so I'll go ahead and analyze it anyway. So what does this model tell us? Well... the obvious. A little guy knocking off an elite team (which autobid teams did only 6 times in the last two seasons) bumps them up about a seed line. A team with an RPI of 110 can be expected to be a seed line above a team with an RPI of 220 (Note that the RPI coefficient is only so low because the variable takes on such large values). If you win the WAC (the 11th ranked conference, by Pomeroy rankings), you can expect that to count for two 2 seed lines above the SWAC champion.

Who you lose to means little - even if all of your losses were to teams with an RPI over 100, you wouldn't expect it to change your seed. 10 losses to teams in the top 100 is worse than 5 losses to teams outside the top 100, simply because they lost that chance to pad their win count. This isn't evidence in support of scheduling crappy teams (I'm looking at you, Brey) - the RPI hit from scheduling all bad teams negates the benefit of piling up wins.

I'm not sure exactly why Non-elite top 50 wins causes seed to go up. I suspect that means that when the committee is seeing teams that they are seeding below the 7th line show up as top 50 wins, they discount those wins (Big difference between Kansas and Siena). Instead of counting them as a "half-win" in the 1-50 column, though, they just disregard them altogether. This may be completely false, but its an interesting theory - I know when I see that one of a team's top 50 wins is against Wichita St., I tend to not give them as much credit.

Summary: To get a better seed, you can A) beat elite teams, B) pile up wins, C) lower your RPI, or D) get a better conference. Sigh. As I said earlier, the obvious. Side Note: D might actually be a case of mixing causal relationships - its possible (and indeed probable) that the same factors that make a team more likely to get a good seed also raise the conference ranking.
Read more!

Thursday, February 4, 2010

A Story about Bulldogs

On my last seed list, a reader made a comment asking why Butler had dropped three lines. When making a new seed list, I don't look at the old one until after I am finished the new (and sometimes, not even then). While this causes inconsistencies in logic, it also allows me to see if I would apply the same logic a second time through. But, it is odd that Butler dropped so many lines - so I took a look. Here's why Butler dropped.
So, its obvious that to drop as far as they did, I had to have 9 teams pass Butler. 9! Actually, Butler passed a couple of teams, namely North Carolina and UConn. I could talk about those, but that would take another whole post. So, actually, there are 11 teams. Those teams that passed Butler were: Richmond, Charlotte, Dayton, Rhode Island, Xavier, Texas A&M, Oklahoma St, Cornell, Northern Iowa, Virginia Tech, Marquette.
The first thing that should stand out to you is the rising Atlantic 10. 5 of the 11 teams are from the Atlantic 10. Obviously something changed in my line of thinking. Thankfully, I remember what this is. Between the 1st and 2nd seed list, two A-10 teams (Charlotte and Richmond, I think) moved from the 50-100 range to the 1-50 range. This means that each NCAA candidate A-10 team now shows up with 1-3 extra top 50 wins. This is why all the A-10 teams got pushed up. At this point, you may be angry that such a change makes a difference. You have to remember that team RPIs now don’t matter – what they will be at the end of the year does.
Before you argue more about that point, take a look at this: http://www.bbstate.com/schools/TEX/sheet. This is a replica of the team sheet that the NCAA selection committee will see come March (with data as of today, obviously). What do you notice? (Exercise: Actually consider this for a second before proceeding) The thing that jumps out at me is the grid of records. To make any sense of that grid, you have to narrow your focus, and my logic tells me to look at the columns for 1-50 and 50-100. The priority of course goes to the first column, since it is primarily those teams that will appear in the NCAA tournament.
Let’s take a quick look at Butler. Let’s assume they win out (which I think will happen). They should then have three wins in the top 50 – Ohio State, Xavier, and Siena. Siena is getting the benefit of good scheduling, and is not (IMO) actually good. That’s what my friend, who is a huge Saints homer, tells me. When the committee is actually scrutinizing teams, I think that they will discount that win a little. However, it will still count for something. When they look at the Ohio State win, chances are good that someone (hint: the Ohio State AD) will bring up the Evan Turner injury. So that win will be discounted a little. And, while Xavier is a solid tourney team (says he who had them as “last in” a week ago), they aren’t a protected seed. So keep this in mind: Butler’s lone “excellent” win has an asterisk, and besides that they will only have two “good” wins. I believe they will end the season 26-4, (18-0) in conference, 3-3 against the top 50, 3-1 against the top 100, with a great road/neutral record.
Anyhow, 6 left to explain. Let’s go to Cornell and Northern Iowa, the cases with lots of wins, but less substance than Butler. Northern Iowa is a case of me deciding “You know what? You were dumb in your first seed list. UNI is probably going to go 17-1 in the Missouri Valley Conference – a decently tough conference. Moreover, they’ll end the season 27-2. I don’t care about the “2”, but wowwie! 27 is a big number!” (If you can’t tell, I’m mocking the way I think the committee thinks) So I gave UNI a bump. (And, truth be told, 17-1 in the MVC is better than 18-0 in the Horizon). Cornell… I don’t know about them. They’re going to go 14-0 in the Ivy League. But they have no good wins, just the “good loss” against Kansas. However, that Kansas game did show that they could potentially beat a #1 seed – something the committee likes out of their 9-13 lines. It makes the tourney more interesting. And on that strength, the sportswriters have voted them into the top 25. If the sportswriters see it some way, chances are good the committee will see it similarly. I actually don’t like Cornell up where they are, but I could see it happening.
Virginia Tech and Marquette: Teams I am giving too much credit because of high Pomeroy rankings, despite being wildly inconsistent. Or am I? Marquette fits the role of Boston College last year, (bad losses + a few top wins), and they’ve already stolen a few against good opponents. But I notably bumped them up 4 seed lines in a WEEK. THAT is more shocking than dropping Butler. Wait… they did win a road game against UConn, which increased my faith in their ability to win big games. Thus, I think they take down both Pitt and Louisville, increasing their top 50 wins to 5. Marquette is a case that I’m willing to accept. However, I am certainly giving VT too much credit, operating under the delusion that they can win all three of Wake Forest, Clemson, and Maryland. Its possible, but not likely. What changed in the last week? They added a surprise conference win (at Virginia), which made me believe they could get to 9-7. Count VT in my “idiot” column (and watch them go on to win all three of those games, just to spite me). However, be sure that when predicting the rest of the games in the year for the next seed list I will pay more attention to consistency of a team.

Two left, and they are both are Big 12. Hmm… Okie State sure didn’t do anything to make their resume better – that Kansas St. win was already there. I think I convinced myself this week that winning at Kansas State made up for their other ills, something I didn’t do the first week. Now, I prefer the way I saw it the first time. Another idiot move. Texas A&M… prior to last night, they were and idiot move. However, their surprise win over Missouri puts them at that 5 top 50 win mark. Just an observation – teams with 1-3 wins over the top 50 rpi are bubblicious. Once they hit 4, its very rare for them to miss. 5? Forgetaboutit.

So there you go - 7 good reasons to change my mind on a team, 2 debatable but acceptable reasons, and 2 idiot reasons. Hopefully by the time the tourney rolls around, there will be fewer of the "idiot" reasonings. However, after reflecting a bit, I think that Butler will probably end up as a 9.
Read more!